fdo-mirrors/NetworkManager
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/NetworkManager/NetworkManager.git synced 2025-12-20 23:20:06 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
NetworkManager/libnm-core/tests/test-general.c

9331 lines
347 KiB
C
Raw Normal View History

all: SPDX header conversion $ find * -type f |xargs perl contrib/scripts/spdx.pl $ git rm contrib/scripts/spdx.pl
2019-09-10 11:19:01 +02:00
// SPDX-License-Identifier: GPL-2.0+
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/*
all: unify format of our Copyright source code comments ```bash readarray -d '' FILES < <( git ls-files -z \ ':(exclude)po' \ ':(exclude)shared/c-rbtree' \ ':(exclude)shared/c-list' \ ':(exclude)shared/c-siphash' \ ':(exclude)shared/c-stdaux' \ ':(exclude)shared/n-acd' \ ':(exclude)shared/n-dhcp4' \ ':(exclude)src/systemd/src' \ ':(exclude)shared/systemd/src' \ ':(exclude)m4' \ ':(exclude)COPYING*' ) sed \ -e 's/^\(--\|#\| \*\) *\(([cC]) *\)\?Copyright \+\(\(([cC])\) \+\)\?\(\(20\|19\)[0-9][0-9]\) *[-–] *\(\(20\|19\)[0-9][0-9]\) \+\([^ ].*\)$/\1 C1pyright#\5 - \7#\9/' \ -e 's/^\(--\|#\| \*\) *\(([cC]) *\)\?Copyright \+\(\(([cC])\) \+\)\?\(\(20\|19\)[0-9][0-9]\) *[,] *\(\(20\|19\)[0-9][0-9]\) \+\([^ ].*\)$/\1 C2pyright#\5, \7#\9/' \ -e 's/^\(--\|#\| \*\) *\(([cC]) *\)\?Copyright \+\(\(([cC])\) \+\)\?\(\(20\|19\)[0-9][0-9]\) \+\([^ ].*\)$/\1 C3pyright#\5#\7/' \ -e 's/^Copyright \(\(20\|19\)[0-9][0-9]\) \+\([^ ].*\)$/C4pyright#\1#\3/' \ -i \ "${FILES[@]}" echo ">>> untouched Copyright lines" git grep Copyright "${FILES[@]}" echo ">>> Copyright lines with unusual extra" git grep '\<C[0-9]pyright#' "${FILES[@]}" | grep -i reserved sed \ -e 's/\<C[0-9]pyright#\([^#]*\)#\(.*\)$/Copyright (C) \1 \2/' \ -i \ "${FILES[@]}" ``` https://gitlab.freedesktop.org/NetworkManager/NetworkManager/merge_requests/298
2019-10-01 09:20:35 +02:00
* Copyright (C) 2008 - 2018 Red Hat, Inc.
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
*/
glib-compat: add g_ptr_array_insert()
2015-04-10 09:16:11 +02:00
#define NM_GLIB_COMPAT_H_TEST
all: cleanup includes and let "nm-default.h" include "config.h" - All internal source files (except "examples", which are not internal) should include "config.h" first. As also all internal source files should include "nm-default.h", let "config.h" be included by "nm-default.h" and include "nm-default.h" as first in every source file. We already wanted to include "nm-default.h" before other headers because it might contains some fixes (like "nm-glib.h" compatibility) that is required first. - After including "nm-default.h", we optinally allow for including the corresponding header file for the source file at hand. The idea is to ensure that each header file is self contained. - Don't include "config.h" or "nm-default.h" in any header file (except "nm-sd-adapt.h"). Public headers anyway must not include these headers, and internal headers are never included after "nm-default.h", as of the first previous point. - Include all internal headers with quotes instead of angle brackets. In practice it doesn't matter, because in our public headers we must include other headers with angle brackets. As we use our public headers also to compile our interal source files, effectively the result must be the same. Still do it for consistency. - Except for <config.h> itself. Include it with angle brackets as suggested by https://www.gnu.org/software/autoconf/manual/autoconf.html#Configuration-Headers
2016-02-19 14:57:48 +01:00
#include "nm-default.h"
all: consistently include config.h config.h should be included from every .c file, and it should be included before any other include. Fix that. (As a side effect of how I did this, this also changes us to consistently use "config.h" rather than <config.h>. To the extent that it matters [which is not much], quotes are more correct anyway, since we're talking about a file in our own build tree, not a system include.)
2014-11-13 10:07:02 -05:00
shared: add nm_utils_g_main_context_create_integrate_source() for integrating a GMainContext in another We will rework NMClient entirely. Then, the synchronous initialization will also use the asynchronous code paths. The difference will be that with synchronous initialization, all D-Bus interaction will be done with an internal GMainContext as current thread default, and that internal context will run until initialization completes. Note that even after initialization completes, it cannot be swapped back to the user's (outer) GMainContext. That is because contexts are essentially the queue for our D-Bus events, and we cannot swap from one queue to the other in a race free manner (or a full resync). In other words, the two contexts are not in sync, so after using the internal context NMClient needs to stick to that (at least, until the name owner gets lost, which gives an opportunity to resync and switch back to the user's main context). We thus need to hook the internal (inner) GMainContext with the user's (outer) context, so when the user iterates the outer context, events on the inner context get dispatched. Add nm_utils_g_main_context_create_integrate_source() to create such a GSource for integrating two contexts. Note that the use-case here is limited: the integrated, inner main context must not be explicitly iterated except from being dispatched by the integrating source. Otherwise, you'd get recursive runs, possible deadlocks and general ugliness. NMClient must show restrain how to use the inner context while it is integrated.
2019-11-02 16:55:43 +01:00
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
shared: split C-only helper "shared/nm-std-aux" utils out of "shared/nm-utils" "shared/nm-utils" contains general purpose utility functions that only depend on glib (and extend glib with some helper functions). We will also add code that does not use glib, hence it would be good if the part of "shared/nm-utils" that does not depend on glib, could be used by these future projects. Also, we use the term "utils" everywhere. While that covers the purpose and content well, having everything called "nm-something-utils" is not great. Instead, call this "nm-std-aux", inspired by "c-util/c-stdaux". (cherry picked from commit b434b9ec0794fc62a2469445d17debf645ba24cc)
2019-04-14 13:36:32 +02:00
#include "nm-std-aux/c-list-util.h"
shared: move most of "shared/nm-utils" to "shared/nm-glib-aux" From the files under "shared/nm-utils" we build an internal library that provides glib-based helper utilities. Move the files of that basic library to a new subdirectory "shared/nm-glib-aux" and rename the helper library "libnm-core-base.la" to "libnm-glib-aux.la". Reasons: - the name "utils" is overused in our code-base. Everything's an "utils". Give this thing a more distinct name. - there were additional files under "shared/nm-utils", which are not part of this internal library "libnm-utils-base.la". All the files that are part of this library should be together in the same directory, but files that are not, should not be there. - the new name should better convey what this library is and what is isn't: it's a set of utilities and helper functions that extend glib with funcitonality that we commonly need. There are still some files left under "shared/nm-utils". They have less a unifying propose to be in their own directory, so I leave them there for now. But at least they are separate from "shared/nm-glib-aux", which has a very clear purpose. (cherry picked from commit 80db06f768e47541eae7d66ef48fbe47bf1a69ce)
2019-04-15 08:16:00 +02:00
#include "nm-glib-aux/nm-enum-utils.h"
shared: add nm_utils_strsplit_quoted() We want to parse "/proc/cmdline". That is space separated with support for quoting and escaping. Our implementation becomes part of stable behavior, and we should interpret the kernel command line the same way as the system does. That means, our implementation should match systemd's.
2020-06-22 19:43:35 +02:00
#include "nm-glib-aux/nm-str-buf.h"
#include "systemd/nm-sd-utils-shared.h"
c-list: add c_list_sort() Add a stable, recursive merge sort for CList. This could be improved by doing an iterative implementation. The recursive implementation's stack depth is not an issue, as it is bound by O(ln(n)). But an iterative implementation would safe the overhead of O(n*log(n)) function calls and be potentially faster.
2017-07-08 13:30:00 +02:00
all: cleanup includes and let "nm-default.h" include "config.h" - All internal source files (except "examples", which are not internal) should include "config.h" first. As also all internal source files should include "nm-default.h", let "config.h" be included by "nm-default.h" and include "nm-default.h" as first in every source file. We already wanted to include "nm-default.h" before other headers because it might contains some fixes (like "nm-glib.h" compatibility) that is required first. - After including "nm-default.h", we optinally allow for including the corresponding header file for the source file at hand. The idea is to ensure that each header file is self contained. - Don't include "config.h" or "nm-default.h" in any header file (except "nm-sd-adapt.h"). Public headers anyway must not include these headers, and internal headers are never included after "nm-default.h", as of the first previous point. - Include all internal headers with quotes instead of angle brackets. In practice it doesn't matter, because in our public headers we must include other headers with angle brackets. As we use our public headers also to compile our interal source files, effectively the result must be the same. Still do it for consistency. - Except for <config.h> itself. Include it with angle brackets as suggested by https://www.gnu.org/software/autoconf/manual/autoconf.html#Configuration-Headers
2016-02-19 14:57:48 +01:00
#include "nm-utils.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
#include "nm-setting-private.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-utils.h"
libnm-core/test: add tests for DNS option utilities
2015-04-01 11:47:42 +02:00
#include "nm-utils-private.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-core-internal.h"
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
#include "nm-core-tests-enum-types.h"
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
#include "nm-team-utils.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-setting-8021x.h"
#include "nm-setting-adsl.h"
#include "nm-setting-bluetooth.h"
#include "nm-setting-bond.h"
#include "nm-setting-bridge.h"
#include "nm-setting-bridge-port.h"
#include "nm-setting-cdma.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
#include "nm-setting-connection.h"
libnm, cli, ifcfg-rh: add NMSettingEthtool setting Note that in NetworkManager API (D-Bus, libnm, and nmcli), the features are called "feature-xyz". The "feature-" prefix is used, because NMSettingEthtool possibly will gain support for options that are not only -K|--offload|--features, for example -C|--coalesce. The "xzy" suffix is either how ethtool utility calls the feature ("tso", "rx"). Or, if ethtool utility specifies no alias for that feature, it's the name from kernel's ETH_SS_FEATURES ("tx-tcp6-segmentation"). If possible, we prefer ethtool utility's naming. Also note, how the features "feature-sg", "feature-tso", and "feature-tx" actually refer to multiple underlying kernel features at once. This too follows what ethtool utility does. The functionality is not yet implemented server-side.
2018-07-16 23:37:55 +02:00
#include "nm-setting-ethtool.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-setting-generic.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
#include "nm-setting-gsm.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-setting-infiniband.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
#include "nm-setting-ip4-config.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-setting-ip6-config.h"
#include "nm-setting-olpc-mesh.h"
#include "nm-setting-ppp.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
#include "nm-setting-pppoe.h"
#include "nm-setting-serial.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-setting-team.h"
#include "nm-setting-team-port.h"
libnm: add NMSettingUser This only adds new API for a NMSettingUser. The setting class is still entirely unused. The point is getting the new API into 1.8.0 release of libnm. It's easier to backport the use of the API to a stable branch then backporting public API. https://bugzilla.gnome.org/show_bug.cgi?id=776276 https://bugzilla.redhat.com/show_bug.cgi?id=1421429
2017-03-24 12:41:04 +01:00
#include "nm-setting-user.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
#include "nm-setting-vlan.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-setting-vpn.h"
#include "nm-setting-wimax.h"
#include "nm-setting-wired.h"
#include "nm-setting-wireless.h"
#include "nm-setting-wireless-security.h"
core: add NMSettingWpan
2018-03-09 10:51:49 +01:00
#include "nm-setting-wpan.h"
libnm-core: add nm-core-types.h, remove cross-includes Add nm-core-types.h, typedefing all of the GObject types in libnm-core; this is needed so that nm-setting.h can reference NMConnection in addition to nm-connection.h referencing NMSetting. Removing the cross-includes from the various headers causes lots of fallout elsewhere. (In particular, nm-utils.h used to include nm-connection.h, which included every setting header, so any file that included nm-utils.h automatically got most of the rest of libnm-core without needing to pay attention to specifics.) Fix this up by including nm-core-internal.h from those files that are now missing includes.
2014-10-21 22:09:52 -04:00
#include "nm-simple-connection.h"
libnm/keyfile: build keyfile code as separate GPL licensed internal library Keyfile support was initially added under GPL-2.0+ license as part of core. It was moved to "libnm-core" in commit 59eb5312a5d6 ('keyfile: merge branch 'th/libnm-keyfile-bgo744699''). "libnm-core" is statically linked with by core and "libnm". In the former case under terms of GPL-2.0+ (good) and in the latter case under terms of LGPL-2.1+ (bad). In fact, to this day, "libnm" doesn't actually use the code. The linker will probably remove all the GPL-2.0+ symbols when compiled with gc-sections or LTO. Still, linking them together in the first place makes "libnm" only available under GPL code (despite the code not actually being used). Instead, move the GPL code to a separate static library "shared/nm-keyfile/libnm-keyfile.la" and only link it to the part that actually uses the code (and which is GPL licensed too). This fixes the license violation. Eventually, it would be very useful to be able to expose keyfile handling via "libnm". However that is not straight forward due to the licensing conflict. https://gitlab.freedesktop.org/NetworkManager/NetworkManager/merge_requests/381
2020-01-02 07:37:59 +01:00
#include "nm-keyfile/nm-keyfile-internal.h"
shared: move most of "shared/nm-utils" to "shared/nm-glib-aux" From the files under "shared/nm-utils" we build an internal library that provides glib-based helper utilities. Move the files of that basic library to a new subdirectory "shared/nm-glib-aux" and rename the helper library "libnm-core-base.la" to "libnm-glib-aux.la". Reasons: - the name "utils" is overused in our code-base. Everything's an "utils". Give this thing a more distinct name. - there were additional files under "shared/nm-utils", which are not part of this internal library "libnm-utils-base.la". All the files that are part of this library should be together in the same directory, but files that are not, should not be there. - the new name should better convey what this library is and what is isn't: it's a set of utilities and helper functions that extend glib with funcitonality that we commonly need. There are still some files left under "shared/nm-utils". They have less a unifying propose to be in their own directory, so I leave them there for now. But at least they are separate from "shared/nm-glib-aux", which has a very clear purpose. (cherry picked from commit 80db06f768e47541eae7d66ef48fbe47bf1a69ce)
2019-04-15 08:16:00 +02:00
#include "nm-glib-aux/nm-dedup-multi.h"
shared: build helper "libnm-libnm-core-{intern|aux}.la" library for libnm-core "libnm-core" implements common functionality for "NetworkManager" and "libnm". Note that clients like "nmcli" cannot access the internal API provided by "libnm-core". So, if nmcli wants to do something that is also done by "libnm-core", , "libnm", or "NetworkManager", the code would have to be duplicated. Instead, such code can be in "libnm-libnm-core-{intern|aux}.la". Note that: 0) "libnm-libnm-core-intern.la" is used by libnm-core itsself. On the other hand, "libnm-libnm-core-aux.la" is not used by libnm-core, but provides utilities on top of it. 1) they both extend "libnm-core" with utlities that are not public API of libnm itself. Maybe part of the code should one day become public API of libnm. On the other hand, this is code for which we may not want to commit to a stable interface or which we don't want to provide as part of the API. 2) "libnm-libnm-core-intern.la" is statically linked by "libnm-core" and thus directly available to "libnm" and "NetworkManager". On the other hand, "libnm-libnm-core-aux.la" may be used by "libnm" and "NetworkManager". Both libraries may be statically linked by libnm clients (like nmcli). 3) it must only use glib, libnm-glib-aux.la, and the public API of libnm-core. This is important: it must not use "libnm-core/nm-core-internal.h" nor "libnm-core/nm-utils-private.h" so the static library is usable by nmcli which couldn't access these. Note that "shared/nm-meta-setting.c" is an entirely different case, because it behaves differently depending on whether linking against "libnm-core" or the client programs. As such, this file must be compiled twice. (cherry picked from commit af07ed01c04867e281cc3982a7ab0d244d4f8e2e)
2019-04-15 09:26:53 +02:00
#include "nm-libnm-core-intern/nm-ethtool-utils.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
#include "test-general-enums.h"
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
shared: move shared files to subdirectory "shared/nm-utils/" The "shared" directory contains files that are possibly used by all components of NetworkManager repository. Some of these files are even copied as-is to other projects (VPN plugins, nm-applet) and used there without modification. Move those files to a separate directory. By moving them to a common directory, it is clearer that they belong together. Also, you can easier compare the copied versions to their original via $ diff -r ./shared/nm-utils/ /path/to/nm-vpn-plugin/shared/nm-utils/
2016-06-15 15:29:10 +02:00
#include "nm-utils/nm-test-utils.h"
all: cleanup includes and let "nm-default.h" include "config.h" - All internal source files (except "examples", which are not internal) should include "config.h" first. As also all internal source files should include "nm-default.h", let "config.h" be included by "nm-default.h" and include "nm-default.h" as first in every source file. We already wanted to include "nm-default.h" before other headers because it might contains some fixes (like "nm-glib.h" compatibility) that is required first. - After including "nm-default.h", we optinally allow for including the corresponding header file for the source file at hand. The idea is to ensure that each header file is self contained. - Don't include "config.h" or "nm-default.h" in any header file (except "nm-sd-adapt.h"). Public headers anyway must not include these headers, and internal headers are never included after "nm-default.h", as of the first previous point. - Include all internal headers with quotes instead of angle brackets. In practice it doesn't matter, because in our public headers we must include other headers with angle brackets. As we use our public headers also to compile our interal source files, effectively the result must be the same. Still do it for consistency. - Except for <config.h> itself. Include it with angle brackets as suggested by https://www.gnu.org/software/autoconf/manual/autoconf.html#Configuration-Headers
2016-02-19 14:57:48 +01:00
all: add C99's "bool" define https://mail.gnome.org/archives/networkmanager-list/2015-November/msg00036.html
2015-11-11 13:33:46 +01:00
/* When passing a "bool" typed argument to a variadic function that
* expects a gboolean, the compiler will promote the integer type
* to have at least size (int). That way:
* g_object_set (obj, PROP_BOOL, bool_val, NULL);
* will just work correctly. */
G_STATIC_ASSERT (sizeof (gboolean) == sizeof (int));
G_STATIC_ASSERT (sizeof (bool) <= sizeof (int));
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
/*****************************************************************************/
shared: add NM_ASCII_SPACES macro
2019-03-20 10:43:54 +01:00
static void
test_nm_ascii_spaces (void)
{
int i;
const char *const S = NM_ASCII_SPACES;
for (i = 0; S[i]; i++)
g_assert (!strchr (&S[i + 1], S[i]));
for (i = 0; S[i] != '\0'; i++)
g_assert (g_ascii_isspace (S[i]));
g_assert (!g_ascii_isspace ((char) 0));
for (i = 1; i < 0x100; i++) {
if (g_ascii_isspace ((char) i))
g_assert (strchr (S, (char) i));
else
g_assert (!strchr (S, (char) i));
}
}
/*****************************************************************************/
all: add helper functions for nm_hash_update*() By using a macro, we don't cast all the types to guint. Instead, we use their native types directly. Hence, we don't need nm_hash_update_uint64() nor nm_hash_update_ptr(). Also, for types smaller then guint like char, we save hashing the all zero bytes.
2017-10-16 12:38:16 +02:00
typedef struct _nm_packed {
int v0;
char v1;
double v2;
guint8 v3;
} TestHashStruct;
static void
_test_hash_struct (int v0, char v1, double v2, guint8 v3)
{
const TestHashStruct s = {
.v0 = v0,
.v1 = v1,
.v2 = v2,
.v3 = v3,
};
NMHashState h;
guint hh;
nm_hash_init (&h, 100);
nm_hash_update (&h, &s, sizeof (s));
hh = nm_hash_complete (&h);
nm_hash_init (&h, 100);
nm_hash_update_val (&h, v0);
nm_hash_update_val (&h, v1);
nm_hash_update_val (&h, v2);
nm_hash_update_val (&h, v3);
g_assert_cmpint (hh, ==, nm_hash_complete (&h));
nm_hash_init (&h, 100);
nm_hash_update_vals (&h, v0, v1, v2, v3);
g_assert_cmpint (hh, ==, nm_hash_complete (&h));
}
all: refactor hashing by introducing NMHashState The privious NM_HASH_* macros directly operated on a guint value and were thus close to the actual implementation. Replace them by adding a NMHashState struct and accessors to update the hash state. This hides the implementation better and would allow us to carry more state. For example, we could switch to siphash24() transparently. For now, we still do a form basically djb2 hashing, albeit with differing start seed. Also add nm_hash_str() and nm_str_hash(): - nm_hash_str() is our own string hashing implementation - nm_str_hash() is our own string implementation, but with a GHashFunc signature, suitable to pass it to g_hash_table_new(). Also, it has this name in order to remind you of g_str_hash(), which it is replacing.
2017-10-13 14:00:22 +02:00
static guint
_test_hash_str (const char *str)
{
NMHashState h;
guint v, v2;
const guint SEED = 10;
nm_hash_init (&h, SEED);
all: use siphash24 for hashing siphash24() is wildly used by projects nowadays. It's certainly slower then our djb hashing that we used before. But quite likely it's fast enough for us, given how wildly it is used. I think it would be hard to profile NetworkManager to show that the performance of hash tables is the issue, be it with djb or siphash24. Certainly with siphash24() it's much harder to exploit the hashing algorithm to cause worst case hash operations (provided that the seed is kept private). Does this better resistance against a denial of service matter for us? Probably not, but let's better be safe then sorry. Note that systemd's implementation uses a different seed for each hash table (at least, after the hash table grows to a certain size). We don't do that and use only one global seed.
2017-10-13 17:16:23 +02:00
nm_hash_update_str0 (&h, str);
all: refactor hashing by introducing NMHashState The privious NM_HASH_* macros directly operated on a guint value and were thus close to the actual implementation. Replace them by adding a NMHashState struct and accessors to update the hash state. This hides the implementation better and would allow us to carry more state. For example, we could switch to siphash24() transparently. For now, we still do a form basically djb2 hashing, albeit with differing start seed. Also add nm_hash_str() and nm_str_hash(): - nm_hash_str() is our own string hashing implementation - nm_str_hash() is our own string implementation, but with a GHashFunc signature, suitable to pass it to g_hash_table_new(). Also, it has this name in order to remind you of g_str_hash(), which it is replacing.
2017-10-13 14:00:22 +02:00
v = nm_hash_complete (&h);
all: use siphash24 for hashing siphash24() is wildly used by projects nowadays. It's certainly slower then our djb hashing that we used before. But quite likely it's fast enough for us, given how wildly it is used. I think it would be hard to profile NetworkManager to show that the performance of hash tables is the issue, be it with djb or siphash24. Certainly with siphash24() it's much harder to exploit the hashing algorithm to cause worst case hash operations (provided that the seed is kept private). Does this better resistance against a denial of service matter for us? Probably not, but let's better be safe then sorry. Note that systemd's implementation uses a different seed for each hash table (at least, after the hash table grows to a certain size). We don't do that and use only one global seed.
2017-10-13 17:16:23 +02:00
/* assert that hashing a string and a buffer yields the
* same result.
*
* I think that is a desirable property. */
nm_hash_init (&h, SEED);
nm_hash_update_mem (&h, str, strlen (str));
v2 = nm_hash_complete (&h);
all: refactor hashing by introducing NMHashState The privious NM_HASH_* macros directly operated on a guint value and were thus close to the actual implementation. Replace them by adding a NMHashState struct and accessors to update the hash state. This hides the implementation better and would allow us to carry more state. For example, we could switch to siphash24() transparently. For now, we still do a form basically djb2 hashing, albeit with differing start seed. Also add nm_hash_str() and nm_str_hash(): - nm_hash_str() is our own string hashing implementation - nm_str_hash() is our own string implementation, but with a GHashFunc signature, suitable to pass it to g_hash_table_new(). Also, it has this name in order to remind you of g_str_hash(), which it is replacing.
2017-10-13 14:00:22 +02:00
g_assert (v == v2);
return v;
}
all: add helper functions for nm_hash_update*() By using a macro, we don't cast all the types to guint. Instead, we use their native types directly. Hence, we don't need nm_hash_update_uint64() nor nm_hash_update_ptr(). Also, for types smaller then guint like char, we save hashing the all zero bytes.
2017-10-16 12:38:16 +02:00
#define _test_hash_vals(type, ...) \
G_STMT_START { \
NMHashState h0, h1, h2, h3; \
const type v[] = { __VA_ARGS__ }; \
guint h; \
guint i; \
\
nm_hash_init (&h0, 10); \
nm_hash_init (&h1, 10); \
nm_hash_init (&h2, 10); \
nm_hash_init (&h3, 10); \
\
/* assert that it doesn't matter, whether we hash the values individually,
* or all at once, or via the convenience macros nm_hash_update_val()
* and nm_hash_update_vals(). */ \
for (i = 0; i < G_N_ELEMENTS (v); i++) { \
nm_hash_update (&h0, &v[i], sizeof (type)); \
nm_hash_update_val (&h1, v[i]); \
} \
nm_hash_update_vals (&h2, __VA_ARGS__); \
nm_hash_update (&h3, v, sizeof (v)); \
\
h = nm_hash_complete (&h0); \
g_assert_cmpint (h, ==, nm_hash_complete (&h1)); \
g_assert_cmpint (h, ==, nm_hash_complete (&h2)); \
g_assert_cmpint (h, ==, nm_hash_complete (&h3)); \
} G_STMT_END
all: refactor hashing by introducing NMHashState The privious NM_HASH_* macros directly operated on a guint value and were thus close to the actual implementation. Replace them by adding a NMHashState struct and accessors to update the hash state. This hides the implementation better and would allow us to carry more state. For example, we could switch to siphash24() transparently. For now, we still do a form basically djb2 hashing, albeit with differing start seed. Also add nm_hash_str() and nm_str_hash(): - nm_hash_str() is our own string hashing implementation - nm_str_hash() is our own string implementation, but with a GHashFunc signature, suitable to pass it to g_hash_table_new(). Also, it has this name in order to remind you of g_str_hash(), which it is replacing.
2017-10-13 14:00:22 +02:00
static void
test_nm_hash (void)
{
shared: add nm_hash_static() to get a static hash key When using siphash24(), the hash value depends on the hashed input and the key from _get_hash_key(). If the input is static, so is also the result of siphash24(), albeit the bits are scrabbled more. Add a nm_hash_static() to get such a static key, but without actually doing siphash24(). The static key is also xored with a static_seed. For that, also mangle the first byte of the hash key using siphash24() itself. That is, because nm_hash_static() only uses the first guint of the random key. Hence, we want that this first guint has all the entropy of the entire key. We use siphash24() itself, to mangle all bits of the 16 byte key into the first guint.
2017-11-15 15:39:50 +01:00
g_assert (nm_hash_static (0));
g_assert (nm_hash_static (777));
shared: optimize nm_hash_str() for NULL to not use siphash24()
2017-11-15 15:14:19 +01:00
g_assert (nm_hash_str (NULL));
g_assert (nm_hash_str (""));
g_assert (nm_hash_str ("a"));
shared: use siphash24() for nm_hash_ptr() siphash24() mixes the bits much better then our naive xor. Don't bypass siphash24(). We supposedly use it for the better hashing properties, so use it also for pointers.
2017-11-15 15:33:32 +01:00
g_assert (nm_hash_ptr (NULL));
g_assert (nm_hash_ptr (""));
g_assert (nm_hash_ptr ("a"));
all: refactor hashing by introducing NMHashState The privious NM_HASH_* macros directly operated on a guint value and were thus close to the actual implementation. Replace them by adding a NMHashState struct and accessors to update the hash state. This hides the implementation better and would allow us to carry more state. For example, we could switch to siphash24() transparently. For now, we still do a form basically djb2 hashing, albeit with differing start seed. Also add nm_hash_str() and nm_str_hash(): - nm_hash_str() is our own string hashing implementation - nm_str_hash() is our own string implementation, but with a GHashFunc signature, suitable to pass it to g_hash_table_new(). Also, it has this name in order to remind you of g_str_hash(), which it is replacing.
2017-10-13 14:00:22 +02:00
_test_hash_str ("");
_test_hash_str ("a");
_test_hash_str ("aa");
_test_hash_str ("diceros bicornis longipes");
all: add helper functions for nm_hash_update*() By using a macro, we don't cast all the types to guint. Instead, we use their native types directly. Hence, we don't need nm_hash_update_uint64() nor nm_hash_update_ptr(). Also, for types smaller then guint like char, we save hashing the all zero bytes.
2017-10-16 12:38:16 +02:00
/* assert that nm_hash_update_vals() is the same as calling nm_hash_update_val() multiple times. */
_test_hash_vals (int, 1);
_test_hash_vals (int, 1, 2);
_test_hash_vals (int, 1, 2, 3);
_test_hash_vals (int, 1, 2, 3, 4);
_test_hash_vals (long, 1l);
_test_hash_vals (long, 1l, 2l, 3l, 4l, 5l);
_test_hash_struct (10, 'a', 5.4, 7);
_test_hash_struct (-10, '\0', -5.4e49, 255);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint8, 1, 0), ==, 0x002);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint8, 1, 1), ==, 0x003);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint8, 1, 1, 0, 0, 0, 0), ==, 0x030);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint8, 1, 1, 0, 0, 0, 1), ==, 0x031);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint8, 0, 0, 1, 1, 0, 0, 0, 1), ==, 0x031);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint16, 0, 0, 1, 1, 0, 0, 0, 1), ==, 0x031);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint16, 0, 0, 0, 1, 1, 0, 0, 0, 1), ==, 0x031);
g_assert_cmpint (NM_HASH_COMBINE_BOOLS (guint16, 1, 0, 0, 1, 1, 0, 0, 0, 1), ==, 0x131);
all: refactor hashing by introducing NMHashState The privious NM_HASH_* macros directly operated on a guint value and were thus close to the actual implementation. Replace them by adding a NMHashState struct and accessors to update the hash state. This hides the implementation better and would allow us to carry more state. For example, we could switch to siphash24() transparently. For now, we still do a form basically djb2 hashing, albeit with differing start seed. Also add nm_hash_str() and nm_str_hash(): - nm_hash_str() is our own string hashing implementation - nm_str_hash() is our own string implementation, but with a GHashFunc signature, suitable to pass it to g_hash_table_new(). Also, it has this name in order to remind you of g_str_hash(), which it is replacing.
2017-10-13 14:00:22 +02:00
}
/*****************************************************************************/
shared: add nm_g_slice_free_fcn() util Useful, when you need a GDestroyNotify function for g_slice_free() of a certain type.
2017-08-15 12:32:37 +02:00
static void
test_nm_g_slice_free_fcn (void)
{
gpointer p;
shared: add static assert for nm_g_slice_free_fcn() argument
2017-12-07 12:31:55 +01:00
struct {
char a1;
char a2;
} xx;
shared: add nm_g_slice_free_fcn() util Useful, when you need a GDestroyNotify function for g_slice_free() of a certain type.
2017-08-15 12:32:37 +02:00
p = g_slice_new (gint64);
(nm_g_slice_free_fcn (gint64)) (p);
p = g_slice_new (gint32);
(nm_g_slice_free_fcn (gint32)) (p);
all: don't use gchar/gshort/gint/glong but C types We commonly don't use the glib typedefs for char/short/int/long, but their C types directly. $ git grep '\<g\(char\|short\|int\|long\|float\|double\)\>' | wc -l 587 $ git grep '\<\(char\|short\|int\|long\|float\|double\)\>' | wc -l 21114 One could argue that using the glib typedefs is preferable in public API (of our glib based libnm library) or where it clearly is related to glib, like during g_object_set (obj, PROPERTY, (gint) value, NULL); However, that argument does not seem strong, because in practice we don't follow that argument today, and seldomly use the glib typedefs. Also, the style guide for this would be hard to formalize, because "using them where clearly related to a glib" is a very loose suggestion. Also note that glib typedefs will always just be typedefs of the underlying C types. There is no danger of glib changing the meaning of these typedefs (because that would be a major API break of glib). A simple style guide is instead: don't use these typedefs. No manual actions, I only ran the bash script: FILES=($(git ls-files '*.[hc]')) sed -i \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\>\( [^ ]\)/\1\2/g' \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\> /\1 /g' \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\>/\1/g' \ "${FILES[@]}"
2018-07-11 07:40:19 +02:00
p = g_slice_new (int);
(nm_g_slice_free_fcn (int)) (p);
shared: add nm_g_slice_free_fcn() util Useful, when you need a GDestroyNotify function for g_slice_free() of a certain type.
2017-08-15 12:32:37 +02:00
p = g_slice_new (gint64);
nm_g_slice_free_fcn_gint64 (p);
shared: add static assert for nm_g_slice_free_fcn() argument
2017-12-07 12:31:55 +01:00
p = g_slice_alloc (sizeof (xx));
(nm_g_slice_free_fcn (xx)) (p);
shared: add nm_g_slice_free_fcn() util Useful, when you need a GDestroyNotify function for g_slice_free() of a certain type.
2017-08-15 12:32:37 +02:00
}
/*****************************************************************************/
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
static void
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
_do_test_nm_utils_strsplit_set_f_one (NMUtilsStrsplitSetFlags flags,
const char *str,
gsize words_len,
const char *const*exp_words)
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
{
shared/tests: add unit tests for new flags of nm_utils_strsplit_set_full() (cherry picked from commit a7d1e14e6deb33dc51902bf76c51755e9915bb1a)
2019-04-16 12:34:11 +02:00
#define DELIMITERS " \n"
#define DELIMITERS_C ' ', '\n'
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
gs_free const char **words = NULL;
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
gsize i, j, k;
const gboolean f_allow_escaping = NM_FLAGS_HAS (flags, NM_UTILS_STRSPLIT_SET_FLAGS_ALLOW_ESCAPING);
const gboolean f_preserve_empty = NM_FLAGS_HAS (flags, NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY);
const char *s1;
gsize initial_offset;
gs_strfreev char **words_g = NULL;
g_assert (!NM_FLAGS_ANY (flags, ~( NM_UTILS_STRSPLIT_SET_FLAGS_ALLOW_ESCAPING
| NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY)));
/* assert that the epected words are valid (and don't contain unescaped delimiters). */
for (i = 0; i < words_len; i++) {
const char *w = exp_words[i];
g_assert (w);
if (!f_preserve_empty)
g_assert (w[0]);
for (k = 0; w[k]; ) {
if ( f_allow_escaping
&& w[k] == '\\') {
k++;
if (w[k] == '\0')
break;
k++;
continue;
}
g_assert (!NM_IN_SET (w[k], DELIMITERS_C));
k++;
}
if (!f_allow_escaping)
g_assert (!NM_STRCHAR_ANY (w, ch, NM_IN_SET (ch, DELIMITERS_C)));
}
initial_offset = (f_preserve_empty || !str)
? 0u
: strspn (str, DELIMITERS);
/* first compare our expected values with what g_strsplit_set() would
* do. */
words_g = str ? g_strsplit_set (str, DELIMITERS, -1) : NULL;
if (str == NULL) {
g_assert_cmpint (words_len, ==, 0);
g_assert (!words_g);
} else if (nm_streq0 (str, "")) {
g_assert_cmpint (words_len, ==, 0);
g_assert (words_g);
g_assert (!words_g[0]);
} else {
g_assert (words_g);
g_assert (words_g[0]);
if (!f_allow_escaping) {
if (!f_preserve_empty) {
for (i = 0, j = 0; words_g[i]; i++) {
if (words_g[i][0] == '\0')
g_free (words_g[i]);
else
words_g[j++] = words_g[i];
}
words_g[j] = NULL;
}
if (f_preserve_empty)
g_assert_cmpint (words_len, >, 0);
for (i = 0; i < words_len; i++) {
g_assert (exp_words[i]);
g_assert_cmpstr (exp_words[i], ==, words_g[i]);
}
g_assert (words_g[words_len] == NULL);
g_assert_cmpint (NM_PTRARRAY_LEN (words_g), ==, words_len);
g_assert (_nm_utils_strv_cmp_n (exp_words, words_len, NM_CAST_STRV_CC (words_g), -1) == 0);
}
shared: add flags argument to nm_utils_strsplit_set() It will be useful to extend nm_utils_strsplit_set() with various flavors and subtly different behaviors. Add a flags argument to support these.
2019-04-03 15:44:19 +02:00
}
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
if ( flags == NM_UTILS_STRSPLIT_SET_FLAGS_NONE
&& nmtst_get_rand_bool ())
words = nm_utils_strsplit_set (str, DELIMITERS);
else if ( flags == NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY
&& nmtst_get_rand_bool ())
words = nm_utils_strsplit_set_with_empty (str, DELIMITERS);
else
words = nm_utils_strsplit_set_full (str, DELIMITERS, flags);
g_assert_cmpint (NM_PTRARRAY_LEN (words), ==, words_len);
if (words_len == 0) {
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
g_assert (!words);
g_assert ( !str
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
|| NM_STRCHAR_ALL (str, ch, NM_IN_SET (ch, DELIMITERS_C)));
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
return;
}
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
g_assert (words);
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
for (i = 0; i < words_len; i++)
g_assert_cmpstr (exp_words[i], ==, words[i]);
g_assert (words[words_len] == NULL);
g_assert (_nm_utils_strv_cmp_n (exp_words, words_len, words, -1) == 0);
s1 = words[0];
g_assert (s1 >= (char *) &words[words_len + 1]);
s1 = &s1[strlen (str)];
for (i = 1; i < words_len; i++) {
g_assert (&(words[i - 1])[strlen (words[i - 1])] < words[i]);
g_assert (words[i] <= s1);
}
/* while strsplit removes all delimiters, we can relatively easily find them
* in the original string. Assert that the original string and the pointer offsets
* of words correspond. In particular, find idx_delim_after and idx_delim_before
* to determine which delimiter was after/before a word. */
{
gsize idx_word_start;
gsize idx_delim_after_old = G_MAXSIZE;
idx_word_start = initial_offset;
for (i = 0; i < words_len; i++) {
const gsize l_i = strlen (words[i]);
gsize idx_delim_after;
gsize idx_delim_before;
/* find the delimiter *after* words[i]. We can do that by looking at the next
* word and calculating the pointer difference.
*
* The delimiter after the very last word is '\0' and requires strlen() to find. */
idx_delim_after = initial_offset + ((words[i] - words[0]) + l_i);
if (idx_delim_after != idx_word_start + l_i) {
g_assert (!f_preserve_empty);
g_assert_cmpint (idx_word_start + l_i, <, idx_delim_after);
idx_word_start = idx_delim_after - l_i;
}
if (i + 1 < words_len) {
gsize x = initial_offset + ((words[i + 1] - words[0]) - 1);
if (idx_delim_after != x) {
g_assert (!f_preserve_empty);
g_assert_cmpint (idx_delim_after, <, x);
for (k = idx_delim_after; k <= x; k++)
g_assert (NM_IN_SET (str[k], DELIMITERS_C));
}
g_assert (NM_IN_SET (str[idx_delim_after], DELIMITERS_C));
} else {
if (f_preserve_empty)
g_assert (NM_IN_SET (str[idx_delim_after], '\0'));
else
g_assert (NM_IN_SET (str[idx_delim_after], '\0', DELIMITERS_C));
}
/* find the delimiter *before* words[i]. */
if (i == 0) {
/* there is only a delimiter *before*, with !f_preserve_empty and leading
* delimiters. */
idx_delim_before = G_MAXSIZE;
if (initial_offset > 0) {
g_assert (!f_preserve_empty);
idx_delim_before = initial_offset - 1;
}
} else
idx_delim_before = initial_offset + (words[i] - words[0]) - 1;
if (idx_delim_before != G_MAXSIZE)
g_assert (NM_IN_SET (str[idx_delim_before], DELIMITERS_C));
if (idx_delim_after_old != idx_delim_before) {
g_assert (!f_preserve_empty);
if (i == 0) {
g_assert_cmpint (initial_offset, >, 0);
g_assert_cmpint (idx_delim_before, !=, G_MAXSIZE);
g_assert_cmpint (idx_delim_before, ==, initial_offset - 1);
} else {
g_assert_cmpint (idx_delim_after_old, !=, G_MAXSIZE);
g_assert_cmpint (idx_delim_before, !=, G_MAXSIZE);
g_assert_cmpint (idx_delim_after_old, <, idx_delim_before);
for (k = idx_delim_after_old; k <= idx_delim_before; k++)
g_assert (NM_IN_SET (str[k], DELIMITERS_C));
}
}
for (k = 0; k < l_i; ) {
if ( f_allow_escaping
&& str[idx_word_start + k] == '\\') {
k++;
if (k >= l_i)
break;
k++;
continue;
}
g_assert (!NM_IN_SET (str[idx_word_start + k], DELIMITERS_C));
k++;
}
g_assert (strncmp (words[i], &str[idx_word_start], l_i) == 0);
if (i > 0) {
const char *s = &(words[i - 1])[strlen (words[i - 1]) + 1];
if (s != words[i]) {
g_assert (!f_preserve_empty);
g_assert (s < words[i]);
}
}
idx_word_start += l_i + 1;
idx_delim_after_old = idx_delim_after;
}
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
}
}
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
static void
_do_test_nm_utils_strsplit_set_f (NMUtilsStrsplitSetFlags flags,
const char *str,
gsize words_len,
const char *const*exp_words)
{
_do_test_nm_utils_strsplit_set_f_one (flags, str, words_len, exp_words);
if (NM_FLAGS_HAS (flags, NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY)) {
gs_unref_ptrarray GPtrArray *exp_words2 = NULL;
gsize k;
exp_words2 = g_ptr_array_new ();
for (k = 0; k < words_len; k++) {
if (exp_words[k][0] != '\0')
g_ptr_array_add (exp_words2, (gpointer) exp_words[k]);
}
_do_test_nm_utils_strsplit_set_f_one (flags & (~NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY),
str,
exp_words2->len,
(const char *const*) exp_words2->pdata);
}
}
#define do_test_nm_utils_strsplit_set_f(flags, str, ...) \
_do_test_nm_utils_strsplit_set_f (flags, \
str, \
NM_NARG (__VA_ARGS__), \
NM_MAKE_STRV (__VA_ARGS__))
#define do_test_nm_utils_strsplit_set(allow_escaping, str, ...) \
do_test_nm_utils_strsplit_set_f ( (allow_escaping) \
? NM_UTILS_STRSPLIT_SET_FLAGS_ALLOW_ESCAPING \
: NM_UTILS_STRSPLIT_SET_FLAGS_NONE, \
str, \
##__VA_ARGS__)
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
shared/tests: add unit tests for new flags of nm_utils_strsplit_set_full() (cherry picked from commit a7d1e14e6deb33dc51902bf76c51755e9915bb1a)
2019-04-16 12:34:11 +02:00
static void
_do_test_nm_utils_strsplit_set_simple (NMUtilsStrsplitSetFlags flags,
const char *str,
gsize words_len,
const char *const*exp_words)
{
gs_free const char **tokens = NULL;
gsize n_tokens;
tokens = nm_utils_strsplit_set_full (str, DELIMITERS, flags);
if (!tokens) {
g_assert_cmpint (words_len, ==, 0);
return;
}
g_assert (str && str[0]);
g_assert_cmpint (words_len, >, 0);
n_tokens = NM_PTRARRAY_LEN (tokens);
if (_nm_utils_strv_cmp_n (exp_words, words_len, tokens, -1) != 0) {
gsize i;
g_print (">>> split \"%s\" (flags %x) got %zu tokens (%zu expected)\n", str, (guint) flags, n_tokens, words_len);
for (i = 0; i < NM_MAX (n_tokens, words_len); i++) {
const char *s1 = i < n_tokens ? tokens[i] : NULL;
const char *s2 = i < words_len ? exp_words[i] : NULL;
g_print (">>> [%zu]: %s - %s%s%s vs. %s%s%s\n",
i,
nm_streq0 (s1, s2) ? "same" : "diff",
NM_PRINT_FMT_QUOTE_STRING (s1),
NM_PRINT_FMT_QUOTE_STRING (s2));
}
g_assert_not_reached ();
}
g_assert_cmpint (words_len, ==, NM_PTRARRAY_LEN (tokens));
}
#define do_test_nm_utils_strsplit_set_simple(flags, str, ...) \
_do_test_nm_utils_strsplit_set_simple ((flags), (str), NM_NARG (__VA_ARGS__), NM_MAKE_STRV (__VA_ARGS__))
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
static void
test_nm_utils_strsplit_set (void)
{
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
gs_unref_ptrarray GPtrArray *words_exp = NULL;
guint test_run;
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_NONE, NULL);
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_NONE, "");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_NONE, " ");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_NONE, "a b", "a", "b");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, NULL);
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, "");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, " ", "", "");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, " ", "", "", "");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, "a ", "a", "", "");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, "a b", "a", "", "b");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, " ab b", "", "ab", "", "b");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, "ab b", "ab", "", "b");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, "abb", "abb");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, "abb bb ", "abb", "", "bb", "");
do_test_nm_utils_strsplit_set_f (NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY, "abb bcb ", "abb", "bcb", "");
shared: add @allow_escaping argument to @nm_utils_strsplit_set
2018-08-09 09:28:30 +02:00
do_test_nm_utils_strsplit_set (FALSE, NULL);
do_test_nm_utils_strsplit_set (FALSE, "");
shared/tests: add unit tests for new flags of nm_utils_strsplit_set_full() (cherry picked from commit a7d1e14e6deb33dc51902bf76c51755e9915bb1a)
2019-04-16 12:34:11 +02:00
do_test_nm_utils_strsplit_set (FALSE, "\n");
do_test_nm_utils_strsplit_set (TRUE, " \t\n", "\t");
shared: add @allow_escaping argument to @nm_utils_strsplit_set
2018-08-09 09:28:30 +02:00
do_test_nm_utils_strsplit_set (FALSE, "a", "a");
do_test_nm_utils_strsplit_set (FALSE, "a b", "a", "b");
do_test_nm_utils_strsplit_set (FALSE, "a\rb", "a\rb");
do_test_nm_utils_strsplit_set (FALSE, " a\rb ", "a\rb");
do_test_nm_utils_strsplit_set (FALSE, " a bbbd afds ere", "a", "bbbd", "afds", "ere");
do_test_nm_utils_strsplit_set (FALSE,
"1 2 3 4 5 6 7 8 9 0 "
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
"1 2 3 4 5 6 7 8 9 0 "
"1 2 3 4 5 6 7 8 9 0",
"1", "2", "3", "4", "5", "6", "7", "8", "9", "0",
"1", "2", "3", "4", "5", "6", "7", "8", "9", "0",
"1", "2", "3", "4", "5", "6", "7", "8", "9", "0");
shared: add @allow_escaping argument to @nm_utils_strsplit_set
2018-08-09 09:28:30 +02:00
do_test_nm_utils_strsplit_set (TRUE, "\\", "\\");
do_test_nm_utils_strsplit_set (TRUE, "\\ ", "\\ ");
do_test_nm_utils_strsplit_set (TRUE, "\\\\", "\\\\");
do_test_nm_utils_strsplit_set (TRUE, "\\\t", "\\\t");
do_test_nm_utils_strsplit_set (TRUE, "foo\\", "foo\\");
do_test_nm_utils_strsplit_set (TRUE, "bar foo\\", "bar", "foo\\");
do_test_nm_utils_strsplit_set (TRUE, "\\ a b\\ \\ c", "\\ a", "b\\ \\ ", "c");
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
words_exp = g_ptr_array_new_with_free_func (g_free);
for (test_run = 0; test_run < 100; test_run++) {
gboolean f_allow_escaping = nmtst_get_rand_bool ();
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
guint words_len = nmtst_get_rand_uint32 () % 100;
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
gs_free char *str = NULL;
guint i;
g_ptr_array_set_size (words_exp, 0);
for (i = 0; i < words_len; i++) {
guint word_len;
char *word;
guint j;
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
word_len = nmtst_get_rand_uint32 ();
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
if ((word_len % 100) < 30)
word_len = 0;
else
word_len = (word_len >> 10) % 100;
word = g_new (char, word_len + 3);
for (j = 0; j < word_len; ) {
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
guint32 p = nmtst_get_rand_uint32 ();
shared: add NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY flag for nm_utils_strsplit_set_full() Previously, nm_utils_strsplit_set_full() would always remove empty tokens. Add a flag NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY to avoid that. This makes nm_utils_strsplit_set_full() return the same result as g_strsplit_set() and a direct replacement for it -- except for "", where we return %NULL.
2019-04-03 17:32:59 +02:00
static const char delimiters_arr[] = { DELIMITERS_C };
static const char regular_chars[] = "abcdefghijklmnopqrstuvwxyz";
if ( !f_allow_escaping
|| (p % 1000) < 700) {
if (((p >> 20) % 100) < 20)
word[j++] = '\\';
word[j++] = regular_chars[(p >> 11) % (G_N_ELEMENTS (regular_chars) - 1)];
continue;
}
word[j++] = '\\';
word[j++] = delimiters_arr[(p >> 11) % G_N_ELEMENTS (delimiters_arr)];
}
word[j] = '\0';
g_ptr_array_add (words_exp, word);
}
g_ptr_array_add (words_exp, NULL);
str = g_strjoinv (" ", (char **) words_exp->pdata);
if ( str[0] == '\0'
&& words_len > 0) {
g_assert (words_len == 1);
g_assert_cmpstr (words_exp->pdata[0], ==, "");
words_len = 0;
}
_do_test_nm_utils_strsplit_set_f ( (f_allow_escaping ? NM_UTILS_STRSPLIT_SET_FLAGS_ALLOW_ESCAPING : NM_UTILS_STRSPLIT_SET_FLAGS_NONE)
| NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY,
str,
words_len,
(const char *const*) words_exp->pdata);
}
shared/tests: add unit tests for new flags of nm_utils_strsplit_set_full() (cherry picked from commit a7d1e14e6deb33dc51902bf76c51755e9915bb1a)
2019-04-16 12:34:11 +02:00
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED, "\t", "\t");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED | NM_UTILS_STRSPLIT_SET_FLAGS_STRSTRIP, "\t");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED | NM_UTILS_STRSPLIT_SET_FLAGS_STRSTRIP | NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY,
"\t", "");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED | NM_UTILS_STRSPLIT_SET_FLAGS_STRSTRIP | NM_UTILS_STRSPLIT_SET_FLAGS_PRESERVE_EMPTY,
"\t\\\t\t\t\\\t", "\t\t\t\t");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED, "\ta", "\ta");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED | NM_UTILS_STRSPLIT_SET_FLAGS_STRSTRIP, "\ta", "a");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED, "\ta\\ b\t\\ ", "\ta b\t ");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED | NM_UTILS_STRSPLIT_SET_FLAGS_STRSTRIP, "\ta\\ b\t\\ \t", "a b\t ");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED, "a\\ b", "a ", "b");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED, "\ta\\ b", "\ta ", "b");
do_test_nm_utils_strsplit_set_simple (NM_UTILS_STRSPLIT_SET_FLAGS_ESCAPED | NM_UTILS_STRSPLIT_SET_FLAGS_STRSTRIP, "\ta\\ b", "a ", "b");
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
}
/*****************************************************************************/
shared: add nm_utils_escaped_tokens_options_*() API This will be used for splitting and escaping option parameters in nmcli (vpn.data).
2020-03-27 09:29:43 +01:00
static char *
_escaped_tokens_create_random_word_full (const char *const*tokens,
gsize n_tokens,
gsize len)
{
GString *gstr = g_string_new (NULL);
gsize i;
char random_token[2] = { 0 };
for (i = 0; i < len; i++) {
const char *token = tokens[nmtst_get_rand_uint32 () % n_tokens];
if (!token[0]) {
do {
random_token[0] = nmtst_get_rand_uint32 ();
} while (random_token[0] == '\0');
token = random_token;
}
g_string_append (gstr, token);
}
/* reallocate the string, so that we don't have any excess memory from
* the GString buffer. This is so that valgrind may better detect an out
* or range access. */
return nm_str_realloc (g_string_free (gstr, FALSE));
}
/* set to 1 to exclude characters that are annoying to see in the debugger
* and printf() output. */
#define ESCAPED_TOKENS_ONLY_NICE_CHARS 0
static char *
_escaped_tokens_create_random_whitespace (void)
{
static const char *tokens[] = {
" ",
#if !ESCAPED_TOKENS_ONLY_NICE_CHARS
"\n",
"\t",
"\r",
"\f",
#endif
};
return _escaped_tokens_create_random_word_full (tokens, G_N_ELEMENTS (tokens), nmtst_get_rand_word_length (NULL) / 4u);
}
static char *
_escaped_tokens_create_random_word (void)
{
static const char *tokens[] = {
"a",
"b",
"c",
" ",
",",
"=",
"\\",
#if !ESCAPED_TOKENS_ONLY_NICE_CHARS
"\n",
"\f",
":",
"",
#endif
};
return _escaped_tokens_create_random_word_full (tokens, G_N_ELEMENTS (tokens), nmtst_get_rand_word_length (NULL));
}
static void
_escaped_tokens_str_append_delimiter (GString *str,
gboolean strict,
gboolean needs_delimiter)
{
guint len = nmtst_get_rand_word_length (NULL) / 10u;
char *s;
again:
if (!strict) {
g_string_append (str, (s = _escaped_tokens_create_random_whitespace ()));
nm_clear_g_free (&s);
}
if (needs_delimiter)
g_string_append_c (str, ',');
if (!strict) {
g_string_append (str, (s = _escaped_tokens_create_random_whitespace ()));
nm_clear_g_free (&s);
if (len-- > 0) {
needs_delimiter = TRUE;
goto again;
}
}
}
static void
_escaped_tokens_split (char *str, const char **out_key, const char **out_val)
{
const char *key;
const char *val;
gsize len = strlen (str);
g_assert (str);
nm_utils_escaped_tokens_options_split (str, &key, &val);
g_assert (key);
g_assert (key == str);
if (val) {
g_assert (val > str);
g_assert (val > key);
g_assert (val <= &str[len]);
}
NM_SET_OUT (out_key, key);
NM_SET_OUT (out_val, val);
}
static void
_escaped_tokens_combine (GString *combined,
const char *key,
const char *val,
gboolean strict,
gboolean allow_append_delimiter_before,
gboolean needs_delimiter_after)
{
gs_free char *escaped_key = NULL;
gs_free char *escaped_val = NULL;
if (allow_append_delimiter_before)
_escaped_tokens_str_append_delimiter (combined, strict, FALSE);
g_string_append (combined, nm_utils_escaped_tokens_options_escape_key (key, &escaped_key));
if (val) {
char *s;
if (!strict) {
g_string_append (combined, (s = _escaped_tokens_create_random_whitespace ()));
nm_clear_g_free (&s);
}
g_string_append_c (combined, '=');
if (!strict) {
g_string_append (combined, (s = _escaped_tokens_create_random_whitespace ()));
nm_clear_g_free (&s);
}
g_string_append (combined, nm_utils_escaped_tokens_options_escape_val (val, &escaped_val));
}
_escaped_tokens_str_append_delimiter (combined, strict, needs_delimiter_after);
}
static void
_escaped_tokens_check_one_impl (const char *expected_key,
const char *expected_val,
const char *expected_combination,
const char *const*other,
gsize n_other)
{
nm_auto_free_gstring GString *combined = g_string_new (NULL);
gsize i;
g_assert (expected_key);
g_assert (expected_combination);
g_assert (other);
_escaped_tokens_combine (combined,
expected_key,
expected_val,
TRUE,
TRUE,
FALSE);
g_assert_cmpstr (combined->str, ==, expected_combination);
for (i = 0; i < n_other + 2u; i++) {
nm_auto_free_gstring GString *str0 = NULL;
gs_free const char **strv_split = NULL;
gs_free char *strv_split0 = NULL;
const char *comb;
const char *key;
const char *val;
if (i == 0)
comb = expected_combination;
else if (i == 1) {
_escaped_tokens_combine (nm_gstring_prepare (&str0),
expected_key,
expected_val,
FALSE,
TRUE,
FALSE);
comb = str0->str;
} else
comb = other[i - 2];
strv_split = nm_utils_escaped_tokens_options_split_list (comb);
if (!strv_split) {
g_assert_cmpstr (expected_key, ==, "");
g_assert_cmpstr (expected_val, ==, NULL);
continue;
}
g_assert (expected_val || expected_key[0]);
g_assert_cmpuint (NM_PTRARRAY_LEN (strv_split), ==, 1u);
strv_split0 = g_strdup (strv_split[0]);
_escaped_tokens_split (strv_split0, &key, &val);
g_assert_cmpstr (key, ==, expected_key);
g_assert_cmpstr (val, ==, expected_val);
}
}
#define _escaped_tokens_check_one(expected_key, expected_val, expected_combination, ...) \
_escaped_tokens_check_one_impl (expected_key, expected_val, expected_combination, NM_MAKE_STRV (__VA_ARGS__), NM_NARG (__VA_ARGS__))
static void
test_nm_utils_escaped_tokens (void)
{
int i_run;
for (i_run = 0; i_run < 1000; i_run++) {
const guint num_options = nmtst_get_rand_word_length (NULL);
gs_unref_ptrarray GPtrArray *options = g_ptr_array_new_with_free_func (g_free);
nm_auto_free_gstring GString *combined = g_string_new (NULL);
gs_free const char **strv_split = NULL;
guint i_option;
guint i;
/* Generate a list of random words for option key-value pairs. */
for (i_option = 0; i_option < 2u * num_options; i_option++) {
char *word = NULL;
if ( i_option % 2u == 1
&& nmtst_get_rand_uint32 () % 5 == 0
&& strlen (options->pdata[options->len - 1]) > 0u) {
/* For some options, leave the value unset and only generate a key.
*
* If key is "", then we cannot do that, because the test below would try
* to append "" to the combined list, which the parser then would drop.
* Only test omitting the value, if strlen() of the key is positive. */
} else
word = _escaped_tokens_create_random_word ();
g_ptr_array_add (options, word);
}
/* Combine the options in one comma separated list, with proper escaping. */
for (i_option = 0; i_option < num_options; i_option++) {
_escaped_tokens_combine (combined,
options->pdata[2u*i_option + 0u],
options->pdata[2u*i_option + 1u],
FALSE,
i_option == 0,
i_option != num_options - 1);
}
/* ensure that we can split and parse the options without difference. */
strv_split = nm_utils_escaped_tokens_options_split_list (combined->str);
for (i_option = 0; i_option < num_options; i_option++) {
const char *expected_key = options->pdata[2u*i_option + 0u];
const char *expected_val = options->pdata[2u*i_option + 1u];
gs_free char *s_split = i_option < NM_PTRARRAY_LEN (strv_split) ? g_strdup (strv_split[i_option]) : NULL;
const char *key = NULL;
const char *val = NULL;
if (s_split)
_escaped_tokens_split (s_split, &key, &val);
if ( !nm_streq0 (key, expected_key)
|| !nm_streq0 (val, expected_val)) {
g_print (">>> ASSERTION IS ABOUT TO FAIL for item %5d of %5d\n", i_option, num_options);
g_print (">>> combined = \"%s\"\n", combined->str);
g_print (">>> %c parsed[%5d].key = \"%s\"\n", nm_streq (key, expected_key) ? ' ' : 'X', i_option, key);
g_print (">>> %c parsed[%5d].val = %s%s%s\n", nm_streq0 (val, expected_val) ? ' ' : 'X', i_option, NM_PRINT_FMT_QUOTE_STRING (val));
for (i = 0; i < num_options; i++) {
g_print (">>> %c original[%5d].key = \"%s\"\n", i == i_option ? '*' : ' ', i, (char *) options->pdata[2u*i + 0u]);
g_print (">>> %c original[%5d].val = %s%s%s\n", i == i_option ? '*' : ' ', i, NM_PRINT_FMT_QUOTE_STRING ((char *) options->pdata[2u*i + 1u]));
}
for (i = 0; i < NM_PTRARRAY_LEN (strv_split); i++)
g_print (">>> split[%5d] = \"%s\"\n", i, strv_split[i]);
}
g_assert_cmpstr (key, ==, expected_key);
g_assert_cmpstr (val, ==, expected_val);
}
g_assert_cmpint (NM_PTRARRAY_LEN (strv_split), ==, num_options);
/* Above we show a full round-trip of random option key-value pairs, that they can
* without loss escape, concatenate, split-list, and split. This proofed that every
* option key-value pair can be represented as a combined string and parsed back.
*
* Now, just check that we can also parse arbitrary random words in nm_utils_escaped_tokens_options_split().
* split() is a non-injective surjective function. As we check the round-trip above for random words, where
* options-split() is the last step, we show that every random word can be the output of the function
* (which shows, the surjective part).
*
* But multiple random input arguments, may map to the same output argument (non-injective).
* Just test whether we can handle random input words without crashing. For that, just use the
* above generate list of random words.
*/
for (i = 0; i < 1u + 2u * i_option; i++) {
gs_free char *str = NULL;
const char *cstr;
if (i == 0)
cstr = combined->str;
else
cstr = options->pdata[i - 1u];
if (!cstr)
continue;
str = g_strdup (cstr);
_escaped_tokens_split (str, NULL, NULL);
}
}
_escaped_tokens_check_one ("", NULL, "");
_escaped_tokens_check_one ("", "", "=", " =");
_escaped_tokens_check_one ("a", "b", "a=b", "a = b");
_escaped_tokens_check_one ("a\\=", "b\\=", "a\\\\\\==b\\\\=", "a\\\\\\==b\\\\\\=");
_escaped_tokens_check_one ("\\=", "\\=", "\\\\\\==\\\\=", "\\\\\\==\\\\\\=");
_escaped_tokens_check_one (" ", "bb=", "\\ =bb=", "\\ =bb\\=");
_escaped_tokens_check_one (" ", "bb\\=", "\\ =bb\\\\=", "\\ =bb\\\\\\=");
_escaped_tokens_check_one ("a b", "a b", "a b=a b");
_escaped_tokens_check_one ("a b", "a b", "a b=a b");
_escaped_tokens_check_one ("a = b", "a = b", "a \\= b=a = b", "a \\= b=a \\= b");
}
/*****************************************************************************/
c-list: add c_list_sort() Add a stable, recursive merge sort for CList. This could be improved by doing an iterative implementation. The recursive implementation's stack depth is not an issue, as it is bound by O(ln(n)). But an iterative implementation would safe the overhead of O(n*log(n)) function calls and be potentially faster.
2017-07-08 13:30:00 +02:00
typedef struct {
int val;
CList lst;
} CListSort;
static int
_c_list_sort_cmp (const CList *lst_a, const CList *lst_b, const void *user_data)
{
const CListSort *a, *b;
g_assert (lst_a);
g_assert (lst_b);
g_assert (lst_a != lst_b);
a = c_list_entry (lst_a, CListSort, lst);
b = c_list_entry (lst_b, CListSort, lst);
if (a->val < b->val)
return -1;
if (a->val > b->val)
return 1;
return 0;
}
static void
shared/tests: improve tests for c_list_sort() Refactor the tests. Especially, test sorting lists with a length of up to 10000 elements (the actual length is determined randomly on each run).
2017-12-28 19:52:39 +01:00
_do_test_c_list_sort (CListSort *elements, guint n_list, gboolean headless)
c-list: add c_list_sort() Add a stable, recursive merge sort for CList. This could be improved by doing an iterative implementation. The recursive implementation's stack depth is not an issue, as it is bound by O(ln(n)). But an iterative implementation would safe the overhead of O(n*log(n)) function calls and be potentially faster.
2017-07-08 13:30:00 +02:00
{
CList head, *iter, *iter_prev, *lst;
shared/tests: improve tests for c_list_sort() Refactor the tests. Especially, test sorting lists with a length of up to 10000 elements (the actual length is determined randomly on each run).
2017-12-28 19:52:39 +01:00
guint i;
c-list: add c_list_sort() Add a stable, recursive merge sort for CList. This could be improved by doing an iterative implementation. The recursive implementation's stack depth is not an issue, as it is bound by O(ln(n)). But an iterative implementation would safe the overhead of O(n*log(n)) function calls and be potentially faster.
2017-07-08 13:30:00 +02:00
const CListSort *el_prev;
shared/tests: improve tests for c_list_sort() Refactor the tests. Especially, test sorting lists with a length of up to 10000 elements (the actual length is determined randomly on each run).
2017-12-28 19:52:39 +01:00
CListSort *el;
c-list: add c_list_sort() Add a stable, recursive merge sort for CList. This could be improved by doing an iterative implementation. The recursive implementation's stack depth is not an issue, as it is bound by O(ln(n)). But an iterative implementation would safe the overhead of O(n*log(n)) function calls and be potentially faster.
2017-07-08 13:30:00 +02:00
c_list_init (&head);
shared/tests: improve tests for c_list_sort() Refactor the tests. Especially, test sorting lists with a length of up to 10000 elements (the actual length is determined randomly on each run).
2017-12-28 19:52:39 +01:00
for (i = 0; i < n_list; i++) {
el = &elements[i];
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
el->val = nmtst_get_rand_uint32 () % (2*n_list);
shared/tests: improve tests for c_list_sort() Refactor the tests. Especially, test sorting lists with a length of up to 10000 elements (the actual length is determined randomly on each run).
2017-12-28 19:52:39 +01:00
c_list_link_tail (&head, &el->lst);
}
if (headless) {
lst = head.next;
c_list_unlink_stale (&head);
lst = c_list_sort_headless (lst, _c_list_sort_cmp, NULL);
g_assert (lst);
g_assert (lst->next);
g_assert (lst->prev);
g_assert (c_list_length (lst) == n_list - 1);
iter_prev = lst->prev;
for (iter = lst; iter != lst; iter = iter->next) {
g_assert (iter);
g_assert (iter->next);
g_assert (iter->prev == iter_prev);
}
c_list_link_before (lst, &head);
} else
c_list_sort (&head, _c_list_sort_cmp, NULL);
g_assert (!c_list_is_empty (&head));
g_assert (c_list_length (&head) == n_list);
el_prev = NULL;
c_list_for_each (iter, &head) {
el = c_list_entry (iter, CListSort, lst);
g_assert (el >= elements && el < &elements[n_list]);
if (el_prev) {
if (el_prev->val == el->val)
g_assert (el_prev < el);
else
g_assert (el_prev->val < el->val);
g_assert (iter->prev == &el_prev->lst);
g_assert (el_prev->lst.next == iter);
}
el_prev = el;
}
g_assert (head.prev == &el_prev->lst);
}
static void
test_c_list_sort (void)
{
const guint N_ELEMENTS = 10000;
guint n_list, repeat;
gs_free CListSort *elements = NULL;
{
CList head;
c_list_init (&head);
c_list_sort (&head, _c_list_sort_cmp, NULL);
g_assert (c_list_length (&head) == 0);
g_assert (c_list_is_empty (&head));
}
elements = g_new0 (CListSort, N_ELEMENTS);
for (n_list = 1; n_list < N_ELEMENTS; n_list++) {
if (n_list > 150) {
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
n_list += nmtst_get_rand_uint32 () % n_list;
shared/tests: improve tests for c_list_sort() Refactor the tests. Especially, test sorting lists with a length of up to 10000 elements (the actual length is determined randomly on each run).
2017-12-28 19:52:39 +01:00
if (n_list >= N_ELEMENTS)
break;
}
{
const guint N_REPEAT = n_list > 50 ? 1 : 5;
for (repeat = 0; repeat < N_REPEAT; repeat++)
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
_do_test_c_list_sort (elements, n_list, nmtst_get_rand_uint32 () % 2);
c-list: add c_list_sort() Add a stable, recursive merge sort for CList. This could be improved by doing an iterative implementation. The recursive implementation's stack depth is not an issue, as it is bound by O(ln(n)). But an iterative implementation would safe the overhead of O(n*log(n)) function calls and be potentially faster.
2017-07-08 13:30:00 +02:00
}
}
}
/*****************************************************************************/
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
typedef struct {
NMDedupMultiObj parent;
guint val;
guint other;
} DedupObj;
static const NMDedupMultiObjClass dedup_obj_class;
static DedupObj *
_dedup_obj_assert (const NMDedupMultiObj *obj)
{
DedupObj *o;
g_assert (obj);
o = (DedupObj *) obj;
g_assert (o->parent.klass == &dedup_obj_class);
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
g_assert (o->parent._ref_count > 0);
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
g_assert (o->val > 0);
return o;
}
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
static const NMDedupMultiObj *
_dedup_obj_clone (const NMDedupMultiObj *obj)
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
{
DedupObj *o, *o2;
o = _dedup_obj_assert (obj);
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
o2 = g_slice_new0 (DedupObj);
o2->parent.klass = &dedup_obj_class;
o2->parent._ref_count = 1;
o2->val = o->val;
o2->other = o->other;
return (NMDedupMultiObj *) o2;
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
}
static void
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
_dedup_obj_destroy (NMDedupMultiObj *obj)
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
{
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
DedupObj *o = (DedupObj *) obj;
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
nm_assert (o->parent._ref_count == 0);
o->parent._ref_count = 1;
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
o = _dedup_obj_assert (obj);
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
g_slice_free (DedupObj, o);
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
}
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
static void
_dedup_obj_full_hash_update (const NMDedupMultiObj *obj, NMHashState *h)
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
{
const DedupObj *o;
o = _dedup_obj_assert (obj);
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
nm_hash_update_vals (h,
o->val,
o->other);
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
}
static gboolean
_dedup_obj_full_equal (const NMDedupMultiObj *obj_a,
const NMDedupMultiObj *obj_b)
{
const DedupObj *o_a = _dedup_obj_assert (obj_a);
const DedupObj *o_b = _dedup_obj_assert (obj_b);
return o_a->val == o_b->val
&& o_a->other == o_b->other;
}
static const NMDedupMultiObjClass dedup_obj_class = {
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
.obj_clone = _dedup_obj_clone,
.obj_destroy = _dedup_obj_destroy,
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
.obj_full_hash_update = _dedup_obj_full_hash_update,
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
.obj_full_equal = _dedup_obj_full_equal,
};
#define DEDUP_OBJ_INIT(val_val, other_other) \
(&((DedupObj) { \
.parent = { \
.klass = &dedup_obj_class, \
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
._ref_count = NM_OBJ_REF_COUNT_STACKINIT, \
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
}, \
.val = (val_val), \
.other = (other_other), \
}))
typedef struct {
NMDedupMultiIdxType parent;
guint partition_size;
guint val_mod;
} DedupIdxType;
static const NMDedupMultiIdxTypeClass dedup_idx_type_class;
static const DedupIdxType *
_dedup_idx_assert (const NMDedupMultiIdxType *idx_type)
{
DedupIdxType *t;
g_assert (idx_type);
t = (DedupIdxType *) idx_type;
g_assert (t->parent.klass == &dedup_idx_type_class);
g_assert (t->partition_size > 0);
g_assert (t->val_mod > 0);
return t;
}
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
static void
_dedup_idx_obj_id_hash_update (const NMDedupMultiIdxType *idx_type,
const NMDedupMultiObj *obj,
NMHashState *h)
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
{
const DedupIdxType *t;
const DedupObj *o;
t = _dedup_idx_assert (idx_type);
o = _dedup_obj_assert (obj);
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
nm_hash_update_val (h, o->val / t->partition_size);
nm_hash_update_val (h, o->val % t->val_mod);
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
}
static gboolean
_dedup_idx_obj_id_equal (const NMDedupMultiIdxType *idx_type,
const NMDedupMultiObj *obj_a,
const NMDedupMultiObj *obj_b)
{
const DedupIdxType *t;
const DedupObj *o_a;
const DedupObj *o_b;
t = _dedup_idx_assert (idx_type);
o_a = _dedup_obj_assert (obj_a);
o_b = _dedup_obj_assert (obj_b);
return (o_a->val / t->partition_size) == (o_b->val / t->partition_size)
&& (o_a->val % t->val_mod) == (o_b->val % t->val_mod);
}
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
static void
_dedup_idx_obj_partition_hash_update (const NMDedupMultiIdxType *idx_type,
const NMDedupMultiObj *obj,
NMHashState *h)
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
{
const DedupIdxType *t;
const DedupObj *o;
t = _dedup_idx_assert (idx_type);
o = _dedup_obj_assert (obj);
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
nm_hash_update_val (h, o->val / t->partition_size);
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
}
static gboolean
_dedup_idx_obj_partition_equal (const NMDedupMultiIdxType *idx_type,
const NMDedupMultiObj *obj_a,
const NMDedupMultiObj *obj_b)
{
const DedupIdxType *t;
const DedupObj *o_a;
const DedupObj *o_b;
t = _dedup_idx_assert (idx_type);
o_a = _dedup_obj_assert (obj_a);
o_b = _dedup_obj_assert (obj_b);
return (o_a->val / t->partition_size) == (o_b->val / t->partition_size);
}
static const NMDedupMultiIdxTypeClass dedup_idx_type_class = {
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
.idx_obj_id_hash_update = _dedup_idx_obj_id_hash_update,
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
.idx_obj_id_equal = _dedup_idx_obj_id_equal,
all: extend hash functions with an NMHashState argument We often want to cascade hashing, meaning, to combine the outcome of various hash functions in a larger hash. Instead of having each hash function return a guint hash value, accept a hash state argument. This saves the overhead of initializing and completing the intermediate hash states. It also avoids loosing entropy when we reduce the larger hash state into the intermediate guint hash value.
2017-10-17 11:53:08 +02:00
.idx_obj_partition_hash_update = _dedup_idx_obj_partition_hash_update,
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
.idx_obj_partition_equal = _dedup_idx_obj_partition_equal,
};
static const DedupIdxType *
DEDUP_IDX_TYPE_INIT (DedupIdxType *idx_type, guint partition_size, guint val_mod)
{
nm_dedup_multi_idx_type_init ((NMDedupMultiIdxType *) idx_type, &dedup_idx_type_class);
idx_type->val_mod = val_mod;
idx_type->partition_size = partition_size;
return idx_type;
}
static gboolean
_dedup_idx_add (NMDedupMultiIndex *idx, const DedupIdxType *idx_type, const DedupObj *obj, NMDedupMultiIdxMode mode, const NMDedupMultiEntry **out_entry)
{
g_assert (idx);
_dedup_idx_assert ((NMDedupMultiIdxType *) idx_type);
if (obj)
_dedup_obj_assert ((NMDedupMultiObj *) obj);
return nm_dedup_multi_index_add (idx, (NMDedupMultiIdxType *) idx_type,
obj, mode, out_entry, NULL);
}
static void
_dedup_head_entry_assert (const NMDedupMultiHeadEntry *entry)
{
g_assert (entry);
g_assert (entry->len > 0);
g_assert (entry->len == c_list_length (&entry->lst_entries_head));
g_assert (entry->idx_type);
g_assert (entry->is_head);
}
static const DedupObj *
_dedup_entry_assert (const NMDedupMultiEntry *entry)
{
g_assert (entry);
g_assert (!c_list_is_empty (&entry->lst_entries));
g_assert (entry->head);
g_assert (!entry->is_head);
g_assert (entry->head != (gpointer) entry);
_dedup_head_entry_assert (entry->head);
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
return _dedup_obj_assert (entry->obj);
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
}
static const DedupIdxType *
_dedup_entry_get_idx_type (const NMDedupMultiEntry *entry)
{
_dedup_entry_assert (entry);
g_assert (entry->head);
g_assert (entry->head->idx_type);
return _dedup_idx_assert (entry->head->idx_type);
}
static void
_dedup_entry_assert_all (const NMDedupMultiEntry *entry, gssize expected_idx, const DedupObj *const*expected_obj)
{
gsize n, i;
CList *iter;
g_assert (entry);
_dedup_entry_assert (entry);
g_assert (expected_obj);
n = NM_PTRARRAY_LEN (expected_obj);
g_assert (n == c_list_length (&entry->lst_entries));
g_assert (expected_idx >= -1 && expected_idx < n);
g_assert (entry->head);
if (expected_idx == -1)
g_assert (entry->head == (gpointer) entry);
else
g_assert (entry->head != (gpointer) entry);
i = 0;
c_list_for_each (iter, &entry->head->lst_entries_head) {
const NMDedupMultiEntry *entry_current = c_list_entry (iter, NMDedupMultiEntry, lst_entries);
const DedupObj *obj_current;
const DedupIdxType *idx_type = _dedup_entry_get_idx_type (entry_current);
obj_current = _dedup_entry_assert (entry_current);
g_assert (obj_current);
g_assert (i < n);
if (expected_idx == i)
g_assert (entry_current == entry);
g_assert (idx_type->parent.klass->idx_obj_partition_equal (&idx_type->parent,
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
entry_current->obj,
c_list_entry (entry->head->lst_entries_head.next, NMDedupMultiEntry, lst_entries)->obj));
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
i++;
}
}
#define _dedup_entry_assert_all(entry, expected_idx, ...) _dedup_entry_assert_all (entry, expected_idx, (const DedupObj *const[]) { __VA_ARGS__, NULL })
static void
test_dedup_multi (void)
{
NMDedupMultiIndex *idx;
DedupIdxType IDX_20_3_a_stack;
const DedupIdxType *const IDX_20_3_a = DEDUP_IDX_TYPE_INIT (&IDX_20_3_a_stack, 20, 3);
const NMDedupMultiEntry *entry1;
idx = nm_dedup_multi_index_new ();
g_assert (_dedup_idx_add (idx, IDX_20_3_a, DEDUP_OBJ_INIT (1, 1), NM_DEDUP_MULTI_IDX_MODE_APPEND, &entry1));
_dedup_entry_assert_all (entry1, 0, DEDUP_OBJ_INIT (1, 1));
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
g_assert (nm_dedup_multi_index_obj_find (idx, (NMDedupMultiObj *) DEDUP_OBJ_INIT (1, 1)));
g_assert (!nm_dedup_multi_index_obj_find (idx, (NMDedupMultiObj *) DEDUP_OBJ_INIT (1, 2)));
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
g_assert (_dedup_idx_add (idx, IDX_20_3_a, DEDUP_OBJ_INIT (1, 2), NM_DEDUP_MULTI_IDX_MODE_APPEND, &entry1));
_dedup_entry_assert_all (entry1, 0, DEDUP_OBJ_INIT (1, 2));
core: remove NMDedupMultiBox object and track NMDedupMultiObj instances directly Implement the reference counting of NMPObject as part of NMDedupMultiObj and get rid of NMDedupMultiBox. With this change, the NMPObject is aware in which NMDedupMultiIndex instance it is tracked. - this saves an additional GSlice allocation for the NMDedupMultiBox. - it is immediately known, whether an NMPObject is tracked by a certain NMDedupMultiIndex or not. This saves an additional hash lookup. - previously, when all idx-types cease to reference an NMDedupMultiObj instance, it was removed. Now, a tracked objects stays in the NMDedupMultiIndex until it's last reference is deleted. This possibly extends the lifetime of the object and we may reuse it better. - it is no longer possible to add one object to more then one NMDedupMultiIndex instance. As we anyway want to have only one instance to deduplicate the objects, this is fine. - the ref-counting implementation is now part of NMDedupMultiObj. Previously, NMDedupMultiIndex could also track objects that were not ref-counted. Hoever, the object anyway *must* implement the NMDedupMultiObj API, so this flexibility is unneeded and was not used. - a downside is, that NMPObject grows by one pointer size, even if it isn't tracked in the NMDedupMultiIndex. But we really want to put all objects into the index for sharing and deduplication. So this downside should be acceptable. Still, code like nmp_object_stackinit*() needs to handle a larger object.
2017-07-02 23:46:06 +02:00
g_assert (!nm_dedup_multi_index_obj_find (idx, (NMDedupMultiObj *) DEDUP_OBJ_INIT (1, 1)));
g_assert (nm_dedup_multi_index_obj_find (idx, (NMDedupMultiObj *) DEDUP_OBJ_INIT (1, 2)));
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
g_assert (_dedup_idx_add (idx, IDX_20_3_a, DEDUP_OBJ_INIT (2, 2), NM_DEDUP_MULTI_IDX_MODE_APPEND, &entry1));
_dedup_entry_assert_all (entry1, 1, DEDUP_OBJ_INIT (1, 2), DEDUP_OBJ_INIT (2, 2));
nm_dedup_multi_index_unref (idx);
}
/*****************************************************************************/
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
static NMConnection *
_connection_new_from_dbus (GVariant *dict, GError **error)
{
return _nm_simple_connection_new_from_dbus (dict, NM_SETTING_PARSE_FLAGS_NORMALIZE, error);
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
vpn_check_func (const char *key, const char *value, gpointer user_data)
{
if (!strcmp (key, "foobar1")) {
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpstr (value, ==, "blahblah1");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
return;
}
if (!strcmp (key, "foobar2")) {
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpstr (value, ==, "blahblah2");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
return;
}
if (!strcmp (key, "foobar3")) {
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpstr (value, ==, "blahblah3");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
return;
}
if (!strcmp (key, "foobar4")) {
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpstr (value, ==, "blahblah4");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
return;
}
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_not_reached ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
vpn_check_empty_func (const char *key, const char *value, gpointer user_data)
{
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_not_reached ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
test_setting_vpn_items (void)
{
libnm,core: add _nm_connection_aggregate() to replace nm_connection_for_each_setting_value() We should no longer use nm_connection_for_each_setting_value() and nm_setting_for_each_value(). It's fundamentally broken as it does not work with properties that are not backed by a GObject property and it cannot be fixed because it is public API. Add an internal function _nm_connection_aggregate() to replace it. Compare the implementation of the aggregation functionality inside libnm with the previous two checks for secret-flags that it replaces: - previous approach broke abstraction and require detailed knowledge of secret flags. Meaning, they must special case NMSettingVpn and GObject-property based secrets. If we implement a new way for implementing secrets (like we will need for WireGuard), then this the new way should only affect libnm-core, not require changes elsewhere. - it's very inefficient to itereate over all settings. It involves cloning and sorting the list of settings, and retrieve and clone all GObject properties. Only to look at secret properties alone. _nm_connection_aggregate() is supposed to be more flexible then just the two new aggregate types that perform a "find-any" search. The @arg argument and boolean return value can suffice to implement different aggregation types in the future. Also fixes the check of NMAgentManager for secret flags for VPNs (NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS). A secret for VPNs is a property that either has a secret or a secret-flag. The previous implementation would only look at present secrets and check their flags. It wouldn't check secret-flags that are NM_SETTING_SECRET_FLAG_NONE, but have no secret.
2019-01-04 11:28:27 +01:00
gs_unref_object NMConnection *connection = NULL;
NMSettingVpn *s_vpn;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm,core: add _nm_connection_aggregate() to replace nm_connection_for_each_setting_value() We should no longer use nm_connection_for_each_setting_value() and nm_setting_for_each_value(). It's fundamentally broken as it does not work with properties that are not backed by a GObject property and it cannot be fixed because it is public API. Add an internal function _nm_connection_aggregate() to replace it. Compare the implementation of the aggregation functionality inside libnm with the previous two checks for secret-flags that it replaces: - previous approach broke abstraction and require detailed knowledge of secret flags. Meaning, they must special case NMSettingVpn and GObject-property based secrets. If we implement a new way for implementing secrets (like we will need for WireGuard), then this the new way should only affect libnm-core, not require changes elsewhere. - it's very inefficient to itereate over all settings. It involves cloning and sorting the list of settings, and retrieve and clone all GObject properties. Only to look at secret properties alone. _nm_connection_aggregate() is supposed to be more flexible then just the two new aggregate types that perform a "find-any" search. The @arg argument and boolean return value can suffice to implement different aggregation types in the future. Also fixes the check of NMAgentManager for secret flags for VPNs (NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS). A secret for VPNs is a property that either has a secret or a secret-flag. The previous implementation would only look at present secrets and check their flags. It wouldn't check secret-flags that are NM_SETTING_SECRET_FLAG_NONE, but have no secret.
2019-01-04 11:28:27 +01:00
connection = nmtst_create_minimal_connection ("vpn-items", NULL, NM_SETTING_VPN_SETTING_NAME, NULL);
s_vpn = nm_connection_get_setting_vpn (connection);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_data_item (s_vpn, "foobar1", "blahblah1");
nm_setting_vpn_add_data_item (s_vpn, "foobar2", "blahblah2");
nm_setting_vpn_add_data_item (s_vpn, "foobar3", "blahblah3");
nm_setting_vpn_add_data_item (s_vpn, "foobar4", "blahblah4");
/* Ensure that added values are all present */
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
nm_setting_vpn_foreach_data_item (s_vpn, vpn_check_func, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_remove_data_item (s_vpn, "foobar1");
nm_setting_vpn_remove_data_item (s_vpn, "foobar2");
nm_setting_vpn_remove_data_item (s_vpn, "foobar3");
nm_setting_vpn_remove_data_item (s_vpn, "foobar4");
libnm,core: add _nm_connection_aggregate() to replace nm_connection_for_each_setting_value() We should no longer use nm_connection_for_each_setting_value() and nm_setting_for_each_value(). It's fundamentally broken as it does not work with properties that are not backed by a GObject property and it cannot be fixed because it is public API. Add an internal function _nm_connection_aggregate() to replace it. Compare the implementation of the aggregation functionality inside libnm with the previous two checks for secret-flags that it replaces: - previous approach broke abstraction and require detailed knowledge of secret flags. Meaning, they must special case NMSettingVpn and GObject-property based secrets. If we implement a new way for implementing secrets (like we will need for WireGuard), then this the new way should only affect libnm-core, not require changes elsewhere. - it's very inefficient to itereate over all settings. It involves cloning and sorting the list of settings, and retrieve and clone all GObject properties. Only to look at secret properties alone. _nm_connection_aggregate() is supposed to be more flexible then just the two new aggregate types that perform a "find-any" search. The @arg argument and boolean return value can suffice to implement different aggregation types in the future. Also fixes the check of NMAgentManager for secret flags for VPNs (NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS). A secret for VPNs is a property that either has a secret or a secret-flag. The previous implementation would only look at present secrets and check their flags. It wouldn't check secret-flags that are NM_SETTING_SECRET_FLAG_NONE, but have no secret.
2019-01-04 11:28:27 +01:00
g_assert (!_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SECRETS, NULL));
g_assert (!_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_secret (s_vpn, "foobar1", "blahblah1");
libnm,core: add _nm_connection_aggregate() to replace nm_connection_for_each_setting_value() We should no longer use nm_connection_for_each_setting_value() and nm_setting_for_each_value(). It's fundamentally broken as it does not work with properties that are not backed by a GObject property and it cannot be fixed because it is public API. Add an internal function _nm_connection_aggregate() to replace it. Compare the implementation of the aggregation functionality inside libnm with the previous two checks for secret-flags that it replaces: - previous approach broke abstraction and require detailed knowledge of secret flags. Meaning, they must special case NMSettingVpn and GObject-property based secrets. If we implement a new way for implementing secrets (like we will need for WireGuard), then this the new way should only affect libnm-core, not require changes elsewhere. - it's very inefficient to itereate over all settings. It involves cloning and sorting the list of settings, and retrieve and clone all GObject properties. Only to look at secret properties alone. _nm_connection_aggregate() is supposed to be more flexible then just the two new aggregate types that perform a "find-any" search. The @arg argument and boolean return value can suffice to implement different aggregation types in the future. Also fixes the check of NMAgentManager for secret flags for VPNs (NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS). A secret for VPNs is a property that either has a secret or a secret-flag. The previous implementation would only look at present secrets and check their flags. It wouldn't check secret-flags that are NM_SETTING_SECRET_FLAG_NONE, but have no secret.
2019-01-04 11:28:27 +01:00
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SECRETS, NULL));
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_secret (s_vpn, "foobar2", "blahblah2");
nm_setting_vpn_add_secret (s_vpn, "foobar3", "blahblah3");
nm_setting_vpn_add_secret (s_vpn, "foobar4", "blahblah4");
/* Ensure that added values are all present */
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
nm_setting_vpn_foreach_secret (s_vpn, vpn_check_func, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_remove_secret (s_vpn, "foobar1");
nm_setting_vpn_remove_secret (s_vpn, "foobar2");
nm_setting_vpn_remove_secret (s_vpn, "foobar3");
libnm,core: add _nm_connection_aggregate() to replace nm_connection_for_each_setting_value() We should no longer use nm_connection_for_each_setting_value() and nm_setting_for_each_value(). It's fundamentally broken as it does not work with properties that are not backed by a GObject property and it cannot be fixed because it is public API. Add an internal function _nm_connection_aggregate() to replace it. Compare the implementation of the aggregation functionality inside libnm with the previous two checks for secret-flags that it replaces: - previous approach broke abstraction and require detailed knowledge of secret flags. Meaning, they must special case NMSettingVpn and GObject-property based secrets. If we implement a new way for implementing secrets (like we will need for WireGuard), then this the new way should only affect libnm-core, not require changes elsewhere. - it's very inefficient to itereate over all settings. It involves cloning and sorting the list of settings, and retrieve and clone all GObject properties. Only to look at secret properties alone. _nm_connection_aggregate() is supposed to be more flexible then just the two new aggregate types that perform a "find-any" search. The @arg argument and boolean return value can suffice to implement different aggregation types in the future. Also fixes the check of NMAgentManager for secret flags for VPNs (NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS). A secret for VPNs is a property that either has a secret or a secret-flag. The previous implementation would only look at present secrets and check their flags. It wouldn't check secret-flags that are NM_SETTING_SECRET_FLAG_NONE, but have no secret.
2019-01-04 11:28:27 +01:00
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SECRETS, NULL));
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
nm_setting_vpn_add_data_item (s_vpn, "foobar4-flags", "blahblah4");
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
nm_setting_vpn_add_data_item (s_vpn, "foobar4-flags", "2");
g_assert (!_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_remove_secret (s_vpn, "foobar4");
libnm,core: add _nm_connection_aggregate() to replace nm_connection_for_each_setting_value() We should no longer use nm_connection_for_each_setting_value() and nm_setting_for_each_value(). It's fundamentally broken as it does not work with properties that are not backed by a GObject property and it cannot be fixed because it is public API. Add an internal function _nm_connection_aggregate() to replace it. Compare the implementation of the aggregation functionality inside libnm with the previous two checks for secret-flags that it replaces: - previous approach broke abstraction and require detailed knowledge of secret flags. Meaning, they must special case NMSettingVpn and GObject-property based secrets. If we implement a new way for implementing secrets (like we will need for WireGuard), then this the new way should only affect libnm-core, not require changes elsewhere. - it's very inefficient to itereate over all settings. It involves cloning and sorting the list of settings, and retrieve and clone all GObject properties. Only to look at secret properties alone. _nm_connection_aggregate() is supposed to be more flexible then just the two new aggregate types that perform a "find-any" search. The @arg argument and boolean return value can suffice to implement different aggregation types in the future. Also fixes the check of NMAgentManager for secret flags for VPNs (NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS). A secret for VPNs is a property that either has a secret or a secret-flag. The previous implementation would only look at present secrets and check their flags. It wouldn't check secret-flags that are NM_SETTING_SECRET_FLAG_NONE, but have no secret.
2019-01-04 11:28:27 +01:00
g_assert (!_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SECRETS, NULL));
g_assert (!_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
nm_setting_vpn_remove_data_item (s_vpn, "foobar4-flags");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Try to add some blank values and make sure they are rejected */
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_data_item (s_vpn, NULL, NULL);
g_test_assert_expected_messages ();
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_data_item (s_vpn, "", "");
g_test_assert_expected_messages ();
nm_setting_vpn_add_data_item (s_vpn, "foobar1", "");
libnm: allow setting empty vpn.data item Until now, nm_setting_vpn_add_data_item() would reject empty data values. This leads for example to an assertion failure, if you write a keyfile that assigns an empty value to a key. Keyfile reader would not check that the value is non-empty before calling nm_setting_vpn_add_data_item(). Anyway, I think we should not require having non-empty data elements. It's an unnecessary and sometimes harmful restriction. NetworkManager doesn't understand not care about the content of the vpn data. That is up the VPN plugins. Sometimes and empty value may be desirable. Also, the NM_SETTING_VPN_DATA property setter wouldn't filter out empty values either. So it was always possible to use some libnm API to set data with empty values. The restriction in nm_setting_vpn_add_data_item() was inconsistent.
2020-03-26 12:41:13 +01:00
g_assert_cmpstr (nm_setting_vpn_get_data_item (s_vpn, "foobar1"), ==, "");
nm_setting_vpn_add_data_item (s_vpn, "foobar1", NULL);
g_assert_cmpstr (nm_setting_vpn_get_data_item (s_vpn, "foobar1"), ==, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_data_item (s_vpn, NULL, "blahblah1");
g_test_assert_expected_messages ();
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_data_item (s_vpn, "", "blahblah1");
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
nm_setting_vpn_foreach_data_item (s_vpn, vpn_check_empty_func, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Try to add some blank secrets and make sure they are rejected */
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_secret (s_vpn, NULL, NULL);
g_test_assert_expected_messages ();
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_secret (s_vpn, "", "");
g_test_assert_expected_messages ();
nm_setting_vpn_add_secret (s_vpn, "foobar1", "");
libnm: allow setting empty vpn.secrets item Like for data, now also allow empty secrets to be added to the VPN setting. For one, this avoids an assertion failure, where keyfile reader wouldn't check whether a secret key is set to the empty word. For data, it's more clear that we want to allow setting empty data values. VPN settings are only interpreted by the VPN plugin, so libnm and the daemon shouldn't prevent empty settings. It can be useful to distinguish between unset (NULL) and empty values. For secrets, it's less clear that empty secrets shall be allowed. I think it should. Of course, the empty secret likely isn't a correct nor valid secret. But libnm cannot validate the secrets anyway. It's up to the VPN plugin to handle this in any way they see fit. Also, already before, the user could set NM_SETTING_VPN_SECRETS to a string dictionary with empty passwords. So, the API didn't fully prevent that. Only certain API wouldn't play along.
2020-03-26 12:41:13 +01:00
nm_setting_vpn_add_secret (s_vpn, "foobar1", NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_secret (s_vpn, NULL, "blahblah1");
g_test_assert_expected_messages ();
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (key && key[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_add_secret (s_vpn, "", "blahblah1");
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
nm_setting_vpn_foreach_secret (s_vpn, vpn_check_empty_func, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
test_setting_vpn_update_secrets (void)
{
NMConnection *connection;
libnm, core, cli, tui: fix the capitalization of various types GLib/Gtk have mostly settled on the convention that two-letter acronyms in type names remain all-caps (eg, "IO"), but longer acronyms become initial-caps-only (eg, "Tcp"). NM was inconsistent, with most long acronyms using initial caps only (Adsl, Cdma, Dcb, Gsm, Olpc, Vlan), but others using all caps (DHCP, PPP, PPPOE, VPN). Fix libnm and src/ to use initial-caps only for all three-or-more-letter-long acronyms (and update nmcli and nmtui for the libnm changes).
2014-06-26 13:44:36 -04:00
NMSettingVpn *s_vpn;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariantBuilder settings_builder, vpn_builder, secrets_builder;
GVariant *settings;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
gboolean success;
GError *error = NULL;
const char *tmp;
const char *key1 = "foobar";
const char *key2 = "blahblah";
const char *val1 = "value1";
const char *val2 = "value2";
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm, core, cli, tui: fix the capitalization of various types GLib/Gtk have mostly settled on the convention that two-letter acronyms in type names remain all-caps (eg, "IO"), but longer acronyms become initial-caps-only (eg, "Tcp"). NM was inconsistent, with most long acronyms using initial caps only (Adsl, Cdma, Dcb, Gsm, Olpc, Vlan), but others using all caps (DHCP, PPP, PPPOE, VPN). Fix libnm and src/ to use initial-caps only for all three-or-more-letter-long acronyms (and update nmcli and nmtui for the libnm changes).
2014-06-26 13:44:36 -04:00
s_vpn = (NMSettingVpn *) nm_setting_vpn_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, NM_SETTING (s_vpn));
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_init (&settings_builder, NM_VARIANT_TYPE_CONNECTION);
g_variant_builder_init (&vpn_builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_init (&secrets_builder, G_VARIANT_TYPE ("a{ss}"));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_add (&secrets_builder, "{ss}", key1, val1);
g_variant_builder_add (&secrets_builder, "{ss}", key2, val2);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_add (&vpn_builder, "{sv}",
NM_SETTING_VPN_SECRETS,
g_variant_builder_end (&secrets_builder));
g_variant_builder_add (&settings_builder, "{sa{sv}}",
NM_SETTING_VPN_SETTING_NAME,
&vpn_builder);
settings = g_variant_builder_end (&settings_builder);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_connection_update_secrets (connection, NM_SETTING_VPN_SETTING_NAME, settings, &error);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_no_error (error);
g_assert (success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Read the secrets back out */
tmp = nm_setting_vpn_get_secret (s_vpn, key1);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (tmp);
g_assert_cmpstr (tmp, ==, val1);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tmp = nm_setting_vpn_get_secret (s_vpn, key2);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (tmp);
g_assert_cmpstr (tmp, ==, val2);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (settings);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (connection);
}
#define TO_DEL_NUM 50
typedef struct {
libnm, core, cli, tui: fix the capitalization of various types GLib/Gtk have mostly settled on the convention that two-letter acronyms in type names remain all-caps (eg, "IO"), but longer acronyms become initial-caps-only (eg, "Tcp"). NM was inconsistent, with most long acronyms using initial caps only (Adsl, Cdma, Dcb, Gsm, Olpc, Vlan), but others using all caps (DHCP, PPP, PPPOE, VPN). Fix libnm and src/ to use initial-caps only for all three-or-more-letter-long acronyms (and update nmcli and nmtui for the libnm changes).
2014-06-26 13:44:36 -04:00
NMSettingVpn *s_vpn;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
char *to_del[TO_DEL_NUM];
guint called;
} IterInfo;
static void
del_iter_func (const char *key, const char *value, gpointer user_data)
{
IterInfo *info = user_data;
int i;
/* Record how many times this function gets called; it should get called
* exactly as many times as there are keys in the hash table, regardless
* of what keys we delete from the table.
*/
info->called++;
/* During the iteration, remove a bunch of stuff from the table */
if (info->called == 1) {
for (i = 0; i < TO_DEL_NUM; i++)
nm_setting_vpn_remove_data_item (info->s_vpn, info->to_del[i]);
}
}
static void
test_setting_vpn_modify_during_foreach (void)
{
libnm, core, cli, tui: fix the capitalization of various types GLib/Gtk have mostly settled on the convention that two-letter acronyms in type names remain all-caps (eg, "IO"), but longer acronyms become initial-caps-only (eg, "Tcp"). NM was inconsistent, with most long acronyms using initial caps only (Adsl, Cdma, Dcb, Gsm, Olpc, Vlan), but others using all caps (DHCP, PPP, PPPOE, VPN). Fix libnm and src/ to use initial-caps only for all three-or-more-letter-long acronyms (and update nmcli and nmtui for the libnm changes).
2014-06-26 13:44:36 -04:00
NMSettingVpn *s_vpn;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
IterInfo info;
char *key, *val;
int i, u = 0;
libnm, core, cli, tui: fix the capitalization of various types GLib/Gtk have mostly settled on the convention that two-letter acronyms in type names remain all-caps (eg, "IO"), but longer acronyms become initial-caps-only (eg, "Tcp"). NM was inconsistent, with most long acronyms using initial caps only (Adsl, Cdma, Dcb, Gsm, Olpc, Vlan), but others using all caps (DHCP, PPP, PPPOE, VPN). Fix libnm and src/ to use initial-caps only for all three-or-more-letter-long acronyms (and update nmcli and nmtui for the libnm changes).
2014-06-26 13:44:36 -04:00
s_vpn = (NMSettingVpn *) nm_setting_vpn_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (s_vpn);
for (i = 0; i < TO_DEL_NUM * 2; i++) {
key = g_strdup_printf ("adsfasdfadf%d", i);
val = g_strdup_printf ("42263236236awt%d", i);
nm_setting_vpn_add_data_item (s_vpn, key, val);
/* Cache some keys to delete */
if (i % 2)
info.to_del[u++] = g_strdup (key);
g_free (key);
g_free (val);
}
/* Iterate over current table keys */
info.s_vpn = s_vpn;
info.called = 0;
nm_setting_vpn_foreach_data_item (s_vpn, del_iter_func, &info);
/* Make sure all the things we removed during iteration are really gone */
for (i = 0; i < TO_DEL_NUM; i++) {
g_assert_cmpstr (nm_setting_vpn_get_data_item (s_vpn, info.to_del[i]), ==, NULL);
g_free (info.to_del[i]);
}
/* And make sure the foreach callback was called the same number of times
* as there were keys in the table at the beginning of the foreach.
*/
g_assert_cmpint (info.called, ==, TO_DEL_NUM * 2);
g_object_unref (s_vpn);
}
static void
test_setting_ip4_config_labels (void)
{
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NMSettingIPConfig *s_ip4;
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
NMIPAddress *addr;
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
GVariant *label;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
GPtrArray *addrs;
libnm-core: change list-of-string and array-of-string properties to G_TYPE_STRV Change all DBUS_TYPE_G_LIST_OF_STRING and DBUS_TYPE_G_ARRAY_OF_STRING properties to G_TYPE_STRV, and update everything accordingly. (This doesn't actually require using _nm_setting_class_transform_property(); dbus-glib is happy to transform between 'as' and G_TYPE_STRV.)
2014-08-21 13:19:53 -04:00
char **labels;
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
NMConnection *conn;
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
GVariant *dict, *dict2, *setting_dict, *value;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
GError *error = NULL;
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (G_OBJECT (s_ip4),
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_MANUAL,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NULL);
/* addr 1 */
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
addr = nm_ip_address_new (AF_INET, "1.2.3.4", 24, &error);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
g_assert_no_error (error);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_address (s_ip4, addr);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
nm_ip_address_unref (addr);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 0);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label == NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: don't serialize empty address-labels If no address in an NMSettingIP4Config has a label, then don't bother serializing an array of empty strings.
2014-10-28 12:49:56 -04:00
/* The 'address-labels' property should be omitted from the serialization if
* there are no non-NULL labels.
*/
conn = nmtst_create_minimal_connection ("label test", NULL, NM_SETTING_WIRED_SETTING_NAME, NULL);
nm_connection_add_setting (conn, nm_setting_duplicate (NM_SETTING (s_ip4)));
dict = nm_connection_to_dbus (conn, NM_CONNECTION_SERIALIZE_ALL);
g_object_unref (conn);
setting_dict = g_variant_lookup_value (dict, NM_SETTING_IP4_CONFIG_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (setting_dict != NULL);
value = g_variant_lookup_value (setting_dict, "address-labels", NULL);
g_assert (value == NULL);
g_variant_unref (setting_dict);
g_variant_unref (dict);
/* Now back to constructing the original s_ip4... */
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* addr 2 */
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
addr = nm_ip_address_new (AF_INET, "2.3.4.5", 24, &error);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
g_assert_no_error (error);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
nm_ip_address_set_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL, g_variant_new_string ("eth0:1"));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_address (s_ip4, addr);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
nm_ip_address_unref (addr);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 1);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label != NULL);
g_assert_cmpstr (g_variant_get_string (label, NULL), ==, "eth0:1");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* addr 3 */
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
addr = nm_ip_address_new (AF_INET, "3.4.5.6", 24, &error);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
g_assert_no_error (error);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
nm_ip_address_set_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_address (s_ip4, addr);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
nm_ip_address_unref (addr);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 2);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label == NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Remove addr 1 and re-verify remaining addresses */
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_remove_address (s_ip4, 0);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 0);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "2.3.4.5");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label != NULL);
g_assert_cmpstr (g_variant_get_string (label, NULL), ==, "eth0:1");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 1);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "3.4.5.6");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label == NULL);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
/* If we serialize as the daemon, the labels should appear in the D-Bus
* serialization under both 'address-labels' and 'address-data'.
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
*/
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
conn = nmtst_create_minimal_connection ("label test", NULL, NM_SETTING_WIRED_SETTING_NAME, NULL);
nm_connection_add_setting (conn, NM_SETTING (s_ip4));
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
_nm_utils_is_manager_process = TRUE;
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
dict = nm_connection_to_dbus (conn, NM_CONNECTION_SERIALIZE_ALL);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
_nm_utils_is_manager_process = FALSE;
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_object_unref (conn);
setting_dict = g_variant_lookup_value (dict, NM_SETTING_IP4_CONFIG_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (setting_dict != NULL);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
value = g_variant_lookup_value (setting_dict, "address-labels", G_VARIANT_TYPE_STRING_ARRAY);
g_assert (value != NULL);
g_variant_get (value, "^as", &labels);
g_assert_cmpint (g_strv_length (labels), ==, 2);
g_assert_cmpstr (labels[0], ==, "eth0:1");
g_assert_cmpstr (labels[1], ==, "");
g_variant_unref (value);
g_strfreev (labels);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
value = g_variant_lookup_value (setting_dict, "address-data", G_VARIANT_TYPE ("aa{sv}"));
addrs = nm_utils_ip_addresses_from_variant (value, AF_INET);
g_variant_unref (value);
g_assert (addrs != NULL);
g_assert_cmpint (addrs->len, ==, 2);
addr = addrs->pdata[0];
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
g_assert (label != NULL);
g_assert_cmpstr (g_variant_get_string (label, NULL), ==, "eth0:1");
addr = addrs->pdata[1];
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
g_assert (label == NULL);
g_ptr_array_unref (addrs);
g_variant_unref (setting_dict);
/* We should be able to deserialize the labels from either 'address-labels'
* or 'address-data'.
*/
dict2 = g_variant_ref (dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_DROP_PROPERTY (NM_SETTING_IP4_CONFIG_SETTING_NAME,
"address-data");
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert_no_error (error);
g_variant_unref (dict);
s_ip4 = nm_connection_get_setting_ip4_config (conn);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 0);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "2.3.4.5");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label != NULL);
g_assert_cmpstr (g_variant_get_string (label, NULL), ==, "eth0:1");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 1);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "3.4.5.6");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label == NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
g_object_unref (conn);
NMTST_VARIANT_EDITOR (dict2,
NMTST_VARIANT_DROP_PROPERTY (NM_SETTING_IP4_CONFIG_SETTING_NAME,
"address-labels");
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict2, &error);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
g_assert_no_error (error);
g_variant_unref (dict2);
s_ip4 = nm_connection_get_setting_ip4_config (conn);
addr = nm_setting_ip_config_get_address (s_ip4, 0);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "2.3.4.5");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
g_assert_cmpstr (g_variant_get_string (label, NULL), ==, "eth0:1");
addr = nm_setting_ip_config_get_address (s_ip4, 1);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "3.4.5.6");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
g_assert (label == NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Test explicit property assignment */
g_object_get (G_OBJECT (s_ip4),
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_ADDRESSES, &addrs,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NULL);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_clear_addresses (s_ip4);
g_assert_cmpint (nm_setting_ip_config_get_num_addresses (s_ip4), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (G_OBJECT (s_ip4),
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_ADDRESSES, addrs,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NULL);
libnm-core: improve NMSettingIP4Config / NMSettingIP6Config property types Make the :addresses and :routes properties be GPtrArrays of NMIP4Address, etc, rather than just reflecting the D-Bus data. Make the :dns properties be arrays of strings rather than arrays of binary IP addresses (and update the corresponding APIs as well).
2014-06-24 12:46:03 -04:00
g_ptr_array_unref (addrs);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_ip4));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
g_assert_cmpint (nm_setting_ip_config_get_num_addresses (s_ip4), ==, 2);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 0);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "2.3.4.5");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label != NULL);
g_assert_cmpstr (g_variant_get_string (label, NULL), ==, "eth0:1");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
addr = nm_setting_ip_config_get_address (s_ip4, 1);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "3.4.5.6");
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
label = nm_ip_address_get_attribute (addr, NM_IP_ADDRESS_ATTRIBUTE_LABEL);
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_assert (label == NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: add NMIPAddress/NMIPRoute attributes, use for labels Add key-value attributes to NMIPAddress and NMIPRoute, and use them to store IPv4 address labels. Demote NMSettingIP4Config:address-labels to a D-Bus-only property, and arrange for :addresses setter to read the labels out of that property when creating the addresses.
2014-09-16 16:38:04 -04:00
g_object_unref (conn);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
static void
test_setting_ip4_config_address_data (void)
{
NMSettingIPConfig *s_ip4;
NMIPAddress *addr;
GPtrArray *addrs;
NMConnection *conn;
GVariant *dict, *setting_dict, *value;
GError *error = NULL;
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
g_object_set (G_OBJECT (s_ip4),
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_MANUAL,
NULL);
/* addr 1 */
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
addr = nm_ip_address_new (AF_INET, "1.2.3.4", 24, &error);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
g_assert_no_error (error);
nm_ip_address_set_attribute (addr, "one", g_variant_new_string ("foo"));
nm_ip_address_set_attribute (addr, "two", g_variant_new_int32 (42));
nm_setting_ip_config_add_address (s_ip4, addr);
nm_ip_address_unref (addr);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_ip4));
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
/* addr 2 */
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
addr = nm_ip_address_new (AF_INET, "2.3.4.5", 24, &error);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
g_assert_no_error (error);
nm_setting_ip_config_add_address (s_ip4, addr);
nm_ip_address_unref (addr);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_ip4));
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
/* The client-side D-Bus serialization should include the attributes in
* "address-data", and should not have an "addresses" property.
*/
conn = nmtst_create_minimal_connection ("address-data test", NULL, NM_SETTING_WIRED_SETTING_NAME, NULL);
nm_connection_add_setting (conn, NM_SETTING (s_ip4));
dict = nm_connection_to_dbus (conn, NM_CONNECTION_SERIALIZE_ALL);
setting_dict = g_variant_lookup_value (dict, NM_SETTING_IP4_CONFIG_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (setting_dict != NULL);
value = g_variant_lookup_value (setting_dict, "addresses", NULL);
g_assert (value == NULL);
value = g_variant_lookup_value (setting_dict, "address-data", G_VARIANT_TYPE ("aa{sv}"));
addrs = nm_utils_ip_addresses_from_variant (value, AF_INET);
g_variant_unref (value);
g_assert (addrs != NULL);
g_assert_cmpint (addrs->len, ==, 2);
addr = addrs->pdata[0];
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "1.2.3.4");
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
value = nm_ip_address_get_attribute (addr, "one");
g_assert (value != NULL);
g_assert_cmpstr (g_variant_get_string (value, NULL), ==, "foo");
value = nm_ip_address_get_attribute (addr, "two");
g_assert (value != NULL);
g_assert_cmpint (g_variant_get_int32 (value), ==, 42);
g_ptr_array_unref (addrs);
g_variant_unref (setting_dict);
g_variant_unref (dict);
/* The daemon-side serialization should include both 'addresses' and 'address-data' */
_nm_utils_is_manager_process = TRUE;
dict = nm_connection_to_dbus (conn, NM_CONNECTION_SERIALIZE_ALL);
_nm_utils_is_manager_process = FALSE;
setting_dict = g_variant_lookup_value (dict, NM_SETTING_IP4_CONFIG_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (setting_dict != NULL);
value = g_variant_lookup_value (setting_dict, "addresses", G_VARIANT_TYPE ("aau"));
g_assert (value != NULL);
g_variant_unref (value);
value = g_variant_lookup_value (setting_dict, "address-data", G_VARIANT_TYPE ("aa{sv}"));
g_assert (value != NULL);
g_variant_unref (value);
g_variant_unref (setting_dict);
g_object_unref (conn);
/* When we reserialize that dictionary as a client, 'address-data' will be preferred. */
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
g_assert_no_error (error);
s_ip4 = nm_connection_get_setting_ip4_config (conn);
addr = nm_setting_ip_config_get_address (s_ip4, 0);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "1.2.3.4");
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
value = nm_ip_address_get_attribute (addr, "one");
g_assert (value != NULL);
g_assert_cmpstr (g_variant_get_string (value, NULL), ==, "foo");
value = nm_ip_address_get_attribute (addr, "two");
g_assert (value != NULL);
g_assert_cmpint (g_variant_get_int32 (value), ==, 42);
/* But on the server side, 'addresses' will have precedence. */
_nm_utils_is_manager_process = TRUE;
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
_nm_utils_is_manager_process = FALSE;
g_assert_no_error (error);
g_variant_unref (dict);
s_ip4 = nm_connection_get_setting_ip4_config (conn);
addr = nm_setting_ip_config_get_address (s_ip4, 0);
libnm/tests: make tested IPv4 addresses non-palindromic For IPv4 addresses, the binary representation is in network-order, contrary to host-order. It's better to choose addresses for testing that are differently on big and little endian systems.
2014-12-19 13:16:30 +01:00
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "1.2.3.4");
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
value = nm_ip_address_get_attribute (addr, "one");
g_assert (value == NULL);
value = nm_ip_address_get_attribute (addr, "two");
g_assert (value == NULL);
g_object_unref (conn);
}
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
static void
test_setting_ip_route_attributes (void)
{
GVariant *variant;
gboolean res, known;
#define TEST_ATTR(name, type, value, family, exp_res, exp_known) \
variant = g_variant_new_ ## type (value); \
res = nm_ip_route_attribute_validate (name, variant, family, &known, NULL); \
g_assert (res == exp_res); \
g_assert (known == exp_known); \
g_variant_unref (variant);
TEST_ATTR ("foo", uint32, 12, AF_INET, FALSE, FALSE);
TEST_ATTR ("tos", byte, 127, AF_INET, TRUE, TRUE);
TEST_ATTR ("tos", string, "0x28", AF_INET, FALSE, TRUE);
TEST_ATTR ("cwnd", uint32, 10, AF_INET, TRUE, TRUE);
TEST_ATTR ("cwnd", string, "11", AF_INET, FALSE, TRUE);
TEST_ATTR ("lock-mtu", boolean, TRUE, AF_INET, TRUE, TRUE);
TEST_ATTR ("lock-mtu", uint32, 1, AF_INET, FALSE, TRUE);
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
TEST_ATTR ("from", string, "fd01::1", AF_INET6, TRUE, TRUE);
TEST_ATTR ("from", string, "fd01::1/64", AF_INET6, TRUE, TRUE);
TEST_ATTR ("from", string, "fd01::1/128", AF_INET6, TRUE, TRUE);
TEST_ATTR ("from", string, "fd01::1/129", AF_INET6, FALSE, TRUE);
TEST_ATTR ("from", string, "fd01::1/a", AF_INET6, FALSE, TRUE);
TEST_ATTR ("from", string, "abc/64", AF_INET6, FALSE, TRUE);
TEST_ATTR ("from", string, "1.2.3.4", AF_INET, FALSE, TRUE);
TEST_ATTR ("from", string, "1.2.3.4", AF_INET6, FALSE, TRUE);
TEST_ATTR ("src", string, "1.2.3.4", AF_INET, TRUE, TRUE);
TEST_ATTR ("src", string, "1.2.3.4", AF_INET6, FALSE, TRUE);
TEST_ATTR ("src", string, "1.2.3.0/24", AF_INET, FALSE, TRUE);
TEST_ATTR ("src", string, "fd01::12", AF_INET6, TRUE, TRUE);
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
platform: add support for local routes Also update unit tests. https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/407 https://bugzilla.redhat.com/show_bug.cgi?id=1821787
2020-05-29 17:53:02 +02:00
TEST_ATTR ("type", string, "local", AF_INET, TRUE, TRUE);
TEST_ATTR ("type", string, "local", AF_INET6, TRUE, TRUE);
TEST_ATTR ("type", string, "unicast", AF_INET, TRUE, TRUE);
TEST_ATTR ("type", string, "unicast", AF_INET6, TRUE, TRUE);
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
#undef TEST_ATTR
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
test_setting_gsm_apn_spaces (void)
{
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSettingGsm *s_gsm = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
const char *tmp;
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (s_gsm);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Trailing space */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar ", NULL);
tmp = nm_setting_gsm_get_apn (s_gsm);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpstr (tmp, ==, "foobar");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Leading space */
g_object_set (s_gsm, NM_SETTING_GSM_APN, " foobar", NULL);
tmp = nm_setting_gsm_get_apn (s_gsm);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpstr (tmp, ==, "foobar");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
test_setting_gsm_apn_bad_chars (void)
{
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSettingGsm *s_gsm = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (s_gsm);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Make sure a valid APN works */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar123.-baz", NULL);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Random invalid chars */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "@#%$@#%@#%", NULL);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Spaces */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar baz", NULL);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* 0 characters long */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "", NULL);
setting-gsm: allow empty apn property in verify() NetworkManager treats "gsm.apn" %NULL as setting an empty APN (""). At least with ModemManager. With oFono, a %NULL APN means not to set the "AccessPointName", so oFono implementation treats %NULL different from "". Soon the meaning will change to allow %NULL to automatically obtain the APN from the mobile-broadband-provider-info. That will be a change in behavior how to treat %NULL. Anyway, since %NULL is accepted and in fact means to actually use "", the empty word should be also accepted to explicitly choose this behavior. This is especially important in combination with changing the meaning of %NULL.
2019-08-25 15:27:37 +02:00
g_assert (nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* 65-character long */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijkl1", NULL);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
test_setting_gsm_apn_underscore (void)
{
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSettingGsm *s_gsm = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
g_assert (s_gsm);
/* 65-character long */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar_baz", NULL);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_gsm));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
test_setting_gsm_without_number (void)
{
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSettingGsm *s_gsm = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
g_assert (s_gsm);
g_object_set (s_gsm, NM_SETTING_GSM_NUMBER, NULL, NULL);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (s_gsm));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (s_gsm, NM_SETTING_GSM_NUMBER, "", NULL);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verify_fails (NM_SETTING (s_gsm), NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
libnm/wwan: add GSM setting device-id, sim-id, and sim-operator-id properties These properties limit whether the connection applies to a certain WWAN modem based on the modem's device ID or SIM ID (as reported by the WWAN management service), or through the MCC/MNC ID of the operator that issued the SIM card.
2015-10-01 12:44:28 -05:00
static void
test_setting_gsm_sim_operator_id (void)
{
gs_unref_object NMSettingGsm *s_gsm = NULL;
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
g_assert (s_gsm);
/* Valid */
g_object_set (s_gsm, NM_SETTING_GSM_SIM_OPERATOR_ID, "12345", NULL);
nmtst_assert_setting_verifies (NM_SETTING (s_gsm));
g_object_set (s_gsm, NM_SETTING_GSM_SIM_OPERATOR_ID, "123456", NULL);
nmtst_assert_setting_verifies (NM_SETTING (s_gsm));
/* Invalid */
g_object_set (s_gsm, NM_SETTING_GSM_SIM_OPERATOR_ID, "", NULL);
nmtst_assert_setting_verify_fails (NM_SETTING (s_gsm), NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_object_set (s_gsm, NM_SETTING_GSM_SIM_OPERATOR_ID, " ", NULL);
nmtst_assert_setting_verify_fails (NM_SETTING (s_gsm), NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_object_set (s_gsm, NM_SETTING_GSM_SIM_OPERATOR_ID, "abcdef", NULL);
nmtst_assert_setting_verify_fails (NM_SETTING (s_gsm), NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static NMSettingWirelessSecurity *
make_test_wsec_setting (const char *detail)
{
NMSettingWirelessSecurity *s_wsec;
s_wsec = (NMSettingWirelessSecurity *) nm_setting_wireless_security_new ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (s_wsec);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (s_wsec,
NM_SETTING_WIRELESS_SECURITY_KEY_MGMT, "wpa-psk",
NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME, "foobarbaz",
NM_SETTING_WIRELESS_SECURITY_PSK, "random psk",
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
NM_SETTING_WIRELESS_SECURITY_PSK_FLAGS, NM_SETTING_SECRET_FLAG_NOT_SAVED,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NM_SETTING_WIRELESS_SECURITY_WEP_KEY0, "aaaaaaaaaa",
NULL);
return s_wsec;
}
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
static gboolean
_variant_contains (GVariant *vardict, const char *key)
{
gs_unref_variant GVariant *value = NULL;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
value = g_variant_lookup_value (vardict, key, NULL);
return !!value;
}
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
test_setting_to_dbus_all (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMSettingWirelessSecurity *s_wsec;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
s_wsec = make_test_wsec_setting ("setting-to-dbus-all");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (NM_SETTING (s_wsec), NULL, NM_CONNECTION_SERIALIZE_ALL, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Make sure all keys are there */
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_KEY_MGMT));
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME));
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_PSK));
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (dict);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (s_wsec);
}
static void
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
test_setting_to_dbus_no_secrets (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMSettingWirelessSecurity *s_wsec;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
s_wsec = make_test_wsec_setting ("setting-to-dbus-no-secrets");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (NM_SETTING (s_wsec), NULL, NM_CONNECTION_SERIALIZE_NO_SECRETS, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Make sure non-secret keys are there */
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_KEY_MGMT));
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Make sure secrets are not there */
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_PSK));
g_assert (!_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (dict);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (s_wsec);
}
static void
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
test_setting_to_dbus_only_secrets (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMSettingWirelessSecurity *s_wsec;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
s_wsec = make_test_wsec_setting ("setting-to-dbus-only-secrets");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (NM_SETTING (s_wsec), NULL, NM_CONNECTION_SERIALIZE_ONLY_SECRETS, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
/* Make sure non-secret keys are not there */
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_KEY_MGMT));
g_assert (!_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
/* Make sure secrets are there */
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_PSK));
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (dict);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (s_wsec);
}
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
static void
test_setting_to_dbus_transform (void)
{
NMSetting *s_wired;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict, *val;
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
const char *test_mac_address = "11:22:33:44:55:66";
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
const guint8 *dbus_mac_address;
guint8 cmp_mac_address[ETH_ALEN];
gsize len;
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
s_wired = nm_setting_wired_new ();
g_object_set (s_wired,
NM_SETTING_WIRED_MAC_ADDRESS, test_mac_address,
NULL);
g_assert_cmpstr (nm_setting_wired_get_mac_address (NM_SETTING_WIRED (s_wired)), ==, test_mac_address);
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (s_wired, NULL, NM_CONNECTION_SERIALIZE_ALL, NULL);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_assert (dict != NULL);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
val = g_variant_lookup_value (dict, NM_SETTING_WIRED_MAC_ADDRESS, G_VARIANT_TYPE_BYTESTRING);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
g_assert (val != NULL);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
dbus_mac_address = g_variant_get_fixed_array (val, &len, 1);
g_assert_cmpint (len, ==, ETH_ALEN);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
nm_utils_hwaddr_aton (test_mac_address, cmp_mac_address, ETH_ALEN);
g_assert (memcmp (dbus_mac_address, cmp_mac_address, ETH_ALEN) == 0);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (val);
g_variant_unref (dict);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
g_object_unref (s_wired);
}
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
static void
test_setting_to_dbus_enum (void)
{
libnm-core: make NMSettingSerial:parity an enum NMSettingSerial:parity was defined as a char-typed property that could have the (case-sensitive!) values 'n', 'E', or 'o'. This is zany. Add an NMSettingSerialParity enum, and use that instead.
2014-09-24 09:12:46 -04:00
NMSetting *s_ip6, *s_wsec, *s_serial;
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
GVariant *dict, *val;
/* enum */
s_ip6 = nm_setting_ip6_config_new ();
g_object_set (s_ip6,
NM_SETTING_IP6_CONFIG_IP6_PRIVACY, NM_SETTING_IP6_CONFIG_PRIVACY_PREFER_TEMP_ADDR,
NULL);
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (s_ip6, NULL, NM_CONNECTION_SERIALIZE_ALL, NULL);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
g_assert (dict != NULL);
val = g_variant_lookup_value (dict, NM_SETTING_IP6_CONFIG_IP6_PRIVACY, G_VARIANT_TYPE_INT32);
g_assert (val != NULL);
g_assert_cmpint (g_variant_get_int32 (val), ==, NM_SETTING_IP6_CONFIG_PRIVACY_PREFER_TEMP_ADDR);
g_variant_unref (val);
g_variant_unref (dict);
g_object_unref (s_ip6);
/* flags (and a transformed enum) */
s_wsec = nm_setting_wireless_security_new ();
g_object_set (s_wsec,
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_TYPE, NM_WEP_KEY_TYPE_KEY,
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_FLAGS, (NM_SETTING_SECRET_FLAG_AGENT_OWNED |
NM_SETTING_SECRET_FLAG_NOT_SAVED),
NULL);
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (s_wsec, NULL, NM_CONNECTION_SERIALIZE_ALL, NULL);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
g_assert (dict != NULL);
val = g_variant_lookup_value (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY_TYPE, G_VARIANT_TYPE_UINT32);
g_assert (val != NULL);
g_assert_cmpint (g_variant_get_uint32 (val), ==, NM_WEP_KEY_TYPE_KEY);
g_variant_unref (val);
val = g_variant_lookup_value (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY_FLAGS, G_VARIANT_TYPE_UINT32);
g_assert (val != NULL);
g_assert_cmpint (g_variant_get_uint32 (val), ==, (NM_SETTING_SECRET_FLAG_AGENT_OWNED |
NM_SETTING_SECRET_FLAG_NOT_SAVED));
g_variant_unref (val);
g_variant_unref (dict);
g_object_unref (s_wsec);
libnm-core: make NMSettingSerial:parity an enum NMSettingSerial:parity was defined as a char-typed property that could have the (case-sensitive!) values 'n', 'E', or 'o'. This is zany. Add an NMSettingSerialParity enum, and use that instead.
2014-09-24 09:12:46 -04:00
/* another transformed enum */
s_serial = nm_setting_serial_new ();
g_object_set (s_serial,
NM_SETTING_SERIAL_PARITY, NM_SETTING_SERIAL_PARITY_ODD,
NULL);
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (s_serial, NULL, NM_CONNECTION_SERIALIZE_ALL, NULL);
libnm-core: make NMSettingSerial:parity an enum NMSettingSerial:parity was defined as a char-typed property that could have the (case-sensitive!) values 'n', 'E', or 'o'. This is zany. Add an NMSettingSerialParity enum, and use that instead.
2014-09-24 09:12:46 -04:00
g_assert (dict != NULL);
val = g_variant_lookup_value (dict, NM_SETTING_SERIAL_PARITY, G_VARIANT_TYPE_BYTE);
g_assert (val != NULL);
g_assert_cmpint (g_variant_get_byte (val), ==, 'o');
g_variant_unref (val);
g_variant_unref (dict);
g_object_unref (s_serial);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
test_connection_to_dbus_setting_name (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMConnection *connection;
NMSettingWirelessSecurity *s_wsec;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
s_wsec = make_test_wsec_setting ("connection-to-dbus-setting-name");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, NM_SETTING (s_wsec));
libnm,core: add _nm_connection_aggregate() to replace nm_connection_for_each_setting_value() We should no longer use nm_connection_for_each_setting_value() and nm_setting_for_each_value(). It's fundamentally broken as it does not work with properties that are not backed by a GObject property and it cannot be fixed because it is public API. Add an internal function _nm_connection_aggregate() to replace it. Compare the implementation of the aggregation functionality inside libnm with the previous two checks for secret-flags that it replaces: - previous approach broke abstraction and require detailed knowledge of secret flags. Meaning, they must special case NMSettingVpn and GObject-property based secrets. If we implement a new way for implementing secrets (like we will need for WireGuard), then this the new way should only affect libnm-core, not require changes elsewhere. - it's very inefficient to itereate over all settings. It involves cloning and sorting the list of settings, and retrieve and clone all GObject properties. Only to look at secret properties alone. _nm_connection_aggregate() is supposed to be more flexible then just the two new aggregate types that perform a "find-any" search. The @arg argument and boolean return value can suffice to implement different aggregation types in the future. Also fixes the check of NMAgentManager for secret flags for VPNs (NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS). A secret for VPNs is a property that either has a secret or a secret-flag. The previous implementation would only look at present secrets and check their flags. It wouldn't check secret-flags that are NM_SETTING_SECRET_FLAG_NONE, but have no secret.
2019-01-04 11:28:27 +01:00
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SECRETS, NULL));
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
g_object_set (s_wsec,
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_FLAGS, NM_SETTING_SECRET_FLAG_NOT_SAVED,
NM_SETTING_WIRELESS_SECURITY_LEAP_PASSWORD_FLAGS, NM_SETTING_SECRET_FLAG_NOT_SAVED,
NULL);
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SECRETS, NULL));
g_assert (!_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
g_object_set (s_wsec,
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_FLAGS, NM_SETTING_SECRET_FLAG_NONE,
NM_SETTING_WIRELESS_SECURITY_LEAP_PASSWORD_FLAGS, NM_SETTING_SECRET_FLAG_NONE,
NULL);
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SECRETS, NULL));
g_assert (_nm_connection_aggregate (connection, NM_CONNECTION_AGGREGATE_ANY_SYSTEM_SECRET_FLAGS, NULL));
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
dict = nm_connection_to_dbus (connection, NM_CONNECTION_SERIALIZE_ALL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
/* Make sure the keys of the first level dict are setting names, not
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
* the GType name of the setting objects.
*/
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (_variant_contains (dict, NM_SETTING_WIRELESS_SECURITY_SETTING_NAME));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (dict);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (connection);
}
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
static void
test_connection_to_dbus_deprecated_props (void)
{
NMConnection *connection;
NMSetting *s_wireless;
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
GBytes *ssid;
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
NMSettingWirelessSecurity *s_wsec;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict, *wireless_dict, *sec_val;
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
connection = nmtst_create_minimal_connection ("test-connection-to-dbus-deprecated-props",
NULL,
NM_SETTING_WIRELESS_SETTING_NAME,
NULL);
s_wireless = nm_setting_wireless_new ();
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
ssid = g_bytes_new ("1234567", 7);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
g_object_set (s_wireless,
NM_SETTING_WIRELESS_SSID, ssid,
NULL);
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
g_bytes_unref (ssid);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
nm_connection_add_setting (connection, s_wireless);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
/* Serialization should not have an 802-11-wireless.security property */
dict = nm_connection_to_dbus (connection, NM_CONNECTION_SERIALIZE_ALL);
g_assert (dict != NULL);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
wireless_dict = g_variant_lookup_value (dict, NM_SETTING_WIRELESS_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (wireless_dict != NULL);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
sec_val = g_variant_lookup_value (wireless_dict, "security", NULL);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
g_assert (sec_val == NULL);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (wireless_dict);
g_variant_unref (dict);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
/* Now add an NMSettingWirelessSecurity and try again */
s_wsec = make_test_wsec_setting ("test-connection-to-dbus-deprecated-props");
nm_connection_add_setting (connection, NM_SETTING (s_wsec));
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
dict = nm_connection_to_dbus (connection, NM_CONNECTION_SERIALIZE_ALL);
g_assert (dict != NULL);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
wireless_dict = g_variant_lookup_value (dict, NM_SETTING_WIRELESS_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (wireless_dict != NULL);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
sec_val = g_variant_lookup_value (wireless_dict, "security", NULL);
g_assert (g_variant_is_of_type (sec_val, G_VARIANT_TYPE_STRING));
g_assert_cmpstr (g_variant_get_string (sec_val, NULL), ==, NM_SETTING_WIRELESS_SECURITY_SETTING_NAME);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (sec_val);
g_variant_unref (wireless_dict);
g_variant_unref (dict);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
g_object_unref (connection);
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
test_setting_new_from_dbus (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMSettingWirelessSecurity *s_wsec;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
s_wsec = make_test_wsec_setting ("setting-new-from-dbus");
libnm: add nm_connection_to_dbus_full() with options argument No options are implemented yet.
2019-06-27 09:07:16 +02:00
dict = _nm_setting_to_dbus (NM_SETTING (s_wsec), NULL, NM_CONNECTION_SERIALIZE_ALL, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (s_wsec);
libnm-core: allow strict and relaxed error behavior for _nm_setting_new_from_dbus() In some situations, we want strict checking of errors, for example when NetworkManager receives a new connection from a client, the connection must make sense as a whole (and since NetworkManager service is backward compatible to the clients and not the other way around, there is no excuse for sending invalid data to the server). In other situations, we want a best-effort behavior. Like when NetworkManager sends a connection to its clients, those clients want to extract as many properties as they understand, but in order to be forward compatible against newer server versions, invalid or unknown properties must be accepted. Previously, a mixture of both was done. Some issues caused a failure to create a new NMSetting, other invalid parts were just silently ignored or triggered a g_warning() in glib. Now allow for both. When doing strict-validation, be more strict and reject all unknown properties and catch when the user sets an invalid argument. On the other hand, allow for a best-effort mode that effectively cannot fail and will return a new NMSetting instance. For now, add NMSettingParseFlags so that the caller can choose the old behavior, strict parsing, or best effort. This patch doesn't have any externally visible change except that no more g_warnings will be emitted.
2016-03-18 13:42:50 +01:00
s_wsec = (NMSettingWirelessSecurity *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_WIRELESS_SECURITY, dict, NULL, NM_SETTING_PARSE_FLAGS_NONE, NULL);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (dict);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (s_wsec);
g_assert_cmpstr (nm_setting_wireless_security_get_key_mgmt (s_wsec), ==, "wpa-psk");
g_assert_cmpstr (nm_setting_wireless_security_get_leap_username (s_wsec), ==, "foobarbaz");
g_assert_cmpstr (nm_setting_wireless_security_get_psk (s_wsec), ==, "random psk");
g_object_unref (s_wsec);
}
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
static void
test_setting_new_from_dbus_transform (void)
{
NMSetting *s_wired;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *dict;
GVariantBuilder builder;
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
const char *test_mac_address = "11:22:33:44:55:66";
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
guint8 dbus_mac_address[ETH_ALEN];
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
GError *error = NULL;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
nm_utils_hwaddr_aton (test_mac_address, dbus_mac_address, ETH_ALEN);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_init (&builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&builder, "{sv}",
NM_SETTING_WIRED_MAC_ADDRESS,
g_variant_new_fixed_array (G_VARIANT_TYPE_BYTE,
dbus_mac_address, ETH_ALEN, 1));
dict = g_variant_builder_end (&builder);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
libnm-core: allow strict and relaxed error behavior for _nm_setting_new_from_dbus() In some situations, we want strict checking of errors, for example when NetworkManager receives a new connection from a client, the connection must make sense as a whole (and since NetworkManager service is backward compatible to the clients and not the other way around, there is no excuse for sending invalid data to the server). In other situations, we want a best-effort behavior. Like when NetworkManager sends a connection to its clients, those clients want to extract as many properties as they understand, but in order to be forward compatible against newer server versions, invalid or unknown properties must be accepted. Previously, a mixture of both was done. Some issues caused a failure to create a new NMSetting, other invalid parts were just silently ignored or triggered a g_warning() in glib. Now allow for both. When doing strict-validation, be more strict and reject all unknown properties and catch when the user sets an invalid argument. On the other hand, allow for a best-effort mode that effectively cannot fail and will return a new NMSetting instance. For now, add NMSettingParseFlags so that the caller can choose the old behavior, strict parsing, or best effort. This patch doesn't have any externally visible change except that no more g_warnings will be emitted.
2016-03-18 13:42:50 +01:00
s_wired = _nm_setting_new_from_dbus (NM_TYPE_SETTING_WIRED, dict, NULL, NM_SETTING_PARSE_FLAGS_NONE, &error);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
g_assert_no_error (error);
g_assert_cmpstr (nm_setting_wired_get_mac_address (NM_SETTING_WIRED (s_wired)), ==, test_mac_address);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (dict);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
g_object_unref (s_wired);
}
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
static void
test_setting_new_from_dbus_enum (void)
{
NMSettingIP6Config *s_ip6;
NMSettingWirelessSecurity *s_wsec;
libnm-core: make NMSettingSerial:parity an enum NMSettingSerial:parity was defined as a char-typed property that could have the (case-sensitive!) values 'n', 'E', or 'o'. This is zany. Add an NMSettingSerialParity enum, and use that instead.
2014-09-24 09:12:46 -04:00
NMSettingSerial *s_serial;
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
GVariant *dict;
GVariantBuilder builder;
GError *error = NULL;
/* enum */
g_variant_builder_init (&builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&builder, "{sv}",
NM_SETTING_IP6_CONFIG_IP6_PRIVACY,
g_variant_new_int32 (NM_SETTING_IP6_CONFIG_PRIVACY_PREFER_TEMP_ADDR));
dict = g_variant_builder_end (&builder);
libnm-core: allow strict and relaxed error behavior for _nm_setting_new_from_dbus() In some situations, we want strict checking of errors, for example when NetworkManager receives a new connection from a client, the connection must make sense as a whole (and since NetworkManager service is backward compatible to the clients and not the other way around, there is no excuse for sending invalid data to the server). In other situations, we want a best-effort behavior. Like when NetworkManager sends a connection to its clients, those clients want to extract as many properties as they understand, but in order to be forward compatible against newer server versions, invalid or unknown properties must be accepted. Previously, a mixture of both was done. Some issues caused a failure to create a new NMSetting, other invalid parts were just silently ignored or triggered a g_warning() in glib. Now allow for both. When doing strict-validation, be more strict and reject all unknown properties and catch when the user sets an invalid argument. On the other hand, allow for a best-effort mode that effectively cannot fail and will return a new NMSetting instance. For now, add NMSettingParseFlags so that the caller can choose the old behavior, strict parsing, or best effort. This patch doesn't have any externally visible change except that no more g_warnings will be emitted.
2016-03-18 13:42:50 +01:00
s_ip6 = (NMSettingIP6Config *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_IP6_CONFIG, dict, NULL, NM_SETTING_PARSE_FLAGS_NONE, &error);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
g_assert_no_error (error);
g_assert_cmpint (nm_setting_ip6_config_get_ip6_privacy (s_ip6), ==, NM_SETTING_IP6_CONFIG_PRIVACY_PREFER_TEMP_ADDR);
g_variant_unref (dict);
g_object_unref (s_ip6);
/* flags (and a transformed enum) */
g_variant_builder_init (&builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&builder, "{sv}",
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_TYPE,
g_variant_new_uint32 (NM_WEP_KEY_TYPE_KEY));
g_variant_builder_add (&builder, "{sv}",
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_FLAGS,
g_variant_new_uint32 (NM_SETTING_SECRET_FLAG_AGENT_OWNED |
NM_SETTING_SECRET_FLAG_NOT_SAVED));
dict = g_variant_builder_end (&builder);
libnm-core: allow strict and relaxed error behavior for _nm_setting_new_from_dbus() In some situations, we want strict checking of errors, for example when NetworkManager receives a new connection from a client, the connection must make sense as a whole (and since NetworkManager service is backward compatible to the clients and not the other way around, there is no excuse for sending invalid data to the server). In other situations, we want a best-effort behavior. Like when NetworkManager sends a connection to its clients, those clients want to extract as many properties as they understand, but in order to be forward compatible against newer server versions, invalid or unknown properties must be accepted. Previously, a mixture of both was done. Some issues caused a failure to create a new NMSetting, other invalid parts were just silently ignored or triggered a g_warning() in glib. Now allow for both. When doing strict-validation, be more strict and reject all unknown properties and catch when the user sets an invalid argument. On the other hand, allow for a best-effort mode that effectively cannot fail and will return a new NMSetting instance. For now, add NMSettingParseFlags so that the caller can choose the old behavior, strict parsing, or best effort. This patch doesn't have any externally visible change except that no more g_warnings will be emitted.
2016-03-18 13:42:50 +01:00
s_wsec = (NMSettingWirelessSecurity *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_WIRELESS_SECURITY, dict, NULL, NM_SETTING_PARSE_FLAGS_NONE, &error);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
g_assert_no_error (error);
g_assert_cmpint (nm_setting_wireless_security_get_wep_key_type (s_wsec), ==, NM_WEP_KEY_TYPE_KEY);
g_assert_cmpint (nm_setting_wireless_security_get_wep_key_flags (s_wsec), ==, (NM_SETTING_SECRET_FLAG_AGENT_OWNED |
NM_SETTING_SECRET_FLAG_NOT_SAVED));
g_variant_unref (dict);
g_object_unref (s_wsec);
libnm-core: make NMSettingSerial:parity an enum NMSettingSerial:parity was defined as a char-typed property that could have the (case-sensitive!) values 'n', 'E', or 'o'. This is zany. Add an NMSettingSerialParity enum, and use that instead.
2014-09-24 09:12:46 -04:00
/* another transformed enum */
g_variant_builder_init (&builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&builder, "{sv}",
NM_SETTING_SERIAL_PARITY,
g_variant_new_byte ('E'));
dict = g_variant_builder_end (&builder);
libnm-core: allow strict and relaxed error behavior for _nm_setting_new_from_dbus() In some situations, we want strict checking of errors, for example when NetworkManager receives a new connection from a client, the connection must make sense as a whole (and since NetworkManager service is backward compatible to the clients and not the other way around, there is no excuse for sending invalid data to the server). In other situations, we want a best-effort behavior. Like when NetworkManager sends a connection to its clients, those clients want to extract as many properties as they understand, but in order to be forward compatible against newer server versions, invalid or unknown properties must be accepted. Previously, a mixture of both was done. Some issues caused a failure to create a new NMSetting, other invalid parts were just silently ignored or triggered a g_warning() in glib. Now allow for both. When doing strict-validation, be more strict and reject all unknown properties and catch when the user sets an invalid argument. On the other hand, allow for a best-effort mode that effectively cannot fail and will return a new NMSetting instance. For now, add NMSettingParseFlags so that the caller can choose the old behavior, strict parsing, or best effort. This patch doesn't have any externally visible change except that no more g_warnings will be emitted.
2016-03-18 13:42:50 +01:00
s_serial = (NMSettingSerial *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_SERIAL, dict, NULL, NM_SETTING_PARSE_FLAGS_NONE, &error);
libnm-core: make NMSettingSerial:parity an enum NMSettingSerial:parity was defined as a char-typed property that could have the (case-sensitive!) values 'n', 'E', or 'o'. This is zany. Add an NMSettingSerialParity enum, and use that instead.
2014-09-24 09:12:46 -04:00
g_assert_no_error (error);
g_assert_cmpint (nm_setting_serial_get_parity (s_serial), ==, NM_SETTING_SERIAL_PARITY_EVEN);
g_variant_unref (dict);
g_object_unref (s_serial);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
}
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
static void
test_setting_new_from_dbus_bad (void)
{
NMSetting *setting;
NMConnection *conn;
GBytes *ssid;
GPtrArray *addrs;
GVariant *orig_dict, *dict;
GError *error = NULL;
/* We want to test:
* - ordinary scalar properties
* - string properties
* - GBytes-valued properties (which are handled specially by set_property_from_dbus())
* - enum/flags-valued properties
* - overridden properties
* - transformed properties
*
* No single setting class has examples of all of these, so we need two settings.
*/
conn = nm_simple_connection_new ();
setting = nm_setting_connection_new ();
g_object_set (setting,
NM_SETTING_CONNECTION_ID, "test",
NM_SETTING_CONNECTION_UUID, "83c5a841-1759-4cdb-bfce-8d4087956497",
NULL);
nm_connection_add_setting (conn, setting);
setting = nm_setting_wireless_new ();
ssid = g_bytes_new ("my-ssid", 7);
g_object_set (setting,
/* scalar */
NM_SETTING_WIRELESS_RATE, 100,
/* string */
NM_SETTING_WIRELESS_MODE, NM_SETTING_WIRELESS_MODE_INFRA,
/* GBytes */
NM_SETTING_WIRELESS_SSID, ssid,
/* transformed */
NM_SETTING_WIRELESS_BSSID, "00:11:22:33:44:55",
NULL);
g_bytes_unref (ssid);
nm_connection_add_setting (conn, setting);
setting = nm_setting_ip6_config_new ();
addrs = g_ptr_array_new_with_free_func ((GDestroyNotify) nm_ip_address_unref);
g_ptr_array_add (addrs, nm_ip_address_new (AF_INET6, "1234::5678", 64, NULL));
g_object_set (setting,
/* enum */
NM_SETTING_IP6_CONFIG_IP6_PRIVACY, NM_SETTING_IP6_CONFIG_PRIVACY_PREFER_PUBLIC_ADDR,
/* overridden */
NM_SETTING_IP_CONFIG_ADDRESSES, addrs,
/* (needed in order to verify()) */
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_AUTO,
NULL);
g_ptr_array_unref (addrs);
nm_connection_add_setting (conn, setting);
orig_dict = nm_connection_to_dbus (conn, NM_CONNECTION_SERIALIZE_ALL);
g_object_unref (conn);
/* sanity-check */
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (orig_dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert_no_error (error);
g_assert (conn);
g_object_unref (conn);
/* Compatible mismatches */
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_WIRELESS_SETTING_NAME,
NM_SETTING_WIRELESS_RATE,
"i", 10);
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert (conn);
g_assert_no_error (error);
setting = nm_connection_get_setting (conn, NM_TYPE_SETTING_WIRELESS);
g_assert (setting);
g_assert_cmpint (nm_setting_wireless_get_rate (NM_SETTING_WIRELESS (setting)), ==, 10);
g_object_unref (conn);
g_variant_unref (dict);
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_IP6_PRIVACY,
"i", NM_SETTING_IP6_CONFIG_PRIVACY_PREFER_TEMP_ADDR);
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert (conn);
g_assert_no_error (error);
setting = nm_connection_get_setting (conn, NM_TYPE_SETTING_IP6_CONFIG);
g_assert (setting);
g_assert_cmpint (nm_setting_ip6_config_get_ip6_privacy (NM_SETTING_IP6_CONFIG (setting)), ==, NM_SETTING_IP6_CONFIG_PRIVACY_PREFER_TEMP_ADDR);
g_object_unref (conn);
g_variant_unref (dict);
/* Incompatible mismatches */
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_WIRELESS_SETTING_NAME,
NM_SETTING_WIRELESS_RATE,
"s", "ten");
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_assert (g_str_has_prefix (error->message, "802-11-wireless.rate:"));
g_clear_error (&error);
g_variant_unref (dict);
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_WIRELESS_SETTING_NAME,
NM_SETTING_WIRELESS_MODE,
"b", FALSE);
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_assert (g_str_has_prefix (error->message, "802-11-wireless.mode:"));
g_clear_error (&error);
g_variant_unref (dict);
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_WIRELESS_SETTING_NAME,
NM_SETTING_WIRELESS_SSID,
"s", "fred");
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_assert (g_str_has_prefix (error->message, "802-11-wireless.ssid:"));
g_clear_error (&error);
g_variant_unref (dict);
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_WIRELESS_SETTING_NAME,
NM_SETTING_WIRELESS_BSSID,
"i", 42);
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_assert (g_str_has_prefix (error->message, "802-11-wireless.bssid:"));
g_clear_error (&error);
g_variant_unref (dict);
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_IP6_PRIVACY,
"s", "private");
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_assert (g_str_has_prefix (error->message, "ipv6.ip6-privacy:"));
g_clear_error (&error);
g_variant_unref (dict);
dict = g_variant_ref (orig_dict);
NMTST_VARIANT_EDITOR (dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_IP_CONFIG_ADDRESSES,
"s", "1234::5678");
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (dict, &error);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
g_assert (g_str_has_prefix (error->message, "ipv6.addresses:"));
g_clear_error (&error);
g_variant_unref (dict);
g_variant_unref (orig_dict);
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static NMConnection *
new_test_connection (void)
{
NMConnection *connection;
NMSetting *setting;
char *uuid;
guint64 timestamp = time (NULL);
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
setting = nm_setting_connection_new ();
uuid = nm_utils_uuid_generate ();
g_object_set (G_OBJECT (setting),
NM_SETTING_CONNECTION_ID, "foobar",
NM_SETTING_CONNECTION_UUID, uuid,
NM_SETTING_CONNECTION_TYPE, NM_SETTING_WIRED_SETTING_NAME,
NM_SETTING_CONNECTION_TIMESTAMP, timestamp,
NULL);
g_free (uuid);
nm_connection_add_setting (connection, setting);
setting = nm_setting_wired_new ();
g_object_set (G_OBJECT (setting),
NM_SETTING_WIRED_MTU, 1592,
NULL);
nm_connection_add_setting (connection, setting);
setting = nm_setting_ip4_config_new ();
g_object_set (G_OBJECT (setting),
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_AUTO,
NM_SETTING_IP_CONFIG_DHCP_HOSTNAME, "eyeofthetiger",
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NULL);
nm_connection_add_setting (connection, setting);
return connection;
}
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
static GVariant *
new_connection_dict (char **out_uuid,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
const char **out_expected_id,
const char **out_expected_ip6_method)
{
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariantBuilder conn_builder, setting_builder;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_init (&conn_builder, NM_VARIANT_TYPE_CONNECTION);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
*out_uuid = nm_utils_uuid_generate ();
*out_expected_id = "My happy connection";
*out_expected_ip6_method = NM_SETTING_IP6_CONFIG_METHOD_LINK_LOCAL;
/* Connection setting */
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_init (&setting_builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&setting_builder, "{sv}",
NM_SETTING_NAME,
g_variant_new_string (NM_SETTING_CONNECTION_SETTING_NAME));
g_variant_builder_add (&setting_builder, "{sv}",
NM_SETTING_CONNECTION_ID,
g_variant_new_string (*out_expected_id));
g_variant_builder_add (&setting_builder, "{sv}",
NM_SETTING_CONNECTION_UUID,
g_variant_new_string (*out_uuid));
g_variant_builder_add (&setting_builder, "{sv}",
NM_SETTING_CONNECTION_TYPE,
g_variant_new_string (NM_SETTING_WIRED_SETTING_NAME));
g_variant_builder_add (&conn_builder, "{sa{sv}}",
NM_SETTING_CONNECTION_SETTING_NAME,
&setting_builder);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Wired setting */
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_init (&setting_builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&conn_builder, "{sa{sv}}",
NM_SETTING_WIRED_SETTING_NAME,
&setting_builder);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* IP6 */
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_builder_init (&setting_builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&setting_builder, "{sv}",
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_METHOD,
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_new_string (*out_expected_ip6_method));
g_variant_builder_add (&conn_builder, "{sa{sv}}",
NM_SETTING_IP6_CONFIG_SETTING_NAME,
&setting_builder);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
return g_variant_builder_end (&conn_builder);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
build: fix -Wstrict-prototypes warnings We disabled -Wstrict-prototypes in commit db9b1df0e47996ff8aaea468a11e1e97f64ee126 . Fix compiler warnings. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-10-29 10:44:31 +01:00
test_connection_replace_settings (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMConnection *connection;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *new_settings;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
GError *error = NULL;
gboolean success;
NMSettingConnection *s_con;
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NMSettingIPConfig *s_ip6;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
char *uuid = NULL;
const char *expected_id = NULL, *expected_method = NULL;
connection = new_test_connection ();
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
new_settings = new_connection_dict (&uuid, &expected_id, &expected_method);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (new_settings);
/* Replace settings and test */
success = nm_connection_replace_settings (connection, new_settings, &error);
g_assert_no_error (error);
g_assert (success);
s_con = nm_connection_get_setting_connection (connection);
g_assert (s_con);
g_assert_cmpstr (nm_setting_connection_get_id (s_con), ==, expected_id);
g_assert_cmpstr (nm_setting_connection_get_uuid (s_con), ==, uuid);
g_assert (nm_connection_get_setting_wired (connection));
g_assert (!nm_connection_get_setting_ip4_config (connection));
s_ip6 = nm_connection_get_setting_ip6_config (connection);
g_assert (s_ip6);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
g_assert_cmpstr (nm_setting_ip_config_get_method (s_ip6), ==, expected_method);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_free (uuid);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (new_settings);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (connection);
}
static void
build: fix -Wstrict-prototypes warnings We disabled -Wstrict-prototypes in commit db9b1df0e47996ff8aaea468a11e1e97f64ee126 . Fix compiler warnings. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-10-29 10:44:31 +01:00
test_connection_replace_settings_from_connection (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMConnection *connection, *replacement;
NMSettingConnection *s_con;
NMSetting *setting;
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
GBytes *ssid;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
char *uuid = NULL;
const char *expected_id = "Awesome connection";
connection = new_test_connection ();
g_assert (connection);
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
replacement = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (replacement);
/* New connection setting */
setting = nm_setting_connection_new ();
g_assert (setting);
uuid = nm_utils_uuid_generate ();
g_object_set (setting,
NM_SETTING_CONNECTION_ID, expected_id,
NM_SETTING_CONNECTION_UUID, uuid,
NM_SETTING_CONNECTION_TYPE, NM_SETTING_WIRELESS_SETTING_NAME,
NULL);
nm_connection_add_setting (replacement, setting);
/* New wifi setting */
setting = nm_setting_wireless_new ();
g_assert (setting);
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
ssid = g_bytes_new ("1234567", 7);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (setting,
NM_SETTING_WIRELESS_SSID, ssid,
NM_SETTING_WIRELESS_MODE, "infrastructure",
NULL);
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
g_bytes_unref (ssid);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (replacement, setting);
/* Replace settings and test */
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
nm_connection_replace_settings_from_connection (connection, replacement);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_con = nm_connection_get_setting_connection (connection);
g_assert (s_con);
g_assert_cmpstr (nm_setting_connection_get_id (s_con), ==, expected_id);
g_assert_cmpstr (nm_setting_connection_get_uuid (s_con), ==, uuid);
g_assert (!nm_connection_get_setting_wired (connection));
g_assert (!nm_connection_get_setting_ip6_config (connection));
g_assert (nm_connection_get_setting_wireless (connection));
g_free (uuid);
g_object_unref (replacement);
g_object_unref (connection);
}
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
static void
test_connection_replace_settings_bad (void)
{
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
NMConnection *connection, *new_connection;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *new_settings;
GVariantBuilder builder, setting_builder;
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
GError *error = NULL;
gboolean success;
NMSettingConnection *s_con;
new_connection = new_test_connection ();
g_assert (nm_connection_verify (new_connection, NULL));
s_con = nm_connection_get_setting_connection (new_connection);
g_object_set (s_con,
NM_SETTING_CONNECTION_UUID, NULL,
NM_SETTING_CONNECTION_ID, "bad-connection",
NULL);
g_assert (!nm_connection_verify (new_connection, NULL));
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
/* nm_connection_replace_settings_from_connection() should succeed */
connection = new_test_connection ();
nm_connection_replace_settings_from_connection (connection, new_connection);
g_assert_cmpstr (nm_connection_get_id (connection), ==, "bad-connection");
g_assert (!nm_connection_verify (connection, NULL));
g_object_unref (connection);
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
/* nm_connection_replace_settings() should succeed */
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
new_settings = nm_connection_to_dbus (new_connection, NM_CONNECTION_SERIALIZE_ALL);
g_assert (new_settings != NULL);
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
connection = new_test_connection ();
success = nm_connection_replace_settings (connection, new_settings, &error);
g_assert_no_error (error);
g_assert (success);
g_assert_cmpstr (nm_connection_get_id (connection), ==, "bad-connection");
g_assert (!nm_connection_verify (connection, NULL));
g_object_unref (connection);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (new_settings);
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
/* But given an invalid dict, it should fail */
g_variant_builder_init (&builder, NM_VARIANT_TYPE_CONNECTION);
g_variant_builder_init (&setting_builder, NM_VARIANT_TYPE_SETTING);
g_variant_builder_add (&builder, "{sa{sv}}",
"ip-over-avian-carrier",
&setting_builder);
new_settings = g_variant_builder_end (&builder);
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
connection = new_test_connection ();
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
success = nm_connection_replace_settings (connection, new_settings, &error);
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_SETTING);
libnm/tests: fix memleaks in tests for valgrind
2015-02-06 13:31:06 +01:00
g_clear_error (&error);
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
g_assert (!success);
libnm-core: fix up connection deserialize/copy/replace semantics libnm-util's connection deserializing/copying/replacing functions have odd semantics where sometimes they both modify a connection AND return an error. libnm-core tried to improve things by guaranteeing that the connection would not be modified if the new settings were invalid, but this ended up breaking a bunch of places that needed to be able to work with invalid connections. So re-fix the functions by reverting back to the old semantics, but having return values that clearly distinguish whether the connection was modified or not. For comparison: - nm_connection_new_from_hash() / nm_simple_connection_new_from_dbus(): - libnm-util: returns a valid connection or NULL. - OLD libnm-core: returned a valid connection or NULL. - NEW libnm-core: returns a valid connection or NULL. - nm_connection_duplicate() / nm_simple_connection_new_clone(): - libnm-util: always succeeds, whether or not the connection is valid. - OLD libnm-core: returned a valid connection or NULL - NEW libnm-core: always succeeds, whether or not the connection is valid. - nm_connection_replace_settings_from_connection(): - libnm-util: always replaces the settings, but returns FALSE if the connection is now invalid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: always replaces the settings, and has no return value. (The modified connection is valid if and only if the replaced-from connection was valid; just like with the libnm-util version.) - nm_connection_replace_settings(): - libnm-util: returns TRUE if the new settings are valid, or FALSE if either (a) the new settings could not be deserialized and the connection is unchanged, or (b) the new settings were deserialized, and the connection was updated, but is now not valid. - OLD libnm-core: either replaced the settings and returned TRUE (if the settings were valid), or else left the connection unchanged and returned FALSE (if not). - NEW libnm-core: returns TRUE if the connection was updated (whether or not it is valid), or FALSE if the new settings could not be deserialized and the connection is unchanged.
2014-09-15 14:05:52 -04:00
g_assert (nm_connection_verify (connection, NULL));
g_object_unref (connection);
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (new_settings);
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
g_object_unref (new_connection);
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
build: fix -Wstrict-prototypes warnings We disabled -Wstrict-prototypes in commit db9b1df0e47996ff8aaea468a11e1e97f64ee126 . Fix compiler warnings. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-10-29 10:44:31 +01:00
test_connection_new_from_dbus (void)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
NMConnection *connection;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *new_settings;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
GError *error = NULL;
NMSettingConnection *s_con;
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NMSettingIPConfig *s_ip6;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
char *uuid = NULL;
const char *expected_id = NULL, *expected_method = NULL;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
new_settings = new_connection_dict (&uuid, &expected_id, &expected_method);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (new_settings);
/* Replace settings and test */
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
connection = _connection_new_from_dbus (new_settings, &error);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert_no_error (error);
g_assert (connection);
s_con = nm_connection_get_setting_connection (connection);
g_assert (s_con);
g_assert_cmpstr (nm_setting_connection_get_id (s_con), ==, expected_id);
g_assert_cmpstr (nm_setting_connection_get_uuid (s_con), ==, uuid);
g_assert (nm_connection_get_setting_wired (connection));
libnm: normalize connections in nm_simple_connection_new_from_dbus() Before we would just call verify() and only return valid connections without attempting to fix them. It is better to use normalize(), because that function is especially there to accept and repair deprecated configurations that would no longer verify(). This changes behavior in the way that the function now accepts connections that would have been rejected before. Since commit b88715e05bc16c40ae68958c58aa86c2ad029a82 normalize() also adds a missing UUID. Hence this also affects the DBUS method 'AddConnection' in that it now accepts connections without UUID. Previously, clients were required to set a UUID for the new connection, now NM core can create a random one if no UUID is set. https://bugzilla.gnome.org/show_bug.cgi?id=740813
2014-11-27 15:59:08 +01:00
g_assert (nm_connection_get_setting_ip4_config (connection));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_ip6 = nm_connection_get_setting_ip6_config (connection);
g_assert (s_ip6);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
g_assert_cmpstr (nm_setting_ip_config_get_method (s_ip6), ==, expected_method);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_free (uuid);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (new_settings);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (connection);
}
static void
check_permission (NMSettingConnection *s_con,
guint32 idx,
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
const char *expected_uname)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
gboolean success;
const char *ptype = NULL, *pitem = NULL, *detail = NULL;
success = nm_setting_connection_get_permission (s_con, 0, &ptype, &pitem, &detail);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpstr (ptype, ==, "user");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (pitem);
g_assert_cmpstr (pitem, ==, expected_uname);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!detail);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
#define TEST_UNAME "asdfasfasdf"
static void
test_setting_connection_permissions_helpers (void)
{
NMSettingConnection *s_con;
gboolean success;
char buf[9] = { 0x61, 0x62, 0x63, 0xff, 0xfe, 0xfd, 0x23, 0x01, 0x00 };
libnm-core: change list-of-string and array-of-string properties to G_TYPE_STRV Change all DBUS_TYPE_G_LIST_OF_STRING and DBUS_TYPE_G_ARRAY_OF_STRING properties to G_TYPE_STRV, and update everything accordingly. (This doesn't actually require using _nm_setting_class_transform_property(); dbus-glib is happy to transform between 'as' and G_TYPE_STRV.)
2014-08-21 13:19:53 -04:00
char **perms;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
const char *expected_perm = "user:" TEST_UNAME ":";
s_con = NM_SETTING_CONNECTION (nm_setting_connection_new ());
/* Ensure a bad [type] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (strcmp (ptype, "user") == 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_setting_connection_add_permission (s_con, "foobar", "blah", NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a bad [type] is rejected */
libnm: don't use strlen() for checking for non-empty string It's well understood that these are NUL terminated strings. We don't need strlen() to check that the strings aren't empty.
2019-03-16 21:21:43 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (ptype && ptype[0]));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_setting_connection_add_permission (s_con, NULL, "blah", NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a bad [item] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (uname));
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (p != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_setting_connection_add_permission (s_con, "user", NULL, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a bad [item] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (uname[0] != '\0'));
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (p != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_setting_connection_add_permission (s_con, "user", "", NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure an [item] with ':' is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (strchr (uname, ':') == NULL));
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (p != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_setting_connection_add_permission (s_con, "user", "ad:asdf", NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a non-UTF-8 [item] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (g_utf8_validate (uname, -1, NULL) == TRUE));
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (p != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_setting_connection_add_permission (s_con, "user", buf, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a non-NULL [detail] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (detail == NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
success = nm_setting_connection_add_permission (s_con, "user", "dafasdf", "asdf");
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (!success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a valid call results in success */
success = nm_setting_connection_add_permission (s_con, "user", TEST_UNAME, NULL);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (success);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 1);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
check_permission (s_con, 0, TEST_UNAME);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Check the actual GObject property just to be paranoid */
libnm-core: change list-of-string and array-of-string properties to G_TYPE_STRV Change all DBUS_TYPE_G_LIST_OF_STRING and DBUS_TYPE_G_ARRAY_OF_STRING properties to G_TYPE_STRV, and update everything accordingly. (This doesn't actually require using _nm_setting_class_transform_property(); dbus-glib is happy to transform between 'as' and G_TYPE_STRV.)
2014-08-21 13:19:53 -04:00
g_object_get (G_OBJECT (s_con), NM_SETTING_CONNECTION_PERMISSIONS, &perms, NULL);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert (perms);
g_assert_cmpint (g_strv_length (perms), ==, 1);
g_assert_cmpstr (perms[0], ==, expected_perm);
libnm-core: change list-of-string and array-of-string properties to G_TYPE_STRV Change all DBUS_TYPE_G_LIST_OF_STRING and DBUS_TYPE_G_ARRAY_OF_STRING properties to G_TYPE_STRV, and update everything accordingly. (This doesn't actually require using _nm_setting_class_transform_property(); dbus-glib is happy to transform between 'as' and G_TYPE_STRV.)
2014-08-21 13:19:53 -04:00
g_strfreev (perms);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Now remove that permission and ensure we have 0 permissions */
nm_setting_connection_remove_permission (s_con, 0);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (s_con);
}
static void
add_permission_property (NMSettingConnection *s_con,
const char *ptype,
const char *pitem,
int pitem_len,
const char *detail)
{
GString *str;
libnm-core: change list-of-string and array-of-string properties to G_TYPE_STRV Change all DBUS_TYPE_G_LIST_OF_STRING and DBUS_TYPE_G_ARRAY_OF_STRING properties to G_TYPE_STRV, and update everything accordingly. (This doesn't actually require using _nm_setting_class_transform_property(); dbus-glib is happy to transform between 'as' and G_TYPE_STRV.)
2014-08-21 13:19:53 -04:00
char *perms[2];
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
str = g_string_sized_new (50);
if (ptype)
g_string_append (str, ptype);
g_string_append_c (str, ':');
if (pitem) {
if (pitem_len >= 0)
g_string_append_len (str, pitem, pitem_len);
else
g_string_append (str, pitem);
}
g_string_append_c (str, ':');
if (detail)
g_string_append (str, detail);
libnm-core: change list-of-string and array-of-string properties to G_TYPE_STRV Change all DBUS_TYPE_G_LIST_OF_STRING and DBUS_TYPE_G_ARRAY_OF_STRING properties to G_TYPE_STRV, and update everything accordingly. (This doesn't actually require using _nm_setting_class_transform_property(); dbus-glib is happy to transform between 'as' and G_TYPE_STRV.)
2014-08-21 13:19:53 -04:00
perms[0] = str->str;
perms[1] = NULL;
g_object_set (G_OBJECT (s_con), NM_SETTING_CONNECTION_PERMISSIONS, perms, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_string_free (str, TRUE);
}
static void
test_setting_connection_permissions_property (void)
{
NMSettingConnection *s_con;
gboolean success;
char buf[9] = { 0x61, 0x62, 0x63, 0xff, 0xfe, 0xfd, 0x23, 0x01, 0x00 };
s_con = NM_SETTING_CONNECTION (nm_setting_connection_new ());
/* Ensure a bad [type] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (strncmp (str, PERM_USER_PREFIX, strlen (PERM_USER_PREFIX)) == 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, "foobar", "blah", -1, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a bad [type] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (strncmp (str, PERM_USER_PREFIX, strlen (PERM_USER_PREFIX)) == 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, NULL, "blah", -1, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a bad [item] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (last_colon > str));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, "user", NULL, -1, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a bad [item] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (last_colon > str));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, "user", "", -1, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure an [item] with ':' in the middle is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (str[i] != ':'));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, "user", "ad:asdf", -1, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure an [item] with ':' at the end is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (str[i] != ':'));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, "user", "adasdfaf:", -1, NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a non-UTF-8 [item] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (g_utf8_validate (str, -1, NULL) == TRUE));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, "user", buf, (int) sizeof (buf), NULL);
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a non-NULL [detail] is rejected */
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (*(last_colon + 1) == '\0'));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_permission_property (s_con, "user", "dafasdf", -1, "asdf");
g_test_assert_expected_messages ();
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Ensure a valid call results in success */
success = nm_setting_connection_add_permission (s_con, "user", TEST_UNAME, NULL);
all: get rid of a handful of unused-but-set variables
2017-12-18 08:36:51 +01:00
g_assert (success);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 1);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
check_permission (s_con, 0, TEST_UNAME);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Now remove that permission and ensure we have 0 permissions */
nm_setting_connection_remove_permission (s_con, 0);
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
g_assert_cmpint (nm_setting_connection_get_num_permissions (s_con), ==, 0);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (s_con);
}
static void
test_connection_compare_same (void)
{
NMConnection *a, *b;
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (nm_connection_compare (a, b, NM_SETTING_COMPARE_FLAG_EXACT));
g_object_unref (a);
g_object_unref (b);
}
static void
test_connection_compare_key_only_in_a (void)
{
NMConnection *a, *b;
NMSettingConnection *s_con;
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_con = (NMSettingConnection *) nm_connection_get_setting (b, NM_TYPE_SETTING_CONNECTION);
g_assert (s_con);
g_object_set (s_con, NM_SETTING_CONNECTION_TIMESTAMP, (guint64) 0, NULL);
g_assert (!nm_connection_compare (a, b, NM_SETTING_COMPARE_FLAG_EXACT));
g_object_unref (a);
g_object_unref (b);
}
static void
test_connection_compare_setting_only_in_a (void)
{
NMConnection *a, *b;
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_remove_setting (b, NM_TYPE_SETTING_IP4_CONFIG);
g_assert (!nm_connection_compare (a, b, NM_SETTING_COMPARE_FLAG_EXACT));
g_object_unref (a);
g_object_unref (b);
}
static void
test_connection_compare_key_only_in_b (void)
{
NMConnection *a, *b;
NMSettingConnection *s_con;
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_con = (NMSettingConnection *) nm_connection_get_setting (b, NM_TYPE_SETTING_CONNECTION);
g_assert (s_con);
g_object_set (s_con, NM_SETTING_CONNECTION_TIMESTAMP, (guint64) 0, NULL);
g_assert (!nm_connection_compare (a, b, NM_SETTING_COMPARE_FLAG_EXACT));
g_object_unref (a);
g_object_unref (b);
}
static void
test_connection_compare_setting_only_in_b (void)
{
NMConnection *a, *b;
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_remove_setting (a, NM_TYPE_SETTING_IP4_CONFIG);
g_assert (!nm_connection_compare (a, b, NM_SETTING_COMPARE_FLAG_EXACT));
g_object_unref (a);
g_object_unref (b);
}
typedef struct {
const char *key_name;
guint32 result;
} DiffKey;
typedef struct {
const char *name;
DiffKey keys[30];
} DiffSetting;
#define ARRAY_LEN(a) (sizeof (a) / sizeof (a[0]))
static void
ensure_diffs (GHashTable *diffs, const DiffSetting *check, gsize n_check)
{
guint i;
g_assert (g_hash_table_size (diffs) == n_check);
/* Loop through the settings */
for (i = 0; i < n_check; i++) {
GHashTable *setting_hash;
guint z = 0;
setting_hash = g_hash_table_lookup (diffs, check[i].name);
g_assert (setting_hash);
/* Get the number of keys to check */
while (check[i].keys[z].key_name)
z++;
g_assert (g_hash_table_size (setting_hash) == z);
/* Now compare the actual keys */
for (z = 0; check[i].keys[z].key_name; z++) {
NMSettingDiffResult result;
result = GPOINTER_TO_UINT (g_hash_table_lookup (setting_hash, check[i].keys[z].key_name));
g_assert (result == check[i].keys[z].result);
}
}
}
static void
test_connection_diff_a_only (void)
{
NMConnection *connection;
GHashTable *out_diffs = NULL;
gboolean same;
const DiffSetting settings[] = {
{ NM_SETTING_CONNECTION_SETTING_NAME, {
{ NM_SETTING_CONNECTION_ID, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_UUID, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add NMSettingConnection:stable-id property This new property be used as token to generate stable-ids instead of the connection's UUID. Later, this will be used by ipv6.addr-gen-mode=stable-privacy, ethernet.cloned-mac-address=stable, and wifi.cloned-mac-address=stable setting. Those generate stable addresses based on the connection's UUID, but allow to use the stable-id instead. This allows multiple connections to generate the same addresses -- on the same machine, because in the above cases a machine dependant key is also hashed.
2016-06-21 16:41:05 +02:00
{ NM_SETTING_CONNECTION_STABLE_ID, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NM_SETTING_CONNECTION_INTERFACE_NAME, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_TYPE, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_TIMESTAMP, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_AUTOCONNECT, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add NMSettingConnection:autoconnect-priority setting The autoconnect priority has only any relevance, if the connection is autoconnect too. The priority defaults to zero, with higher numbers meaning preferred. The valid range is limited to [-999,999]. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-25 20:39:40 +02:00
{ NM_SETTING_CONNECTION_AUTOCONNECT_PRIORITY, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: introduce connection.autoconnect-retries property While technically it's already possible to implement a fail-over mechanism using multiple connections (for example, defining a higher priority DHCP connection with short DHCP timeout and a lower priority one with static address), in practice this doesn't work well as we try to autoactivate each connection 4 times before switching to the next one. Introduce a connection.autoconnect-retries property that can be used to change the number of retries. The special value 0 means infinite and can be used to try the connection forever. A -1 value means the global configured default, which is equal to 4 unless overridden. https://bugzilla.gnome.org/show_bug.cgi?id=763524
2016-10-08 10:27:15 +02:00
{ NM_SETTING_CONNECTION_AUTOCONNECT_RETRIES, NM_SETTING_DIFF_RESULT_IN_A },
all: add connection.multi-connect property for wildcard profiles Add a new option that allows to activate a profile multiple times (at the same time). Previoulsy, all profiles were implicitly NM_SETTING_CONNECTION_MULTI_CONNECT_SINGLE, meaning, that activating a profile that is already active will deactivate it first. This will make more sense, as we also add more match-options how profiles can be restricted to particular devices. We already have connection.type, connection.interface-name, and (ethernet|wifi).mac-address to restrict a profile to particular devices. For example, it is however not possible to specify a wildcard like "eth*" to match a profile to a set of devices by interface-name. That is another missing feature, and once we extend the matching capabilities, it makes more sense to activate a profile multiple times. See also https://bugzilla.redhat.com/show_bug.cgi?id=997998, which previously changed that a connection is restricted to a single activation at a time. This work relaxes that again. This only adds the new property, it is not used nor implemented yet. https://bugzilla.redhat.com/show_bug.cgi?id=1555012
2018-04-10 11:45:35 +02:00
{ NM_SETTING_CONNECTION_MULTI_CONNECT, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NM_SETTING_CONNECTION_READ_ONLY, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_PERMISSIONS, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_ZONE, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_MASTER, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add autoconnect-slaves property to NMSettingConnection The property is used for controlling whether slaves should be brought up with a master connection. If 0, activating the master will not activate slaves. But if set to 1, activating the master will bring up slaves as well. The property can have the third state (-1), meaning that the value is default. That is either a value set in the configuration file for the property, or 0.
2015-05-06 10:54:35 +02:00
{ NM_SETTING_CONNECTION_AUTOCONNECT_SLAVES, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NM_SETTING_CONNECTION_SECONDARIES, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_GATEWAY_PING_TIMEOUT, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: add 'metered' property to NMSettingConnection Add a 'metered' enum property to NMSettingConnection with possible values: unknown,yes,no. The value indicates the presence of limitations in the amount of traffic flowing through the connection.
2015-04-27 17:45:53 +02:00
{ NM_SETTING_CONNECTION_METERED, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add 'lldp' property to NMSettingConnection Add the 'lldp' property to NMSettingConnection, which specifies whether the reception and parsing of LLDP frames to discover neighbor devices should be enabled.
2015-10-07 11:48:30 +02:00
{ NM_SETTING_CONNECTION_LLDP, NM_SETTING_DIFF_RESULT_IN_A },
all: move setting 802-1x.auth-retries to connection.auth-retries The number of authentication retires is useful also for passwords aside 802-1x settings. For example, src/devices/wifi/nm-device-wifi.c also has a retry counter and uses a hard-coded value of 3. Move the setting, so that it can be used in general. Although it is still not implemented for other settings. This is an API and ABI break.
2017-11-02 09:25:40 +01:00
{ NM_SETTING_CONNECTION_AUTH_RETRIES, NM_SETTING_DIFF_RESULT_IN_A },
mdns: add new connection property. Add support for mDNS as a connection-level property. Update ifcfg-rh and keyfile plugins to support it.
2017-11-10 14:49:27 +02:00
{ NM_SETTING_CONNECTION_MDNS, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: add connection.llmnr property
2018-08-31 10:48:18 +02:00
{ NM_SETTING_CONNECTION_LLMNR, NM_SETTING_DIFF_RESULT_IN_A },
dhcp: add support for MUD URL (RFC 8520) [thaller@redhat.com: rewritten commit message] https://tools.ietf.org/html/rfc8520 https://blog.apnic.net/2019/05/14/protecting-the-internet-of-things-with-mud/ https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/402 https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/463
2020-04-08 07:30:08 +00:00
{ NM_SETTING_CONNECTION_MUD_URL, NM_SETTING_DIFF_RESULT_IN_A },
libnm,cli,ifcfg-rh: add connection:wait-device-timeout property Initscripts already honor the DEVTIMEOUT variable (rh #1171917). Don't make this a property only supported by initscripts. Every useful property should also be supported by keyfile and it should be accessible via D-Bus. Also, I will soon drop NMSIfcfgConnection, so handling this would require extra code. It's easier when DEVTIMEOUT is a regular property of the connection profile. The property is not yet implemented. ifcfg-rh still uses the old implementation, and keyfile is not yet adjusted. Since both keyfile and ifcfg-rh will both be rewritten soon, this property will be implemented then.
2019-07-01 10:01:28 +02:00
{ NM_SETTING_CONNECTION_WAIT_DEVICE_TIMEOUT, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NULL, NM_SETTING_DIFF_RESULT_UNKNOWN }
} },
{ NM_SETTING_WIRED_SETTING_NAME, {
{ NM_SETTING_WIRED_PORT, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_SPEED, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_DUPLEX, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_AUTO_NEGOTIATE, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_MAC_ADDRESS, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_CLONED_MAC_ADDRESS, NM_SETTING_DIFF_RESULT_IN_A },
all: make MAC address randomization algorithm configurable For the per-connection settings "ethernet.cloned-mac-address" and "wifi.cloned-mac-address", and for the per-device setting "wifi.scan-rand-mac-address", we may generate MAC addresses using either the "random" or "stable" algorithm. Add new properties "generate-mac-address-mask" that allow to configure which bits of the MAC address will be scrambled. By default, the "random" and "stable" algorithms scamble all bits of the MAC address, including the OUI part and generate a locally- administered, unicast address. By specifying a MAC address mask, we can now configure to perserve parts of the current MAC address of the device. For example, setting "FF:FF:FF:00:00:00" will preserve the first 3 octects of the current MAC address. One can also explicitly specify a MAC address to use instead of the current MAC address. For example, "FF:FF:FF:00:00:00 68:F7:28:00:00:00" sets the OUI part of the MAC address to "68:F7:28" while scrambling the last 3 octects. Similarly, "02:00:00:00:00:00 00:00:00:00:00:00" will scamble all bits of the MAC address, except clearing the second-least significant bit. Thus, creating a burned-in address, globally administered. One can also supply a list of MAC addresses like "FF:FF:FF:00:00:00 68:F7:28:00:00:00 00:0C:29:00:00:00 ..." in which case a MAC address is choosen randomly. To fully scamble the MAC address one can configure "02:00:00:00:00:00 00:00:00:00:00:00 02:00:00:00:00:00". which also randomly creates either a locally or globally administered address. With this, the following macchanger options can be implemented: `macchanger --random` This is the default if no mask is configured. -> "" while is the same as: -> "00:00:00:00:00:00" -> "02:00:00:00:00:00 02:00:00:00:00:00" `macchanger --random --bia` -> "02:00:00:00:00:00 00:00:00:00:00:00" `macchanger --ending` This option cannot be fully implemented, because macchanger uses the current MAC address but also implies --bia. -> "FF:FF:FF:00:00:00" This would yields the same result only if the current MAC address is already a burned-in address too. Otherwise, it has not the same effect as --ending. -> "FF:FF:FF:00:00:00 <MAC_ADDR>" Alternatively, instead of using the current MAC address, spell the OUI part out. But again, that is not really the same as macchanger does because you explictly have to name the OUI part to use. `machanger --another` `machanger --another_any` -> "FF:FF:FF:00:00:00 <MAC_ADDR> <MAC_ADDR> ..." "$(printf "FF:FF:FF:00:00:00 %s\n" "$(sed -n 's/^\([0-9a-fA-F][0-9a-fA-F]\) \([0-9a-fA-F][0-9a-fA-F]\) \([0-9a-fA-F][0-9a-fA-F]\) .*/\1:\2:\3:00:00:00/p' /usr/share/macchanger/wireless.list | xargs)")"
2016-06-22 20:31:39 +02:00
{ NM_SETTING_WIRED_GENERATE_MAC_ADDRESS_MASK, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NM_SETTING_WIRED_MAC_ADDRESS_BLACKLIST, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_MTU, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_S390_SUBCHANNELS, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_S390_NETTYPE, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_S390_OPTIONS, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: add Wake-on-LAN properties to NMSettingWired
2015-05-11 21:59:51 +02:00
{ NM_SETTING_WIRED_WAKE_ON_LAN, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_WIRED_WAKE_ON_LAN_PASSWORD, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NULL, NM_SETTING_DIFF_RESULT_UNKNOWN },
} },
{ NM_SETTING_IP4_CONFIG_SETTING_NAME, {
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
{ NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_IP_CONFIG_DNS, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_IP_CONFIG_DNS_SEARCH, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: add dns-options property to NMSettingIPConfig
2015-03-26 08:57:02 +01:00
{ NM_SETTING_IP_CONFIG_DNS_OPTIONS, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
{ NM_SETTING_IP_CONFIG_ADDRESSES, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
{ NM_SETTING_IP_CONFIG_GATEWAY, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
{ NM_SETTING_IP_CONFIG_ROUTES, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add NMSettingIPConfig:route-metric https://bugzilla.gnome.org/show_bug.cgi?id=735512 https://bugzilla.redhat.com/show_bug.cgi?id=663730 Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-27 19:57:24 +02:00
{ NM_SETTING_IP_CONFIG_ROUTE_METRIC, NM_SETTING_DIFF_RESULT_IN_A },
all: rework configuring route table support by adding "route-table" setting We added "ipv4.route-table-sync" and "ipv6.route-table-sync" to not change behavior for users that configured policy routing outside of NetworkManager, for example, via a dispatcher script. Users had to explicitly opt-in for NetworkManager to fully manage all routing tables. These settings were awkward. Replace them with new settings "ipv4.route-table" and "ipv6.route-table". Note that this commit breaks API/ABI on the unstable development branch by removing recently added API. As before, a connection will have no route-table set by default. This has the meaning that policy-routing is not enabled and only the main table will be fully synced. Once the user sets a table, we recognize that and NetworkManager manages all routing tables. The new route-table setting has other important uses: analog to "ipv4.route-metric", it is the default that applies to all routes. Currently it only works for static routes, not DHCP, SLAAC, default-route, etc. That will be implemented later. For static routes, each route still can explicitly set a table, and overwrite the per-connection setting in "ipv4.route-table" and "ipv6.route-table".
2017-09-28 08:40:41 +02:00
{ NM_SETTING_IP_CONFIG_ROUTE_TABLE, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add API to NMSettingIPConfig for routing rules
2019-03-14 12:04:21 +01:00
{ NM_SETTING_IP_CONFIG_ROUTING_RULES, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
{ NM_SETTING_IP_CONFIG_IGNORE_AUTO_ROUTES, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_IP_CONFIG_IGNORE_AUTO_DNS, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add NMSettingIPConfig:route-metric https://bugzilla.gnome.org/show_bug.cgi?id=735512 https://bugzilla.redhat.com/show_bug.cgi?id=663730 Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-27 19:57:24 +02:00
{ NM_SETTING_IP4_CONFIG_DHCP_CLIENT_ID, NM_SETTING_DIFF_RESULT_IN_A },
libnm,core,cli: move dhcp-timeout property to generic NMSettingIPConfig The property applies to both IPv4 and IPv6 and so it should not be in NMSettingIP4Config but in the base class.
2016-02-03 11:12:07 +01:00
{ NM_SETTING_IP_CONFIG_DHCP_TIMEOUT, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
{ NM_SETTING_IP_CONFIG_DHCP_SEND_HOSTNAME, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_IP_CONFIG_DHCP_HOSTNAME, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add ipvx.dhcp-hostname-flags properties When using the dhclient DHCP backend users can tweak the behavior in the dhclient configuration file. One of the options that was reported as useful in the past was the FQDN flags [1] [2]. Add native support for FQDN flags to NM by introducing new ipv{4,6}.dhcp-hostname-flags properties. [1] https://bugzilla.redhat.com/show_bug.cgi?id=1684595 [2] https://bugzilla.redhat.com/show_bug.cgi?id=1255507
2019-07-09 11:26:02 +02:00
{ NM_SETTING_IP_CONFIG_DHCP_HOSTNAME_FLAGS,NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: add a 'dhcp-fqdn' property to NMSettingIP4Config The property contains the fully qualified domain name to be sent to DHCP server using the FQDN option. The property is mutually exclusive with 'dhcp-hostname'.
2015-10-13 14:10:01 +02:00
{ NM_SETTING_IP4_CONFIG_DHCP_FQDN, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
{ NM_SETTING_IP_CONFIG_NEVER_DEFAULT, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_IP_CONFIG_MAY_FAIL, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add NMSettingIPConfig 'dad-timeout' property The property is used to control duplicate address detection: * -1 means default value * 0 means no DAD is performed * > 0 means timeout (in milliseconds) for arping responses [bgalvani: moved setting from NMSettingIP4Config]
2015-09-07 18:00:10 +02:00
{ NM_SETTING_IP_CONFIG_DAD_TIMEOUT, NM_SETTING_DIFF_RESULT_IN_A },
libnm-core: add dns-priority to NMSettingIPConfig
2016-04-23 15:57:04 +02:00
{ NM_SETTING_IP_CONFIG_DNS_PRIORITY, NM_SETTING_DIFF_RESULT_IN_A },
all: add ipvX.dhcp-iaid properties Add new ipv4.dhcp-iaid and ipv6.dhcp-iaid properties to specify a DHCP IAID.
2019-10-15 10:27:10 +02:00
{ NM_SETTING_IP_CONFIG_DHCP_IAID, NM_SETTING_DIFF_RESULT_IN_A },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NULL, NM_SETTING_DIFF_RESULT_UNKNOWN },
} },
};
connection = new_test_connection ();
same = nm_connection_diff (connection, NULL, NM_SETTING_COMPARE_FLAG_EXACT, &out_diffs);
g_assert (same == FALSE);
g_assert (out_diffs != NULL);
g_assert (g_hash_table_size (out_diffs) > 0);
ensure_diffs (out_diffs, settings, ARRAY_LEN (settings));
g_hash_table_destroy (out_diffs);
g_object_unref (connection);
}
static void
test_connection_diff_same (void)
{
NMConnection *a, *b;
GHashTable *out_diffs = NULL;
gboolean same;
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
same = nm_connection_diff (a, b, NM_SETTING_COMPARE_FLAG_EXACT, &out_diffs);
g_assert (same == TRUE);
g_assert (out_diffs == NULL);
g_object_unref (a);
g_object_unref (b);
}
static void
test_connection_diff_different (void)
{
NMConnection *a, *b;
GHashTable *out_diffs = NULL;
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NMSettingIPConfig *s_ip4;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
gboolean same;
const DiffSetting settings[] = {
{ NM_SETTING_IP4_CONFIG_SETTING_NAME, {
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
{ NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_DIFF_RESULT_IN_A | NM_SETTING_DIFF_RESULT_IN_B },
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{ NULL, NM_SETTING_DIFF_RESULT_UNKNOWN },
} },
};
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
s_ip4 = nm_connection_get_setting_ip4_config (a);
g_assert (s_ip4);
g_object_set (G_OBJECT (s_ip4),
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_MANUAL,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NULL);
same = nm_connection_diff (a, b, NM_SETTING_COMPARE_FLAG_EXACT, &out_diffs);
g_assert (same == FALSE);
g_assert (out_diffs != NULL);
g_assert (g_hash_table_size (out_diffs) > 0);
ensure_diffs (out_diffs, settings, ARRAY_LEN (settings));
g_hash_table_destroy (out_diffs);
g_object_unref (a);
g_object_unref (b);
}
static void
test_connection_diff_no_secrets (void)
{
NMConnection *a, *b;
GHashTable *out_diffs = NULL;
NMSetting *s_pppoe;
gboolean same;
const DiffSetting settings[] = {
{ NM_SETTING_PPPOE_SETTING_NAME, {
{ NM_SETTING_PPPOE_PASSWORD, NM_SETTING_DIFF_RESULT_IN_B },
{ NULL, NM_SETTING_DIFF_RESULT_UNKNOWN },
} },
};
a = new_test_connection ();
s_pppoe = nm_setting_pppoe_new ();
g_object_set (G_OBJECT (s_pppoe),
NM_SETTING_PPPOE_USERNAME, "thomas",
NULL);
nm_connection_add_setting (a, s_pppoe);
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Add a secret to B */
s_pppoe = NM_SETTING (nm_connection_get_setting_pppoe (b));
g_assert (s_pppoe);
g_object_set (G_OBJECT (s_pppoe),
NM_SETTING_PPPOE_PASSWORD, "secretpassword",
NULL);
/* Make sure the diff returns no results as secrets are ignored */
same = nm_connection_diff (a, b, NM_SETTING_COMPARE_FLAG_IGNORE_SECRETS, &out_diffs);
g_assert (same == TRUE);
g_assert (out_diffs == NULL);
/* Now make sure the diff returns results if secrets are not ignored */
same = nm_connection_diff (a, b, NM_SETTING_COMPARE_FLAG_EXACT, &out_diffs);
g_assert (same == FALSE);
g_assert (out_diffs != NULL);
g_assert (g_hash_table_size (out_diffs) > 0);
ensure_diffs (out_diffs, settings, ARRAY_LEN (settings));
g_hash_table_destroy (out_diffs);
g_object_unref (a);
g_object_unref (b);
}
static void
test_connection_diff_inferrable (void)
{
NMConnection *a, *b;
GHashTable *out_diffs = NULL;
gboolean same;
NMSettingConnection *s_con;
NMSettingWired *s_wired;
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NMSettingIPConfig *s_ip4;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
char *uuid;
const DiffSetting settings[] = {
{ NM_SETTING_CONNECTION_SETTING_NAME, {
{ NM_SETTING_CONNECTION_INTERFACE_NAME, NM_SETTING_DIFF_RESULT_IN_A },
{ NULL, NM_SETTING_DIFF_RESULT_UNKNOWN },
} },
};
a = new_test_connection ();
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
b = nm_simple_connection_new_clone (a);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Change the UUID, wired MTU, and set ignore-auto-dns */
s_con = nm_connection_get_setting_connection (a);
g_assert (s_con);
uuid = nm_utils_uuid_generate ();
g_object_set (G_OBJECT (s_con),
NM_SETTING_CONNECTION_UUID, uuid,
NM_SETTING_CONNECTION_ID, "really neat connection",
NULL);
g_free (uuid);
s_wired = nm_connection_get_setting_wired (a);
g_assert (s_wired);
g_object_set (G_OBJECT (s_wired), NM_SETTING_WIRED_MTU, 300, NULL);
s_ip4 = nm_connection_get_setting_ip4_config (a);
g_assert (s_ip4);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
g_object_set (G_OBJECT (s_ip4), NM_SETTING_IP_CONFIG_IGNORE_AUTO_DNS, TRUE, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Make sure the diff returns no results as secrets are ignored */
same = nm_connection_diff (a, b, NM_SETTING_COMPARE_FLAG_INFERRABLE, &out_diffs);
g_assert (same == TRUE);
g_assert (out_diffs == NULL);
/* And change a INFERRABLE property to ensure that it shows up in the diff results */
g_object_set (G_OBJECT (s_con), NM_SETTING_CONNECTION_INTERFACE_NAME, "usb0", NULL);
/* Make sure the diff returns no results as secrets are ignored */
same = nm_connection_diff (a, b, NM_SETTING_COMPARE_FLAG_INFERRABLE, &out_diffs);
g_assert (same == FALSE);
g_assert (out_diffs != NULL);
g_assert (g_hash_table_size (out_diffs) > 0);
ensure_diffs (out_diffs, settings, ARRAY_LEN (settings));
g_hash_table_destroy (out_diffs);
g_object_unref (a);
g_object_unref (b);
}
static void
add_generic_settings (NMConnection *connection, const char *ctype)
{
NMSetting *setting;
char *uuid;
uuid = nm_utils_uuid_generate ();
setting = nm_setting_connection_new ();
g_object_set (setting,
NM_SETTING_CONNECTION_ID, "asdfasdfadf",
NM_SETTING_CONNECTION_TYPE, ctype,
NM_SETTING_CONNECTION_UUID, uuid,
NULL);
nm_connection_add_setting (connection, setting);
g_free (uuid);
setting = nm_setting_ip4_config_new ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
g_object_set (setting, NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_AUTO, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, setting);
setting = nm_setting_ip6_config_new ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
g_object_set (setting, NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_AUTO, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, setting);
}
static void
test_connection_good_base_types (void)
{
NMConnection *connection;
NMSetting *setting;
gboolean success;
GError *error = NULL;
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
GBytes *ssid;
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
const char *bdaddr = "11:22:33:44:55:66";
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Try a basic wired connection */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_WIRED_SETTING_NAME);
setting = nm_setting_wired_new ();
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
g_assert_no_error (error);
g_assert (success);
g_object_unref (connection);
/* Try a wired PPPoE connection */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_PPPOE_SETTING_NAME);
setting = nm_setting_pppoe_new ();
g_object_set (setting, NM_SETTING_PPPOE_USERNAME, "bob smith", NULL);
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
g_assert_no_error (error);
g_assert (success);
g_object_unref (connection);
/* Wifi connection */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_WIRELESS_SETTING_NAME);
setting = nm_setting_wireless_new ();
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
ssid = g_bytes_new ("1234567", 7);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (setting,
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
NM_SETTING_WIRELESS_SSID, ssid,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NM_SETTING_WIRELESS_MODE, "infrastructure",
NULL);
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
g_bytes_unref (ssid);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
g_assert_no_error (error);
g_assert (success);
g_object_unref (connection);
/* Bluetooth connection */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_BLUETOOTH_SETTING_NAME);
setting = nm_setting_bluetooth_new ();
g_object_set (setting,
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
NM_SETTING_BLUETOOTH_BDADDR, bdaddr,
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NM_SETTING_CONNECTION_TYPE, NM_SETTING_BLUETOOTH_TYPE_PANU,
NULL);
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
g_assert_no_error (error);
g_assert (success);
g_object_unref (connection);
/* WiMAX connection */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_WIMAX_SETTING_NAME);
setting = nm_setting_wimax_new ();
g_object_set (setting, NM_SETTING_WIMAX_NETWORK_NAME, "CLEAR", NULL);
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
g_assert_no_error (error);
g_assert (success);
g_object_unref (connection);
/* GSM connection */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_GSM_SETTING_NAME);
setting = nm_setting_gsm_new ();
g_object_set (setting,
NM_SETTING_GSM_APN, "metered.billing.sucks",
NULL);
nm_connection_add_setting (connection, setting);
/* CDMA connection */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_CDMA_SETTING_NAME);
setting = nm_setting_cdma_new ();
g_object_set (setting,
NM_SETTING_CDMA_NUMBER, "#777",
NM_SETTING_CDMA_USERNAME, "foobar@vzw.com",
NULL);
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
g_assert_no_error (error);
g_assert (success);
g_object_unref (connection);
}
static void
test_connection_bad_base_types (void)
{
NMConnection *connection;
NMSetting *setting;
gboolean success;
GError *error = NULL;
/* Test various non-base connection types to make sure they are rejected;
* using a fake 'wired' connection so the rest of it verifies
*/
/* Connection setting */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_CONNECTION_SETTING_NAME);
setting = nm_setting_wired_new ();
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm-core: clean up NMConnectionError Rename NM_CONNECTION_ERROR_UNKNOWN to NM_CONNECTION_ERROR_FAILED, following GError best practices. Replace NM_CONNECTION_ERROR_CONNECTION_SETTING_NOT_FOUND ("no NMSettingConnection") with a more generic NM_CONNECTION_ERROR_MISSING_SETTING. Use that new code in a few places that had previously been using NM_CONNECTION_ERROR_SETTING_NOT_FOUND, which was supposed to mean "the setting that you asked about doesn't exist", not "the connection is invalid because it's missing a required setting". Clarify that NM_CONNECTION_ERROR_INVALID_SETTING can be used for any "invalid or inappropriate NMSetting", not just a "conflicting" one. (But fix a case in nm_connection_update_secrets() that was returning INVALID_SETTING when it should have been return-if-failing instead.) For both MISSING_SETTING and INVALID_SETTING, always prefix the error message with "setting-name: ", just like we do with the various NMSetting MISSING_PROPERTY and INVALID_PROPERTY errors. And make sure that the error message is marked for localization. Drop NM_CONNECTION_ERROR_CONNECTION_TYPE_INVALID, which is pretty pointless; it was only used in the case where connection.type was the name of a valid setting type that is not a base setting type. Instead, just return NM_SETTING_CONNECTION_ERROR_INVALID_PROPERTY for connection.type in this case (which is what the code already did when connection.type was completely unrecognized).
2014-10-11 15:02:16 -04:00
g_assert (g_str_has_prefix (error->message, "connection.type: "));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (success == FALSE);
g_object_unref (connection);
g_clear_error (&error);
/* PPP setting */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_PPP_SETTING_NAME);
setting = nm_setting_wired_new ();
nm_connection_add_setting (connection, setting);
setting = nm_setting_ppp_new ();
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm-core: clean up NMConnectionError Rename NM_CONNECTION_ERROR_UNKNOWN to NM_CONNECTION_ERROR_FAILED, following GError best practices. Replace NM_CONNECTION_ERROR_CONNECTION_SETTING_NOT_FOUND ("no NMSettingConnection") with a more generic NM_CONNECTION_ERROR_MISSING_SETTING. Use that new code in a few places that had previously been using NM_CONNECTION_ERROR_SETTING_NOT_FOUND, which was supposed to mean "the setting that you asked about doesn't exist", not "the connection is invalid because it's missing a required setting". Clarify that NM_CONNECTION_ERROR_INVALID_SETTING can be used for any "invalid or inappropriate NMSetting", not just a "conflicting" one. (But fix a case in nm_connection_update_secrets() that was returning INVALID_SETTING when it should have been return-if-failing instead.) For both MISSING_SETTING and INVALID_SETTING, always prefix the error message with "setting-name: ", just like we do with the various NMSetting MISSING_PROPERTY and INVALID_PROPERTY errors. And make sure that the error message is marked for localization. Drop NM_CONNECTION_ERROR_CONNECTION_TYPE_INVALID, which is pretty pointless; it was only used in the case where connection.type was the name of a valid setting type that is not a base setting type. Instead, just return NM_SETTING_CONNECTION_ERROR_INVALID_PROPERTY for connection.type in this case (which is what the code already did when connection.type was completely unrecognized).
2014-10-11 15:02:16 -04:00
g_assert (g_str_has_prefix (error->message, "connection.type: "));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (success == FALSE);
g_object_unref (connection);
g_clear_error (&error);
/* Serial setting */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_SERIAL_SETTING_NAME);
setting = nm_setting_wired_new ();
nm_connection_add_setting (connection, setting);
setting = nm_setting_serial_new ();
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm-core: clean up NMConnectionError Rename NM_CONNECTION_ERROR_UNKNOWN to NM_CONNECTION_ERROR_FAILED, following GError best practices. Replace NM_CONNECTION_ERROR_CONNECTION_SETTING_NOT_FOUND ("no NMSettingConnection") with a more generic NM_CONNECTION_ERROR_MISSING_SETTING. Use that new code in a few places that had previously been using NM_CONNECTION_ERROR_SETTING_NOT_FOUND, which was supposed to mean "the setting that you asked about doesn't exist", not "the connection is invalid because it's missing a required setting". Clarify that NM_CONNECTION_ERROR_INVALID_SETTING can be used for any "invalid or inappropriate NMSetting", not just a "conflicting" one. (But fix a case in nm_connection_update_secrets() that was returning INVALID_SETTING when it should have been return-if-failing instead.) For both MISSING_SETTING and INVALID_SETTING, always prefix the error message with "setting-name: ", just like we do with the various NMSetting MISSING_PROPERTY and INVALID_PROPERTY errors. And make sure that the error message is marked for localization. Drop NM_CONNECTION_ERROR_CONNECTION_TYPE_INVALID, which is pretty pointless; it was only used in the case where connection.type was the name of a valid setting type that is not a base setting type. Instead, just return NM_SETTING_CONNECTION_ERROR_INVALID_PROPERTY for connection.type in this case (which is what the code already did when connection.type was completely unrecognized).
2014-10-11 15:02:16 -04:00
g_assert (g_str_has_prefix (error->message, "connection.type: "));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (success == FALSE);
g_object_unref (connection);
g_clear_error (&error);
/* IP4 setting */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_IP4_CONFIG_SETTING_NAME);
setting = nm_setting_wired_new ();
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm-core: clean up NMConnectionError Rename NM_CONNECTION_ERROR_UNKNOWN to NM_CONNECTION_ERROR_FAILED, following GError best practices. Replace NM_CONNECTION_ERROR_CONNECTION_SETTING_NOT_FOUND ("no NMSettingConnection") with a more generic NM_CONNECTION_ERROR_MISSING_SETTING. Use that new code in a few places that had previously been using NM_CONNECTION_ERROR_SETTING_NOT_FOUND, which was supposed to mean "the setting that you asked about doesn't exist", not "the connection is invalid because it's missing a required setting". Clarify that NM_CONNECTION_ERROR_INVALID_SETTING can be used for any "invalid or inappropriate NMSetting", not just a "conflicting" one. (But fix a case in nm_connection_update_secrets() that was returning INVALID_SETTING when it should have been return-if-failing instead.) For both MISSING_SETTING and INVALID_SETTING, always prefix the error message with "setting-name: ", just like we do with the various NMSetting MISSING_PROPERTY and INVALID_PROPERTY errors. And make sure that the error message is marked for localization. Drop NM_CONNECTION_ERROR_CONNECTION_TYPE_INVALID, which is pretty pointless; it was only used in the case where connection.type was the name of a valid setting type that is not a base setting type. Instead, just return NM_SETTING_CONNECTION_ERROR_INVALID_PROPERTY for connection.type in this case (which is what the code already did when connection.type was completely unrecognized).
2014-10-11 15:02:16 -04:00
g_assert (g_str_has_prefix (error->message, "connection.type: "));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (success == FALSE);
g_object_unref (connection);
g_clear_error (&error);
/* IP6 setting */
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
add_generic_settings (connection, NM_SETTING_IP6_CONFIG_SETTING_NAME);
setting = nm_setting_wired_new ();
nm_connection_add_setting (connection, setting);
success = nm_connection_verify (connection, &error);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm-core: clean up NMConnectionError Rename NM_CONNECTION_ERROR_UNKNOWN to NM_CONNECTION_ERROR_FAILED, following GError best practices. Replace NM_CONNECTION_ERROR_CONNECTION_SETTING_NOT_FOUND ("no NMSettingConnection") with a more generic NM_CONNECTION_ERROR_MISSING_SETTING. Use that new code in a few places that had previously been using NM_CONNECTION_ERROR_SETTING_NOT_FOUND, which was supposed to mean "the setting that you asked about doesn't exist", not "the connection is invalid because it's missing a required setting". Clarify that NM_CONNECTION_ERROR_INVALID_SETTING can be used for any "invalid or inappropriate NMSetting", not just a "conflicting" one. (But fix a case in nm_connection_update_secrets() that was returning INVALID_SETTING when it should have been return-if-failing instead.) For both MISSING_SETTING and INVALID_SETTING, always prefix the error message with "setting-name: ", just like we do with the various NMSetting MISSING_PROPERTY and INVALID_PROPERTY errors. And make sure that the error message is marked for localization. Drop NM_CONNECTION_ERROR_CONNECTION_TYPE_INVALID, which is pretty pointless; it was only used in the case where connection.type was the name of a valid setting type that is not a base setting type. Instead, just return NM_SETTING_CONNECTION_ERROR_INVALID_PROPERTY for connection.type in this case (which is what the code already did when connection.type was completely unrecognized).
2014-10-11 15:02:16 -04:00
g_assert (g_str_has_prefix (error->message, "connection.type: "));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (success == FALSE);
g_object_unref (connection);
g_clear_error (&error);
}
static void
test_setting_compare_id (void)
{
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSetting *old = NULL, *new = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
gboolean success;
old = nm_setting_connection_new ();
g_object_set (old,
NM_SETTING_CONNECTION_ID, "really awesome cool connection",
NM_SETTING_CONNECTION_UUID, "fbbd59d5-acab-4e30-8f86-258d272617e7",
NM_SETTING_CONNECTION_AUTOCONNECT, FALSE,
NULL);
new = nm_setting_duplicate (old);
g_object_set (new, NM_SETTING_CONNECTION_ID, "some different connection id", NULL);
/* First make sure they are different */
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (success == FALSE);
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_IGNORE_ID);
g_assert (success);
}
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
static void
test_setting_compare_addresses (void)
{
gs_unref_object NMSetting *s1 = NULL, *s2 = NULL;
gboolean success;
NMIPAddress *a;
GHashTable *result = NULL;
s1 = nm_setting_ip4_config_new ();
s2 = nm_setting_ip4_config_new ();
a = nm_ip_address_new (AF_INET, "192.168.7.5", 24, NULL);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
nm_ip_address_set_attribute (a, NM_IP_ADDRESS_ATTRIBUTE_LABEL, g_variant_new_string ("xoxoxo"));
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
nm_setting_ip_config_add_address ((NMSettingIPConfig *) s1, a);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
nm_ip_address_set_attribute (a, NM_IP_ADDRESS_ATTRIBUTE_LABEL, g_variant_new_string ("hello"));
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
nm_setting_ip_config_add_address ((NMSettingIPConfig *) s2, a);
nm_ip_address_unref (a);
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
if (nmtst_get_rand_uint32 () % 2)
tests: replace NMTST_SWAP() by new NM_SWAP() macro NMTST_SWAP() used memcpy() for copying the value, while NM_SWAP() uses a temporary variable with typeof(). I think the latter is preferable. Also, the macro is essentially doing the same thing.
2020-03-23 08:04:33 +01:00
NM_SWAP (s1, s2);
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
success = nm_setting_compare (s1, s2, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (!success);
success = nm_setting_diff (s1, s2, NM_SETTING_COMPARE_FLAG_EXACT, FALSE, &result);
g_assert (!success);
all: use nm_clear_pointer() instead of g_clear_pointer() g_clear_pointer() would always cast the destroy notify function pointer to GDestroyNotify. That means, it lost some type safety, like GPtrArray *ptr_arr = ... g_clear_pointer (&ptr_arr, g_array_unref); Since glib 2.58 ([1]), g_clear_pointer() is also more type safe. But this is not used by NetworkManager, because we don't set GLIB_VERSION_MIN_REQUIRED to 2.58. [1] https://gitlab.gnome.org/GNOME/glib/-/commit/f9a9902aac826ab4aecc25f6eb533a418a4fa559 We have nm_clear_pointer() to avoid this issue for a long time (pre 1.12.0). Possibly we should redefine in our source tree g_clear_pointer() as nm_clear_pointer(). However, I don't like to patch glib functions with our own variant. Arguably, we do patch g_clear_error() in such a manner. But there the point is to make the function inlinable. Also, nm_clear_pointer() returns a boolean that indicates whether anything was cleared. That is sometimes useful. I think we should just consistently use nm_clear_pointer() instead, which does always the preferable thing. Replace: sed 's/\<g_clear_pointer *(\([^;]*\), *\([a-z_A-Z0-9]\+\) *)/nm_clear_pointer (\1, \2)/g' $(git grep -l g_clear_pointer) -i
2020-03-23 11:09:24 +01:00
nm_clear_pointer (&result, g_hash_table_unref);
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
}
static void
test_setting_compare_routes (void)
{
gs_unref_object NMSetting *s1 = NULL, *s2 = NULL;
gboolean success;
NMIPRoute *r;
GHashTable *result = NULL;
s1 = nm_setting_ip4_config_new ();
s2 = nm_setting_ip4_config_new ();
r = nm_ip_route_new (AF_INET, "192.168.12.0", 24, "192.168.11.1", 473, NULL);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
nm_ip_route_set_attribute (r, NM_IP_ADDRESS_ATTRIBUTE_LABEL, g_variant_new_string ("xoxoxo"));
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
nm_setting_ip_config_add_route ((NMSettingIPConfig *) s1, r);
libnm: add NM_IP_ADDRESS_ATTRIBUTE_LABEL define There is only one supported attribute for addresses. The "lable". Give it a #define.
2017-12-11 14:06:10 +01:00
nm_ip_route_set_attribute (r, NM_IP_ADDRESS_ATTRIBUTE_LABEL, g_variant_new_string ("hello"));
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
nm_setting_ip_config_add_route ((NMSettingIPConfig *) s2, r);
nm_ip_route_unref (r);
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
if (nmtst_get_rand_uint32 () % 2)
tests: replace NMTST_SWAP() by new NM_SWAP() macro NMTST_SWAP() used memcpy() for copying the value, while NM_SWAP() uses a temporary variable with typeof(). I think the latter is preferable. Also, the macro is essentially doing the same thing.
2020-03-23 08:04:33 +01:00
NM_SWAP (s1, s2);
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
success = nm_setting_compare (s1, s2, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (!success);
success = nm_setting_diff (s1, s2, NM_SETTING_COMPARE_FLAG_EXACT, FALSE, &result);
g_assert (!success);
all: use nm_clear_pointer() instead of g_clear_pointer() g_clear_pointer() would always cast the destroy notify function pointer to GDestroyNotify. That means, it lost some type safety, like GPtrArray *ptr_arr = ... g_clear_pointer (&ptr_arr, g_array_unref); Since glib 2.58 ([1]), g_clear_pointer() is also more type safe. But this is not used by NetworkManager, because we don't set GLIB_VERSION_MIN_REQUIRED to 2.58. [1] https://gitlab.gnome.org/GNOME/glib/-/commit/f9a9902aac826ab4aecc25f6eb533a418a4fa559 We have nm_clear_pointer() to avoid this issue for a long time (pre 1.12.0). Possibly we should redefine in our source tree g_clear_pointer() as nm_clear_pointer(). However, I don't like to patch glib functions with our own variant. Arguably, we do patch g_clear_error() in such a manner. But there the point is to make the function inlinable. Also, nm_clear_pointer() returns a boolean that indicates whether anything was cleared. That is sometimes useful. I think we should just consistently use nm_clear_pointer() instead, which does always the preferable thing. Replace: sed 's/\<g_clear_pointer *(\([^;]*\), *\([a-z_A-Z0-9]\+\) *)/nm_clear_pointer (\1, \2)/g' $(git grep -l g_clear_pointer) -i
2020-03-23 11:09:24 +01:00
nm_clear_pointer (&result, g_hash_table_unref);
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
}
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
static void
test_setting_compare_wired_cloned_mac_address (void)
{
gs_unref_object NMSetting *old = NULL, *new = NULL;
gboolean success;
gs_free char *str1 = NULL;
old = nm_setting_wired_new ();
g_object_set (old,
NM_SETTING_WIRED_CLONED_MAC_ADDRESS, "stable",
NULL);
g_assert_cmpstr ("stable", ==, nm_setting_wired_get_cloned_mac_address ((NMSettingWired *) old));
g_object_get (old, NM_SETTING_WIRED_CLONED_MAC_ADDRESS, &str1, NULL);
g_assert_cmpstr ("stable", ==, str1);
all: use nm_clear_g_free() instead of g_clear_pointer() I think it's preferable to use nm_clear_g_free() instead of g_clear_pointer(, g_free). The reasons are not very strong, but I think it is overall preferable to have a shorthand for this frequently used functionality. sed 's/\<g_clear_pointer *(\([^;]*\), *\(g_free\) *)/nm_clear_g_free (\1)/g' $(git grep -l g_clear_pointer) -i
2020-03-23 11:00:43 +01:00
nm_clear_g_free (&str1);
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
new = nm_setting_duplicate (old);
g_object_set (new, NM_SETTING_WIRED_CLONED_MAC_ADDRESS, "11:22:33:44:55:66", NULL);
g_assert_cmpstr ("11:22:33:44:55:66", ==, nm_setting_wired_get_cloned_mac_address ((NMSettingWired *) new));
g_object_get (new, NM_SETTING_WIRED_CLONED_MAC_ADDRESS, &str1, NULL);
g_assert_cmpstr ("11:22:33:44:55:66", ==, str1);
all: use nm_clear_g_free() instead of g_clear_pointer() I think it's preferable to use nm_clear_g_free() instead of g_clear_pointer(, g_free). The reasons are not very strong, but I think it is overall preferable to have a shorthand for this frequently used functionality. sed 's/\<g_clear_pointer *(\([^;]*\), *\(g_free\) *)/nm_clear_g_free (\1)/g' $(git grep -l g_clear_pointer) -i
2020-03-23 11:00:43 +01:00
nm_clear_g_free (&str1);
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (!success);
g_clear_object (&new);
new = nm_setting_duplicate (old);
g_object_set (new, NM_SETTING_WIRED_CLONED_MAC_ADDRESS, "stable-bia", NULL);
g_assert_cmpstr ("stable-bia", ==, nm_setting_wired_get_cloned_mac_address ((NMSettingWired *) new));
g_object_get (new, NM_SETTING_WIRED_CLONED_MAC_ADDRESS, &str1, NULL);
g_assert_cmpstr ("stable-bia", ==, str1);
all: use nm_clear_g_free() instead of g_clear_pointer() I think it's preferable to use nm_clear_g_free() instead of g_clear_pointer(, g_free). The reasons are not very strong, but I think it is overall preferable to have a shorthand for this frequently used functionality. sed 's/\<g_clear_pointer *(\([^;]*\), *\(g_free\) *)/nm_clear_g_free (\1)/g' $(git grep -l g_clear_pointer) -i
2020-03-23 11:00:43 +01:00
nm_clear_g_free (&str1);
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (!success);
g_clear_object (&new);
}
static void
test_setting_compare_wireless_cloned_mac_address (void)
{
gs_unref_object NMSetting *old = NULL, *new = NULL;
gboolean success;
gs_free char *str1 = NULL;
old = nm_setting_wireless_new ();
g_object_set (old,
NM_SETTING_WIRELESS_CLONED_MAC_ADDRESS, "stable",
NULL);
g_assert_cmpstr ("stable", ==, nm_setting_wireless_get_cloned_mac_address ((NMSettingWireless *) old));
g_object_get (old, NM_SETTING_WIRELESS_CLONED_MAC_ADDRESS, &str1, NULL);
g_assert_cmpstr ("stable", ==, str1);
all: use nm_clear_g_free() instead of g_clear_pointer() I think it's preferable to use nm_clear_g_free() instead of g_clear_pointer(, g_free). The reasons are not very strong, but I think it is overall preferable to have a shorthand for this frequently used functionality. sed 's/\<g_clear_pointer *(\([^;]*\), *\(g_free\) *)/nm_clear_g_free (\1)/g' $(git grep -l g_clear_pointer) -i
2020-03-23 11:00:43 +01:00
nm_clear_g_free (&str1);
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
new = nm_setting_duplicate (old);
g_object_set (new, NM_SETTING_WIRELESS_CLONED_MAC_ADDRESS, "11:22:33:44:55:66", NULL);
g_assert_cmpstr ("11:22:33:44:55:66", ==, nm_setting_wireless_get_cloned_mac_address ((NMSettingWireless *) new));
g_object_get (new, NM_SETTING_WIRELESS_CLONED_MAC_ADDRESS, &str1, NULL);
g_assert_cmpstr ("11:22:33:44:55:66", ==, str1);
all: use nm_clear_g_free() instead of g_clear_pointer() I think it's preferable to use nm_clear_g_free() instead of g_clear_pointer(, g_free). The reasons are not very strong, but I think it is overall preferable to have a shorthand for this frequently used functionality. sed 's/\<g_clear_pointer *(\([^;]*\), *\(g_free\) *)/nm_clear_g_free (\1)/g' $(git grep -l g_clear_pointer) -i
2020-03-23 11:00:43 +01:00
nm_clear_g_free (&str1);
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (!success);
g_clear_object (&new);
new = nm_setting_duplicate (old);
g_object_set (new, NM_SETTING_WIRELESS_CLONED_MAC_ADDRESS, "stable-bia", NULL);
g_assert_cmpstr ("stable-bia", ==, nm_setting_wireless_get_cloned_mac_address ((NMSettingWireless *) new));
g_object_get (new, NM_SETTING_WIRELESS_CLONED_MAC_ADDRESS, &str1, NULL);
g_assert_cmpstr ("stable-bia", ==, str1);
shared: add nm_clear_g_free()
2016-09-28 15:34:52 +02:00
nm_clear_g_free (&str1);
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (!success);
g_clear_object (&new);
}
libnm-core: add NM_SETTING_COMPARE_FLAG_IGNORE_TIMESTAMP flag for ignoring timestamp when comparing settings or connections.
2014-11-18 14:20:12 +01:00
static void
test_setting_compare_timestamp (void)
{
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSetting *old = NULL, *new = NULL;
libnm-core: add NM_SETTING_COMPARE_FLAG_IGNORE_TIMESTAMP flag for ignoring timestamp when comparing settings or connections.
2014-11-18 14:20:12 +01:00
gboolean success;
old = nm_setting_connection_new ();
g_object_set (old,
NM_SETTING_CONNECTION_ID, "ignore timestamp connection",
NM_SETTING_CONNECTION_UUID, "b047a198-0e0a-4f0e-a653-eea09bb35e40",
NM_SETTING_CONNECTION_AUTOCONNECT, FALSE,
libnm-core: Add missing type cast Fixes testing on 32-bit arches: /core/general/test_setting_compare_timestamp: (./test-general:29331): GLib-GObject-WARNING **: g_object_set_valist: object class `NMSettingConnection' has no property named `$?\xff\xff\x89t$0\x89|$4\xe8\u001c\x98\xff\xff\x85\xc0tM\x83\xf8\xfft3\x8dT$(\xc7D$\u0008' /bin/sh: line 5: 29331 Trace/breakpoint trap ${dir}$tst FAIL: test-general Fixes: 093a3c88d03bf91ed65e80e45261451a50e974df
2014-11-19 16:01:43 +01:00
NM_SETTING_CONNECTION_TIMESTAMP, (guint64) 1234567890,
libnm-core: add NM_SETTING_COMPARE_FLAG_IGNORE_TIMESTAMP flag for ignoring timestamp when comparing settings or connections.
2014-11-18 14:20:12 +01:00
NULL);
new = nm_setting_duplicate (old);
libnm-core: Add missing type cast Fixes testing on 32-bit arches: /core/general/test_setting_compare_timestamp: (./test-general:29331): GLib-GObject-WARNING **: g_object_set_valist: object class `NMSettingConnection' has no property named `$?\xff\xff\x89t$0\x89|$4\xe8\u001c\x98\xff\xff\x85\xc0tM\x83\xf8\xfft3\x8dT$(\xc7D$\u0008' /bin/sh: line 5: 29331 Trace/breakpoint trap ${dir}$tst FAIL: test-general Fixes: 093a3c88d03bf91ed65e80e45261451a50e974df
2014-11-19 16:01:43 +01:00
g_object_set (new, NM_SETTING_CONNECTION_TIMESTAMP, (guint64) 1416316539, NULL);
libnm-core: add NM_SETTING_COMPARE_FLAG_IGNORE_TIMESTAMP flag for ignoring timestamp when comparing settings or connections.
2014-11-18 14:20:12 +01:00
/* First make sure they are different */
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (success == FALSE);
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_IGNORE_TIMESTAMP);
g_assert (success);
}
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
typedef struct {
NMSettingSecretFlags secret_flags;
NMSettingCompareFlags comp_flags;
gboolean remove_secret;
} TestDataCompareSecrets;
static TestDataCompareSecrets *
test_data_compare_secrets_new (NMSettingSecretFlags secret_flags,
NMSettingCompareFlags comp_flags,
gboolean remove_secret)
{
TestDataCompareSecrets *data = g_new0 (TestDataCompareSecrets, 1);
data->secret_flags = secret_flags;
data->comp_flags = comp_flags;
data->remove_secret = remove_secret;
return data;
}
libnm/tests: add tests for comparing settings with different secret-flags The flags NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS and NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS act on the secret flags to decide whether to ignore a secret. But there was not test how this behaved, if the two settings had differing flags.
2019-01-09 14:41:55 +01:00
static void
_test_compare_secrets_check_diff (NMSetting *a,
NMSetting *b,
NMSettingCompareFlags flags,
gboolean exp_same_psk,
gboolean exp_same_psk_flags)
{
gs_unref_hashtable GHashTable *h = NULL;
NMSettingDiffResult _RESULT_IN_A = NM_SETTING_DIFF_RESULT_IN_A;
NMSettingDiffResult _RESULT_IN_B = NM_SETTING_DIFF_RESULT_IN_B;
gboolean invert_results;
gboolean diff_result;
NMSettingSecretFlags a_psk_flags = nm_setting_wireless_security_get_psk_flags (NM_SETTING_WIRELESS_SECURITY (a));
NMSettingSecretFlags b_psk_flags = nm_setting_wireless_security_get_psk_flags (NM_SETTING_WIRELESS_SECURITY (b));
const char *a_psk = nm_setting_wireless_security_get_psk (NM_SETTING_WIRELESS_SECURITY (a));
const char *b_psk = nm_setting_wireless_security_get_psk (NM_SETTING_WIRELESS_SECURITY (b));
g_assert (NM_IS_SETTING_WIRELESS_SECURITY (a));
g_assert (NM_IS_SETTING_WIRELESS_SECURITY (b));
invert_results = nmtst_get_rand_bool ();
if (invert_results) {
_RESULT_IN_A = NM_SETTING_DIFF_RESULT_IN_B;
_RESULT_IN_B = NM_SETTING_DIFF_RESULT_IN_A;
}
diff_result = nm_setting_diff (a, b, flags, invert_results, &h);
g_assert (exp_same_psk_flags == (a_psk_flags == b_psk_flags));
if (nm_streq0 (a_psk, b_psk))
g_assert (exp_same_psk);
else {
if (flags == NM_SETTING_COMPARE_FLAG_EXACT)
g_assert (!exp_same_psk);
else if (flags == NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS) {
if ( !NM_FLAGS_HAS (a_psk_flags, NM_SETTING_SECRET_FLAG_AGENT_OWNED)
&& !NM_FLAGS_HAS (b_psk_flags, NM_SETTING_SECRET_FLAG_AGENT_OWNED))
g_assert (!exp_same_psk);
else if ( !NM_FLAGS_HAS (a_psk_flags, NM_SETTING_SECRET_FLAG_AGENT_OWNED)
&& NM_FLAGS_HAS (b_psk_flags, NM_SETTING_SECRET_FLAG_AGENT_OWNED))
g_assert (!exp_same_psk);
else
g_assert (exp_same_psk);
} else if (flags == NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS) {
if ( !NM_FLAGS_HAS (a_psk_flags, NM_SETTING_SECRET_FLAG_NOT_SAVED)
&& !NM_FLAGS_HAS (b_psk_flags, NM_SETTING_SECRET_FLAG_NOT_SAVED))
g_assert (!exp_same_psk);
else if ( !NM_FLAGS_HAS (a_psk_flags, NM_SETTING_SECRET_FLAG_NOT_SAVED)
&& NM_FLAGS_HAS (b_psk_flags, NM_SETTING_SECRET_FLAG_NOT_SAVED))
g_assert (!exp_same_psk);
else
g_assert (exp_same_psk);
} else if (flags == NM_SETTING_COMPARE_FLAG_IGNORE_SECRETS)
g_assert (exp_same_psk);
else
g_assert_not_reached ();
}
g_assert (diff_result == (exp_same_psk && exp_same_psk_flags));
g_assert (diff_result == (!h));
if (!diff_result) {
if (flags == NM_SETTING_COMPARE_FLAG_EXACT)
g_assert (!exp_same_psk);
else if ( NM_IN_SET (flags, NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS,
NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS)
&& (a_psk_flags != b_psk_flags)
&& nm_setting_wireless_security_get_psk_flags (NM_SETTING_WIRELESS_SECURITY (a)) == NM_SETTING_SECRET_FLAG_NONE)
g_assert (!exp_same_psk);
else
g_assert (exp_same_psk);
g_assert ((!exp_same_psk) == g_hash_table_contains (h, NM_SETTING_WIRELESS_SECURITY_PSK));
if (!exp_same_psk) {
if (nm_setting_wireless_security_get_psk (NM_SETTING_WIRELESS_SECURITY (a)))
g_assert_cmpint (GPOINTER_TO_UINT (g_hash_table_lookup (h, NM_SETTING_WIRELESS_SECURITY_PSK)), ==, _RESULT_IN_A);
else
g_assert_cmpint (GPOINTER_TO_UINT (g_hash_table_lookup (h, NM_SETTING_WIRELESS_SECURITY_PSK)), ==, _RESULT_IN_B);
}
g_assert ((!exp_same_psk_flags) == g_hash_table_contains (h, NM_SETTING_WIRELESS_SECURITY_PSK_FLAGS));
if (!exp_same_psk_flags) {
if (nm_setting_wireless_security_get_psk_flags (NM_SETTING_WIRELESS_SECURITY (a)) != NM_SETTING_SECRET_FLAG_NONE)
g_assert_cmpint (GPOINTER_TO_UINT (g_hash_table_lookup (h, NM_SETTING_WIRELESS_SECURITY_PSK_FLAGS)), ==, _RESULT_IN_A);
else
g_assert_cmpint (GPOINTER_TO_UINT (g_hash_table_lookup (h, NM_SETTING_WIRELESS_SECURITY_PSK_FLAGS)), ==, _RESULT_IN_B);
}
g_assert_cmpint (g_hash_table_size (h), ==, (!exp_same_psk) + (!exp_same_psk_flags));
}
g_assert (diff_result == nm_setting_compare (a, b, flags));
g_assert (diff_result == nm_setting_compare (b, a, flags));
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
test_setting_compare_secrets (gconstpointer test_data)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
const TestDataCompareSecrets *data = test_data;
libnm/tests: add tests for comparing settings with different secret-flags The flags NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS and NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS act on the secret flags to decide whether to ignore a secret. But there was not test how this behaved, if the two settings had differing flags.
2019-01-09 14:41:55 +01:00
gs_unref_object NMConnection *conn_old = NULL;
gs_unref_object NMConnection *conn_new = NULL;
gs_unref_object NMSetting *old = NULL;
gs_unref_object NMSetting *new = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Make sure that a connection with transient/unsaved secrets compares
* successfully to the same connection without those secrets.
*/
old = nm_setting_wireless_security_new ();
g_object_set (old,
NM_SETTING_WIRELESS_SECURITY_KEY_MGMT, "wpa-psk",
NM_SETTING_WIRELESS_SECURITY_PSK, "really cool psk",
NULL);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
nm_setting_set_secret_flags (old, NM_SETTING_WIRELESS_SECURITY_PSK, data->secret_flags, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
new = nm_setting_duplicate (old);
libnm/tests: add tests for comparing settings with different secret-flags The flags NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS and NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS act on the secret flags to decide whether to ignore a secret. But there was not test how this behaved, if the two settings had differing flags.
2019-01-09 14:41:55 +01:00
if (data->remove_secret)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (new, NM_SETTING_WIRELESS_SECURITY_PSK, NULL, NULL);
libnm/tests: add tests for comparing settings with different secret-flags The flags NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS and NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS act on the secret flags to decide whether to ignore a secret. But there was not test how this behaved, if the two settings had differing flags.
2019-01-09 14:41:55 +01:00
g_assert ((!data->remove_secret) == nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT));
g_assert ((!data->remove_secret) == nm_setting_compare (new, old, NM_SETTING_COMPARE_FLAG_EXACT));
_test_compare_secrets_check_diff (old, new, NM_SETTING_COMPARE_FLAG_EXACT, !data->remove_secret, TRUE);
_test_compare_secrets_check_diff (new, old, NM_SETTING_COMPARE_FLAG_EXACT, !data->remove_secret, TRUE);
g_assert (nm_setting_compare (old, new, data->comp_flags));
g_assert (nm_setting_compare (new, old, data->comp_flags));
_test_compare_secrets_check_diff (old, new, data->comp_flags, TRUE, TRUE);
_test_compare_secrets_check_diff (new, old, data->comp_flags, TRUE, TRUE);
/* OK. Try again, but this time not only change the secret, also let the secret flags differ... */
if (data->secret_flags != NM_SETTING_SECRET_FLAG_NONE) {
nm_setting_set_secret_flags (new, NM_SETTING_WIRELESS_SECURITY_PSK, NM_SETTING_SECRET_FLAG_NONE, NULL);
_test_compare_secrets_check_diff (old, new, NM_SETTING_COMPARE_FLAG_EXACT, FALSE, FALSE);
_test_compare_secrets_check_diff (new, old, NM_SETTING_COMPARE_FLAG_EXACT, FALSE, FALSE);
_test_compare_secrets_check_diff (old, new, data->comp_flags, TRUE, FALSE);
_test_compare_secrets_check_diff (new, old, data->comp_flags, FALSE, FALSE);
nm_setting_set_secret_flags (new, NM_SETTING_WIRELESS_SECURITY_PSK, data->secret_flags, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
libnm/tests: add tests for comparing settings with different secret-flags The flags NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS and NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS act on the secret flags to decide whether to ignore a secret. But there was not test how this behaved, if the two settings had differing flags.
2019-01-09 14:41:55 +01:00
conn_old = nmtst_create_minimal_connection ("test-compare-secrets", NULL, NM_SETTING_WIRELESS_SETTING_NAME, NULL);
nm_connection_add_setting (conn_old, nm_setting_duplicate (old));
conn_new = nm_simple_connection_new_clone (conn_old);
nm_connection_add_setting (conn_new, nm_setting_duplicate (new));
g_assert ((!data->remove_secret) == nm_connection_compare (conn_old, conn_new, NM_SETTING_COMPARE_FLAG_EXACT));
g_assert ((!data->remove_secret) == nm_connection_compare (conn_new, conn_old, NM_SETTING_COMPARE_FLAG_EXACT));
g_assert (nm_connection_compare (conn_old, conn_new, data->comp_flags));
g_assert (nm_connection_compare (conn_new, conn_old, data->comp_flags));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
static void
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
test_setting_compare_vpn_secrets (gconstpointer test_data)
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
{
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
const TestDataCompareSecrets *data = test_data;
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSetting *old = NULL, *new = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
gboolean success;
/* Make sure that a connection with transient/unsaved secrets compares
* successfully to the same connection without those secrets.
*/
old = nm_setting_vpn_new ();
nm_setting_vpn_add_secret (NM_SETTING_VPN (old), "foobarbaz", "really secret password");
nm_setting_vpn_add_secret (NM_SETTING_VPN (old), "asdfasdfasdf", "really adfasdfasdfasdf");
nm_setting_vpn_add_secret (NM_SETTING_VPN (old), "0123456778", "abcdefghijklmnpqrstuvqxyz");
nm_setting_vpn_add_secret (NM_SETTING_VPN (old), "borkbork", "yet another really secret password");
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
nm_setting_set_secret_flags (old, "borkbork", data->secret_flags, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* Clear "borkbork" from the duplicated setting */
new = nm_setting_duplicate (old);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
if (data->remove_secret) {
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_setting_vpn_remove_secret (NM_SETTING_VPN (new), "borkbork");
/* First make sure they are different */
success = nm_setting_compare (old, new, NM_SETTING_COMPARE_FLAG_EXACT);
g_assert (success == FALSE);
}
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
success = nm_setting_compare (old, new, data->comp_flags);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (success);
}
static void
test_hwaddr_aton_ether_normal (void)
{
guint8 buf[100];
guint8 expected[ETH_ALEN] = { 0x00, 0x11, 0x22, 0x33, 0x44, 0x55 };
libnm-core, etc: drop type-based hwaddr funcs, port to length-based ones Drop the arptype-based nm_utils_hwaddr funcs, and rename the length-based ones to no longer have _len in their names. This also switches nm_utils_hwaddr_atoba() to using a length rather than an arptype, and adds a length argument to nm_utils_hwaddr_valid() (making nm_utils_hwaddr_valid() now a replacement for nm_utils_hwaddr_aton() in some places, where we were only using aton() to do validity checking).
2014-07-04 15:59:19 -04:00
g_assert (nm_utils_hwaddr_aton ("00:11:22:33:44:55", buf, ETH_ALEN) != NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (memcmp (buf, expected, sizeof (expected)) == 0);
}
static void
test_hwaddr_aton_ib_normal (void)
{
guint8 buf[100];
const char *source = "00:11:22:33:44:55:66:77:88:99:01:12:23:34:45:56:67:78:89:90";
guint8 expected[INFINIBAND_ALEN] = { 0x00, 0x11, 0x22, 0x33, 0x44, 0x55, 0x66,
0x77, 0x88, 0x99, 0x01, 0x12, 0x23, 0x34, 0x45, 0x56, 0x67, 0x78, 0x89,
0x90 };
libnm-core, etc: drop type-based hwaddr funcs, port to length-based ones Drop the arptype-based nm_utils_hwaddr funcs, and rename the length-based ones to no longer have _len in their names. This also switches nm_utils_hwaddr_atoba() to using a length rather than an arptype, and adds a length argument to nm_utils_hwaddr_valid() (making nm_utils_hwaddr_valid() now a replacement for nm_utils_hwaddr_aton() in some places, where we were only using aton() to do validity checking).
2014-07-04 15:59:19 -04:00
g_assert (nm_utils_hwaddr_aton (source, buf, INFINIBAND_ALEN) != NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (memcmp (buf, expected, sizeof (expected)) == 0);
}
static void
test_hwaddr_aton_no_leading_zeros (void)
{
guint8 buf[100];
guint8 expected[ETH_ALEN] = { 0x00, 0x1A, 0x2B, 0x03, 0x44, 0x05 };
libnm-core, etc: drop type-based hwaddr funcs, port to length-based ones Drop the arptype-based nm_utils_hwaddr funcs, and rename the length-based ones to no longer have _len in their names. This also switches nm_utils_hwaddr_atoba() to using a length rather than an arptype, and adds a length argument to nm_utils_hwaddr_valid() (making nm_utils_hwaddr_valid() now a replacement for nm_utils_hwaddr_aton() in some places, where we were only using aton() to do validity checking).
2014-07-04 15:59:19 -04:00
g_assert (nm_utils_hwaddr_aton ("0:1a:2B:3:44:5", buf, ETH_ALEN) != NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert (memcmp (buf, expected, sizeof (expected)) == 0);
}
static void
test_hwaddr_aton_malformed (void)
{
guint8 buf[100];
libnm-core, etc: drop type-based hwaddr funcs, port to length-based ones Drop the arptype-based nm_utils_hwaddr funcs, and rename the length-based ones to no longer have _len in their names. This also switches nm_utils_hwaddr_atoba() to using a length rather than an arptype, and adds a length argument to nm_utils_hwaddr_valid() (making nm_utils_hwaddr_valid() now a replacement for nm_utils_hwaddr_aton() in some places, where we were only using aton() to do validity checking).
2014-07-04 15:59:19 -04:00
g_assert (nm_utils_hwaddr_aton ("0:1a:2B:3:a@%%", buf, ETH_ALEN) == NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
libnm-core, etc: add nm_utils_hwaddr_matches() Add nm_utils_hwaddr_matches(), for comparing hardware addresses for equality, allowing either binary or ASCII hardware addresses to be passed, and handling the special rules for InfiniBand hardware addresses automatically. Update code to use it.
2014-06-16 11:30:47 -04:00
static void
test_hwaddr_equal (void)
{
const char *string = "00:1a:2b:03:44:05";
const char *upper_string = "00:1A:2B:03:44:05";
const char *bad_string = "0:1a:2b:3:44:5";
const guint8 binary[ETH_ALEN] = { 0x00, 0x1A, 0x2B, 0x03, 0x44, 0x05 };
const char *other_string = "1a:2b:03:44:05:00";
const guint8 other_binary[ETH_ALEN] = { 0x1A, 0x2B, 0x03, 0x44, 0x05, 0x00 };
const char *long_string = "00:1a:2b:03:44:05:06:07";
const guint8 long_binary[8] = { 0x00, 0x1A, 0x2B, 0x03, 0x44, 0x05, 0x06, 0x07 };
const char *null_string = "00:00:00:00:00:00";
const guint8 null_binary[ETH_ALEN] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 };
g_assert (nm_utils_hwaddr_matches (string, -1, string, -1));
g_assert (nm_utils_hwaddr_matches (string, -1, upper_string, -1));
g_assert (nm_utils_hwaddr_matches (string, -1, bad_string, -1));
g_assert (nm_utils_hwaddr_matches (string, -1, binary, sizeof (binary)));
g_assert (!nm_utils_hwaddr_matches (string, -1, other_string, -1));
g_assert (!nm_utils_hwaddr_matches (string, -1, other_binary, sizeof (other_binary)));
g_assert (!nm_utils_hwaddr_matches (string, -1, long_string, -1));
g_assert (!nm_utils_hwaddr_matches (string, -1, long_binary, sizeof (long_binary)));
g_assert (!nm_utils_hwaddr_matches (string, -1, null_string, -1));
g_assert (!nm_utils_hwaddr_matches (string, -1, null_binary, sizeof (null_binary)));
g_assert (!nm_utils_hwaddr_matches (string, -1, NULL, ETH_ALEN));
g_assert (nm_utils_hwaddr_matches (binary, sizeof (binary), string, -1));
g_assert (nm_utils_hwaddr_matches (binary, sizeof (binary), upper_string, -1));
g_assert (nm_utils_hwaddr_matches (binary, sizeof (binary), bad_string, -1));
g_assert (nm_utils_hwaddr_matches (binary, sizeof (binary), binary, sizeof (binary)));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), other_string, -1));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), other_binary, sizeof (other_binary)));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), long_string, -1));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), long_binary, sizeof (long_binary)));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), null_string, -1));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), null_binary, sizeof (null_binary)));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), NULL, ETH_ALEN));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, string, -1));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, upper_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, bad_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, binary, sizeof (binary)));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, other_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, other_binary, sizeof (other_binary)));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, long_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, long_binary, sizeof (long_binary)));
g_assert (nm_utils_hwaddr_matches (null_string, -1, null_string, -1));
g_assert (nm_utils_hwaddr_matches (null_string, -1, null_binary, sizeof (null_binary)));
g_assert (nm_utils_hwaddr_matches (null_string, -1, NULL, ETH_ALEN));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), string, -1));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), upper_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), bad_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), binary, sizeof (binary)));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), other_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), other_binary, sizeof (other_binary)));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), long_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), long_binary, sizeof (long_binary)));
g_assert (nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), null_string, -1));
g_assert (nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), null_binary, sizeof (null_binary)));
g_assert (nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), NULL, ETH_ALEN));
utils: make nm_utils_hwaddr_matches() accept NULL This essentially aligns the implementation with the documentation. It is also rather useful, since it allows us to use the value returned by nm_setting_wired_get_mac_address() directly, and that one can indeed be NULL.
2019-11-14 15:08:31 +01:00
libnm: don't compare invalid mac addresses as equal in nm_utils_hwaddr_matches() By passing as length of the MAC addresses -1 for both arguments, one could get through to compare empty strings, NULL, and addresses longer than the maximum. Such addresses are not valid, and they should never compare equal (not even to themselves). This is a change in behavior of public API, but it never made sense to claim two addresses are equal, when they are not even valid addresses. Also, avoid undefined behavior with "NULL, -1, NULL, -1" arguments, where we would call memcmp() with zero length and NULL arguments. UBSan flags that too.
2020-05-13 22:48:34 +02:00
g_assert (!nm_utils_hwaddr_matches (NULL, -1, NULL, -1));
utils: make nm_utils_hwaddr_matches() accept NULL This essentially aligns the implementation with the documentation. It is also rather useful, since it allows us to use the value returned by nm_setting_wired_get_mac_address() directly, and that one can indeed be NULL.
2019-11-14 15:08:31 +01:00
g_assert (!nm_utils_hwaddr_matches (NULL, -1, string, -1));
g_assert (!nm_utils_hwaddr_matches (string, -1, NULL, -1));
g_assert (!nm_utils_hwaddr_matches (NULL, -1, null_string, -1));
g_assert (!nm_utils_hwaddr_matches (null_string, -1, NULL, -1));
g_assert (!nm_utils_hwaddr_matches (NULL, -1, binary, sizeof (binary)));
g_assert (!nm_utils_hwaddr_matches (binary, sizeof (binary), NULL, -1));
g_assert (!nm_utils_hwaddr_matches (NULL, -1, null_binary, sizeof (null_binary)));
g_assert (!nm_utils_hwaddr_matches (null_binary, sizeof (null_binary), NULL, -1));
libnm-core, etc: add nm_utils_hwaddr_matches() Add nm_utils_hwaddr_matches(), for comparing hardware addresses for equality, allowing either binary or ASCII hardware addresses to be passed, and handling the special rules for InfiniBand hardware addresses automatically. Update code to use it.
2014-06-16 11:30:47 -04:00
}
libnm-core: add nm_utils_hwaddr_canonical() Add a helper function for converting hardware addresses to canonical form.
2014-10-28 08:32:25 -04:00
static void
test_hwaddr_canonical (void)
{
const char *string = "00:1A:2B:03:44:05";
const char *lower_string = "00:1a:2b:03:44:05";
const char *short_string = "0:1a:2b:3:44:5";
libnm-core: canonicalize hardware addresses in settings Convert hardware addresses to canonical form (uppercase, leading zeros, colons) when setting them on/adding them to NMSetting properties.
2014-10-28 08:56:07 -04:00
const char *hyphen_string = "00-1a-2b-03-44-05";
libnm-core: add nm_utils_hwaddr_canonical() Add a helper function for converting hardware addresses to canonical form.
2014-10-28 08:32:25 -04:00
const char *invalid_string = "00:1A:2B";
char *canonical;
canonical = nm_utils_hwaddr_canonical (string, ETH_ALEN);
g_assert_cmpstr (canonical, ==, string);
g_free (canonical);
canonical = nm_utils_hwaddr_canonical (lower_string, ETH_ALEN);
g_assert_cmpstr (canonical, ==, string);
g_free (canonical);
canonical = nm_utils_hwaddr_canonical (short_string, ETH_ALEN);
g_assert_cmpstr (canonical, ==, string);
g_free (canonical);
libnm-core: canonicalize hardware addresses in settings Convert hardware addresses to canonical form (uppercase, leading zeros, colons) when setting them on/adding them to NMSetting properties.
2014-10-28 08:56:07 -04:00
canonical = nm_utils_hwaddr_canonical (hyphen_string, ETH_ALEN);
g_assert_cmpstr (canonical, ==, string);
g_free (canonical);
libnm-core: add nm_utils_hwaddr_canonical() Add a helper function for converting hardware addresses to canonical form.
2014-10-28 08:32:25 -04:00
canonical = nm_utils_hwaddr_canonical (invalid_string, ETH_ALEN);
g_assert_cmpstr (canonical, ==, NULL);
canonical = nm_utils_hwaddr_canonical (invalid_string, -1);
g_assert_cmpstr (canonical, ==, invalid_string);
g_free (canonical);
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
test_connection_changed_cb (NMConnection *connection, gboolean *data)
{
*data = TRUE;
}
core: implement nm_utils_ip4_netmask_to_prefix() via __builtin_ctz() Taken from systemd's in4_addr_netmask_to_prefixlen(). Yes, this adds the requirement that "int" is 32 bits. But systemd already has the same requirement in u32ctz(), hence we anyway cannot build on other architectures. If that is ever necessary, it's easy to adjust.
2018-12-14 16:45:53 +01:00
static guint32
_netmask_to_prefix (guint32 netmask)
{
guint32 prefix;
guint8 v;
const guint8 *p = (guint8 *) &netmask;
if (p[3]) {
prefix = 24;
v = p[3];
} else if (p[2]) {
prefix = 16;
v = p[2];
} else if (p[1]) {
prefix = 8;
v = p[1];
} else {
prefix = 0;
v = p[0];
}
while (v) {
prefix++;
v <<= 1;
}
g_assert_cmpint (prefix, <=, 32);
/* we re-implemented the netmask-to-prefix code differently. Check
* that they agree. */
g_assert_cmpint (prefix, ==, nm_utils_ip4_netmask_to_prefix (netmask));
return prefix;
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
test_ip4_prefix_to_netmask (void)
{
int i;
for (i = 0; i<=32; i++) {
all: use _nm_utils_ip4_*() utils functions
2017-09-05 15:25:34 +02:00
guint32 netmask = _nm_utils_ip4_prefix_to_netmask (i);
core: implement nm_utils_ip4_netmask_to_prefix() via __builtin_ctz() Taken from systemd's in4_addr_netmask_to_prefixlen(). Yes, this adds the requirement that "int" is 32 bits. But systemd already has the same requirement in u32ctz(), hence we anyway cannot build on other architectures. If that is ever necessary, it's easy to adjust.
2018-12-14 16:45:53 +01:00
int plen = _netmask_to_prefix (netmask);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert_cmpint (i, ==, plen);
{
guint32 msk = 0x80000000;
guint32 netmask2 = 0;
guint32 prefix = i;
while (prefix > 0) {
netmask2 |= msk;
msk >>= 1;
prefix--;
}
g_assert_cmpint (netmask, ==, (guint32) htonl (netmask2));
}
}
}
static void
test_ip4_netmask_to_prefix (void)
{
int i, j;
GRand *rand = g_rand_new ();
g_rand_set_seed (rand, 1);
for (i = 2; i<=32; i++) {
all: use _nm_utils_ip4_*() utils functions
2017-09-05 15:25:34 +02:00
guint32 netmask = _nm_utils_ip4_prefix_to_netmask (i);
guint32 netmask_lowest_bit = netmask & ~_nm_utils_ip4_prefix_to_netmask (i-1);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
core: implement nm_utils_ip4_netmask_to_prefix() via __builtin_ctz() Taken from systemd's in4_addr_netmask_to_prefixlen(). Yes, this adds the requirement that "int" is 32 bits. But systemd already has the same requirement in u32ctz(), hence we anyway cannot build on other architectures. If that is ever necessary, it's easy to adjust.
2018-12-14 16:45:53 +01:00
g_assert_cmpint (i, ==, _netmask_to_prefix (netmask));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
for (j = 0; j < 2*i; j++) {
guint32 r = g_rand_int (rand);
guint32 netmask_holey;
guint32 prefix_holey;
netmask_holey = (netmask & r) | netmask_lowest_bit;
if (netmask_holey == netmask)
continue;
/* create an invalid netmask with holes and check that the function
* returns the longest prefix. */
core: implement nm_utils_ip4_netmask_to_prefix() via __builtin_ctz() Taken from systemd's in4_addr_netmask_to_prefixlen(). Yes, this adds the requirement that "int" is 32 bits. But systemd already has the same requirement in u32ctz(), hence we anyway cannot build on other architectures. If that is ever necessary, it's easy to adjust.
2018-12-14 16:45:53 +01:00
prefix_holey = _netmask_to_prefix (netmask_holey);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert_cmpint (i, ==, prefix_holey);
}
}
g_rand_free (rand);
}
#define ASSERT_CHANGED(statement) \
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
G_STMT_START { \
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
changed = FALSE; \
statement; \
g_assert (changed); \
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
} G_STMT_END
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
#define ASSERT_UNCHANGED(statement) \
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
G_STMT_START { \
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
changed = FALSE; \
statement; \
g_assert (!changed); \
libnm/tests: convert test cases to use g_assert() instead of ASSERT()
2016-02-14 14:09:21 +01:00
} G_STMT_END
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
test_connection_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
connection = new_test_connection ();
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
/* Add new setting */
ASSERT_CHANGED (nm_connection_add_setting (connection, nm_setting_vlan_new ()));
/* Remove existing setting */
ASSERT_CHANGED (nm_connection_remove_setting (connection, NM_TYPE_SETTING_VLAN));
/* Remove non-existing setting */
ASSERT_UNCHANGED (nm_connection_remove_setting (connection, NM_TYPE_SETTING_VLAN));
g_object_unref (connection);
}
static void
test_setting_connection_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
NMSettingConnection *s_con;
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_free char *uuid = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
s_con = (NMSettingConnection *) nm_setting_connection_new ();
nm_connection_add_setting (connection, NM_SETTING (s_con));
ASSERT_CHANGED (g_object_set (s_con, NM_SETTING_CONNECTION_ID, "adfadfasdfaf", NULL));
ASSERT_CHANGED (nm_setting_connection_add_permission (s_con, "user", "billsmith", NULL));
ASSERT_CHANGED (nm_setting_connection_remove_permission (s_con, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (iter != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_connection_remove_permission (s_con, 1));
g_test_assert_expected_messages ();
uuid = nm_utils_uuid_generate ();
ASSERT_CHANGED (nm_setting_connection_add_secondary (s_con, uuid));
ASSERT_CHANGED (nm_setting_connection_remove_secondary (s_con, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (elt != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_connection_remove_secondary (s_con, 1));
g_test_assert_expected_messages ();
g_object_unref (connection);
}
static void
test_setting_bond_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
NMSettingBond *s_bond;
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
s_bond = (NMSettingBond *) nm_setting_bond_new ();
nm_connection_add_setting (connection, NM_SETTING (s_bond));
ASSERT_CHANGED (nm_setting_bond_add_option (s_bond, NM_SETTING_BOND_OPTION_DOWNDELAY, "10"));
ASSERT_CHANGED (nm_setting_bond_remove_option (s_bond, NM_SETTING_BOND_OPTION_DOWNDELAY));
ASSERT_UNCHANGED (nm_setting_bond_remove_option (s_bond, NM_SETTING_BOND_OPTION_UPDELAY));
g_object_unref (connection);
}
static void
test_setting_ip4_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NMSettingIPConfig *s_ip4;
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
NMIPAddress *addr;
NMIPRoute *route;
GError *error = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, NM_SETTING (s_ip4));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_dns (s_ip4, "11.22.0.0"));
ASSERT_CHANGED (nm_setting_ip_config_remove_dns (s_ip4, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->dns->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_dns (s_ip4, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_dns (s_ip4, "33.44.0.0");
ASSERT_CHANGED (nm_setting_ip_config_clear_dns (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_dns_search (s_ip4, "foobar.com"));
ASSERT_CHANGED (nm_setting_ip_config_remove_dns_search (s_ip4, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->dns_search->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_dns_search (s_ip4, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_dns_search (s_ip4, "foobar.com"));
ASSERT_CHANGED (nm_setting_ip_config_clear_dns_searches (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
addr = nm_ip_address_new (AF_INET, "22.33.0.0", 24, &error);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
g_assert_no_error (error);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_address (s_ip4, addr));
ASSERT_CHANGED (nm_setting_ip_config_remove_address (s_ip4, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->addresses->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_address (s_ip4, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_address (s_ip4, addr);
ASSERT_CHANGED (nm_setting_ip_config_clear_addresses (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
route = nm_ip_route_new (AF_INET, "22.33.0.0", 24, NULL, 0, &error);
g_assert_no_error (error);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_route (s_ip4, route));
ASSERT_CHANGED (nm_setting_ip_config_remove_route (s_ip4, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->routes->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_route (s_ip4, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_route (s_ip4, route);
ASSERT_CHANGED (nm_setting_ip_config_clear_routes (s_ip4));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: add dns-options property to NMSettingIPConfig
2015-03-26 08:57:02 +01:00
ASSERT_CHANGED (nm_setting_ip_config_add_dns_option (s_ip4, "debug"));
ASSERT_CHANGED (nm_setting_ip_config_remove_dns_option (s_ip4, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->dns_options->len));
libnm-core: add dns-options property to NMSettingIPConfig
2015-03-26 08:57:02 +01:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_dns_option (s_ip4, 1));
g_test_assert_expected_messages ();
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
nm_ip_address_unref (addr);
nm_ip_route_unref (route);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (connection);
}
static void
test_setting_ip6_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NMSettingIPConfig *s_ip6;
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
NMIPAddress *addr;
NMIPRoute *route;
GError *error = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
s_ip6 = (NMSettingIPConfig *) nm_setting_ip6_config_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, NM_SETTING (s_ip6));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_dns (s_ip6, "1:2:3::4:5:6"));
ASSERT_CHANGED (nm_setting_ip_config_remove_dns (s_ip6, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->dns->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_dns (s_ip6, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_dns (s_ip6, "1:2:3::4:5:6");
ASSERT_CHANGED (nm_setting_ip_config_clear_dns (s_ip6));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_dns_search (s_ip6, "foobar.com"));
ASSERT_CHANGED (nm_setting_ip_config_remove_dns_search (s_ip6, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->dns_search->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_dns_search (s_ip6, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_dns_search (s_ip6, "foobar.com");
ASSERT_CHANGED (nm_setting_ip_config_clear_dns_searches (s_ip6));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
addr = nm_ip_address_new (AF_INET6, "1:2:3::4:5:6", 64, &error);
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
g_assert_no_error (error);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_address (s_ip6, addr));
ASSERT_CHANGED (nm_setting_ip_config_remove_address (s_ip6, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->addresses->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_address (s_ip6, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_address (s_ip6, addr);
ASSERT_CHANGED (nm_setting_ip_config_clear_addresses (s_ip6));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
route = nm_ip_route_new (AF_INET6, "1:2:3::4:5:6", 128, NULL, 0, &error);
g_assert_no_error (error);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_CHANGED (nm_setting_ip_config_add_route (s_ip6, route));
ASSERT_CHANGED (nm_setting_ip_config_remove_route (s_ip6, 0));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx >= 0 && idx < priv->routes->len));
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
ASSERT_UNCHANGED (nm_setting_ip_config_remove_route (s_ip6, 1));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_test_assert_expected_messages ();
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
nm_setting_ip_config_add_route (s_ip6, route);
ASSERT_CHANGED (nm_setting_ip_config_clear_routes (s_ip6));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, all: merge IPv4 and IPv6 address/route types Merge NMIP4Address and NMIP6Address into NMIPAddress, and NMIP4Route and NMIP6Route into NMIPRoute. The new types represent IP addresses as strings, rather than in binary, and so are address-family agnostic.
2014-09-16 16:42:46 -04:00
nm_ip_address_unref (addr);
nm_ip_route_unref (route);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_unref (connection);
}
static void
test_setting_vlan_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
NMSettingVlan *s_vlan;
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
s_vlan = (NMSettingVlan *) nm_setting_vlan_new ();
nm_connection_add_setting (connection, NM_SETTING (s_vlan));
ASSERT_CHANGED (nm_setting_vlan_add_priority (s_vlan, NM_VLAN_INGRESS_MAP, 1, 3));
ASSERT_CHANGED (nm_setting_vlan_remove_priority (s_vlan, NM_VLAN_INGRESS_MAP, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx < g_slist_length (list)));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_vlan_remove_priority (s_vlan, NM_VLAN_INGRESS_MAP, 1));
g_test_assert_expected_messages ();
ASSERT_CHANGED (nm_setting_vlan_add_priority_str (s_vlan, NM_VLAN_INGRESS_MAP, "1:3"));
ASSERT_CHANGED (nm_setting_vlan_clear_priorities (s_vlan, NM_VLAN_INGRESS_MAP));
ASSERT_CHANGED (nm_setting_vlan_add_priority (s_vlan, NM_VLAN_EGRESS_MAP, 1, 3));
ASSERT_CHANGED (nm_setting_vlan_remove_priority (s_vlan, NM_VLAN_EGRESS_MAP, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (idx < g_slist_length (list)));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_vlan_remove_priority (s_vlan, NM_VLAN_EGRESS_MAP, 1));
g_test_assert_expected_messages ();
ASSERT_CHANGED (nm_setting_vlan_add_priority_str (s_vlan, NM_VLAN_EGRESS_MAP, "1:3"));
ASSERT_CHANGED (nm_setting_vlan_clear_priorities (s_vlan, NM_VLAN_EGRESS_MAP));
g_object_unref (connection);
}
static void
test_setting_vpn_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
libnm, core, cli, tui: fix the capitalization of various types GLib/Gtk have mostly settled on the convention that two-letter acronyms in type names remain all-caps (eg, "IO"), but longer acronyms become initial-caps-only (eg, "Tcp"). NM was inconsistent, with most long acronyms using initial caps only (Adsl, Cdma, Dcb, Gsm, Olpc, Vlan), but others using all caps (DHCP, PPP, PPPOE, VPN). Fix libnm and src/ to use initial-caps only for all three-or-more-letter-long acronyms (and update nmcli and nmtui for the libnm changes).
2014-06-26 13:44:36 -04:00
NMSettingVpn *s_vpn;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
libnm, core, cli, tui: fix the capitalization of various types GLib/Gtk have mostly settled on the convention that two-letter acronyms in type names remain all-caps (eg, "IO"), but longer acronyms become initial-caps-only (eg, "Tcp"). NM was inconsistent, with most long acronyms using initial caps only (Adsl, Cdma, Dcb, Gsm, Olpc, Vlan), but others using all caps (DHCP, PPP, PPPOE, VPN). Fix libnm and src/ to use initial-caps only for all three-or-more-letter-long acronyms (and update nmcli and nmtui for the libnm changes).
2014-06-26 13:44:36 -04:00
s_vpn = (NMSettingVpn *) nm_setting_vpn_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
nm_connection_add_setting (connection, NM_SETTING (s_vpn));
ASSERT_CHANGED (nm_setting_vpn_add_data_item (s_vpn, "foobar", "baz"));
ASSERT_CHANGED (nm_setting_vpn_remove_data_item (s_vpn, "foobar"));
ASSERT_UNCHANGED (nm_setting_vpn_remove_data_item (s_vpn, "not added"));
ASSERT_CHANGED (nm_setting_vpn_add_secret (s_vpn, "foobar", "baz"));
ASSERT_CHANGED (nm_setting_vpn_remove_secret (s_vpn, "foobar"));
ASSERT_UNCHANGED (nm_setting_vpn_remove_secret (s_vpn, "not added"));
g_object_unref (connection);
}
static void
test_setting_wired_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
NMSettingWired *s_wired;
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
s_wired = (NMSettingWired *) nm_setting_wired_new ();
nm_connection_add_setting (connection, NM_SETTING (s_wired));
ASSERT_CHANGED (nm_setting_wired_add_s390_option (s_wired, "portno", "1"));
ASSERT_CHANGED (nm_setting_wired_remove_s390_option (s_wired, "portno"));
ASSERT_UNCHANGED (nm_setting_wired_remove_s390_option (s_wired, "layer2"));
g_object_unref (connection);
}
static void
test_setting_wireless_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
NMSettingWireless *s_wifi;
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
s_wifi = (NMSettingWireless *) nm_setting_wireless_new ();
nm_connection_add_setting (connection, NM_SETTING (s_wifi));
ASSERT_CHANGED (nm_setting_wireless_add_seen_bssid (s_wifi, "00:11:22:33:44:55"));
g_object_unref (connection);
}
static void
test_setting_wireless_security_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
NMSettingWirelessSecurity *s_wsec;
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
s_wsec = (NMSettingWirelessSecurity *) nm_setting_wireless_security_new ();
nm_connection_add_setting (connection, NM_SETTING (s_wsec));
/* Protos */
ASSERT_CHANGED (nm_setting_wireless_security_add_proto (s_wsec, "wpa"));
ASSERT_CHANGED (nm_setting_wireless_security_remove_proto (s_wsec, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (elt != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_wireless_security_remove_proto (s_wsec, 1));
g_test_assert_expected_messages ();
nm_setting_wireless_security_add_proto (s_wsec, "wep");
ASSERT_CHANGED (nm_setting_wireless_security_clear_protos (s_wsec));
/* Pairwise ciphers */
ASSERT_CHANGED (nm_setting_wireless_security_add_pairwise (s_wsec, "tkip"));
ASSERT_CHANGED (nm_setting_wireless_security_remove_pairwise (s_wsec, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (elt != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_wireless_security_remove_pairwise (s_wsec, 1));
g_test_assert_expected_messages ();
nm_setting_wireless_security_add_pairwise (s_wsec, "tkip");
ASSERT_CHANGED (nm_setting_wireless_security_clear_pairwise (s_wsec));
/* Group ciphers */
ASSERT_CHANGED (nm_setting_wireless_security_add_group (s_wsec, "ccmp"));
ASSERT_CHANGED (nm_setting_wireless_security_remove_group (s_wsec, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (elt != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_wireless_security_remove_group (s_wsec, 1));
g_test_assert_expected_messages ();
nm_setting_wireless_security_add_group (s_wsec, "tkip");
ASSERT_CHANGED (nm_setting_wireless_security_clear_groups (s_wsec));
/* WEP key secret flags */
ASSERT_CHANGED (g_assert (nm_setting_set_secret_flags (NM_SETTING (s_wsec), "wep-key0", NM_SETTING_SECRET_FLAG_AGENT_OWNED, NULL)));
ASSERT_CHANGED (g_assert (nm_setting_set_secret_flags (NM_SETTING (s_wsec), "wep-key1", NM_SETTING_SECRET_FLAG_AGENT_OWNED, NULL)));
ASSERT_CHANGED (g_assert (nm_setting_set_secret_flags (NM_SETTING (s_wsec), "wep-key2", NM_SETTING_SECRET_FLAG_AGENT_OWNED, NULL)));
ASSERT_CHANGED (g_assert (nm_setting_set_secret_flags (NM_SETTING (s_wsec), "wep-key3", NM_SETTING_SECRET_FLAG_AGENT_OWNED, NULL)));
g_object_unref (connection);
}
static void
test_setting_802_1x_changed_signal (void)
{
NMConnection *connection;
gboolean changed = FALSE;
NMSetting8021x *s_8021x;
libnm-core, libnm, core: make NMConnection an interface The fact that NMRemoteConnection has to be an NMConnection and therefore can't be an NMObject means that it needs to reimplement bits of NMObject functionality (and likewise NMObject needs some special magic to deal with it). Likewise, we will need a daemon-side equivalent of NMObject as part of the gdbus port, and we would want NMSettingsConnection to be able to inherit from this as well. Solve this problem by making NMConnection into an interface, and having NMRemoteConnection and NMSettingsConnection implement it. (We use some hacks to keep the GHashTable of NMSettings objects inside nm-connection.c rather than having to be implemented by the implementations.) Since NMConnection is no longer an instantiable type, this adds NMSimpleConnection to replace the various non-D-Bus-based uses of NMConnection throughout the code. nm_connection_new() becomes nm_simple_connection_new(), nm_connection_new_from_hash() becomes nm_simple_connection_new_from_hash(), and nm_connection_duplicate() becomes nm_simple_connection_new_clone().
2014-08-13 14:34:29 -04:00
connection = nm_simple_connection_new ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_signal_connect (connection,
NM_CONNECTION_CHANGED,
(GCallback) test_connection_changed_cb,
&changed);
s_8021x = (NMSetting8021x *) nm_setting_802_1x_new ();
nm_connection_add_setting (connection, NM_SETTING (s_8021x));
/* EAP methods */
ASSERT_CHANGED (nm_setting_802_1x_add_eap_method (s_8021x, "tls"));
ASSERT_CHANGED (nm_setting_802_1x_remove_eap_method (s_8021x, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (elt != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_802_1x_remove_eap_method (s_8021x, 1));
g_test_assert_expected_messages ();
nm_setting_802_1x_add_eap_method (s_8021x, "ttls");
ASSERT_CHANGED (nm_setting_802_1x_clear_eap_methods (s_8021x));
/* alternate subject matches */
ASSERT_CHANGED (nm_setting_802_1x_add_altsubject_match (s_8021x, "EMAIL:server@example.com"));
ASSERT_CHANGED (nm_setting_802_1x_remove_altsubject_match (s_8021x, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (elt != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_802_1x_remove_altsubject_match (s_8021x, 1));
g_test_assert_expected_messages ();
nm_setting_802_1x_add_altsubject_match (s_8021x, "EMAIL:server@example.com");
ASSERT_CHANGED (nm_setting_802_1x_clear_altsubject_matches (s_8021x));
/* phase2 alternate subject matches */
ASSERT_CHANGED (nm_setting_802_1x_add_phase2_altsubject_match (s_8021x, "EMAIL:server@example.com"));
ASSERT_CHANGED (nm_setting_802_1x_remove_phase2_altsubject_match (s_8021x, 0));
tests: use NMTST_EXPECT*() macros Tests are commonly created via copy&paste. Hence, it's better to express a certain concept explicitly via a function or macro. This way, the implementation of the concept can be adjusted at one place, without requiring to change all the callers. Also, the macro is shorter, and brevity is better for tests so it's easier to understand what the test does. Without being bothered by noise from the redundant information. Also, the macro knows better which message to expect. For example, messages inside "src" are prepended by nm-logging.c with a level and a timestamp. The expect macro is aware of that and tests for it #define NMTST_EXPECT_NM_ERROR(msg) NMTST_EXPECT_NM (G_LOG_LEVEL_MESSAGE, "*<error> [*] "msg) This again allows the caller to ignore this prefix, but still assert more strictly.
2018-01-02 11:21:20 +01:00
NMTST_EXPECT_LIBNM_CRITICAL (NMTST_G_RETURN_MSG (elt != NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
ASSERT_UNCHANGED (nm_setting_802_1x_remove_phase2_altsubject_match (s_8021x, 1));
g_test_assert_expected_messages ();
nm_setting_802_1x_add_phase2_altsubject_match (s_8021x, "EMAIL:server@example.com");
ASSERT_CHANGED (nm_setting_802_1x_clear_phase2_altsubject_matches (s_8021x));
g_object_unref (connection);
}
static void
test_setting_old_uuid (void)
{
test: initialize auto-destructed pointers Otherwise the compiler complains that they could be left uninitialized in case the function returns too early. Fixes: 76745817c3e77b7d6c22290a7564b73224147516
2015-02-09 15:17:53 +01:00
gs_unref_object NMSetting *setting = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* NetworkManager-0.9.4.0 generated 40-character UUIDs with no dashes,
* like this one. Test that we maintain compatibility. */
const char *uuid = "f43bec2cdd60e5da381ebb1eb1fa39f3cc52660c";
setting = nm_setting_connection_new ();
g_object_set (G_OBJECT (setting),
NM_SETTING_CONNECTION_ID, "uuidtest",
NM_SETTING_CONNECTION_UUID, uuid,
NM_SETTING_CONNECTION_TYPE, NM_SETTING_WIRED_SETTING_NAME,
NULL);
tests: mute coverity for CHECKED_RETURN (CWE-252) in tests Error: CHECKED_RETURN (CWE-252): [#def12] NetworkManager-0.9.11.0/libnm-core/tests/test-general.c:348: check_return: Calling "nm_setting_verify" without checking return value (as is done elsewhere 37 out of 45 times). ...
2014-12-12 14:42:27 +01:00
nmtst_assert_setting_verifies (NM_SETTING (setting));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm: normalize missing connection UUID Extend nm_connection_normalize() to add a connection UUID in case it is unset.
2014-12-01 13:03:23 +01:00
static void
test_connection_normalize_uuid (void)
{
gs_unref_object NMConnection *con = NULL;
con = nmtst_create_minimal_connection ("test1", NULL, NM_SETTING_WIRED_SETTING_NAME, NULL);
nmtst_assert_connection_verifies_and_normalizable (con);
g_object_set (nm_connection_get_setting_connection (con),
NM_SETTING_CONNECTION_UUID, NULL,
NULL);
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm: normalize missing connection UUID Extend nm_connection_normalize() to add a connection UUID in case it is unset.
2014-12-01 13:03:23 +01:00
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/*
* Test normalization of interface-name
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
*/
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
static void
test_connection_normalize_virtual_iface_name (void)
{
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
NMConnection *con = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NMSettingConnection *s_con;
NMSettingVlan *s_vlan;
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
GVariant *connection_dict, *setting_dict, *var;
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
GError *error = NULL;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
const char *IFACE_NAME = "iface";
const char *IFACE_VIRT = "iface-X";
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
con = nmtst_create_minimal_connection ("test1",
"22001632-bbb4-4616-b277-363dce3dfb5b",
NM_SETTING_VLAN_SETTING_NAME,
&s_con);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
nm_connection_add_setting (con,
g_object_new (NM_TYPE_SETTING_IP4_CONFIG,
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_AUTO,
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
NULL));
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
nm_connection_add_setting (con,
g_object_new (NM_TYPE_SETTING_IP6_CONFIG,
libnm-core: extract NMSettingIPConfig superclass out of IP4, IP6 classes Split a base NMSettingIPConfig class out of NMSettingIP4Config and NMSettingIP6Config, and update things accordingly. Further simplifications of now-redundant IPv4-vs-IPv6 code are possible, and should happen in the future.
2014-10-19 17:30:10 -04:00
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_AUTO,
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
NULL));
s_vlan = nm_connection_get_setting_vlan (con);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_object_set (G_OBJECT (s_vlan),
NM_SETTING_VLAN_PARENT, "eth0",
NULL);
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
g_object_set (G_OBJECT (s_con), NM_SETTING_CONNECTION_INTERFACE_NAME, IFACE_NAME, NULL);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
g_assert_cmpstr (nm_connection_get_interface_name (con), ==, IFACE_NAME);
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
connection_dict = nm_connection_to_dbus (con, NM_CONNECTION_SERIALIZE_ALL);
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
g_object_unref (con);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
/* Serialized form should include vlan.interface-name as well. */
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
setting_dict = g_variant_lookup_value (connection_dict, NM_SETTING_VLAN_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (setting_dict != NULL);
var = g_variant_lookup_value (setting_dict, "interface-name", NULL);
g_assert (var != NULL);
g_assert (g_variant_is_of_type (var, G_VARIANT_TYPE_STRING));
g_assert_cmpstr (g_variant_get_string (var, NULL), ==, IFACE_NAME);
g_variant_unref (setting_dict);
g_variant_unref (var);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
settings: simplify property setter from GVariant to NMSettingConnection:interface-name The interface-name property has several deprecated aliases, like "bridge.interface-name". For backward compatibility, we keep handling them. In particular, the "missing_from_dbus_fcn" handler is set. This handles the case where GVariant only contains the deprecated form, but not "connection.interface-name". Previously, from_dbus_fcn() would check whether the deprecated form was present, and -- only if that form was invalid -- prefer it. The idea was to fail validation if the deprecated property was invalid. I think that is not necessary. Just completely ignore the deprecated property, if the new property is present. What might make sense is to check whether the deprecated and the new form are both present, that they are identical. However, I don't think that is worth the effort.
2020-02-17 17:40:17 +01:00
/* If vlan.interface-name will be ignored. */
libnm-core: update tests to be fully variant-based Use nmtst_variant_editor() to rewrite the libnm-core tests to work entirely in connection dictionaries, not connection hash tables.
2014-09-05 14:40:37 -04:00
NMTST_VARIANT_EDITOR (connection_dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_VLAN_SETTING_NAME,
"interface-name",
"s",
":::this-is-not-a-valid-interface-name:::");
);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
con = _connection_new_from_dbus (connection_dict, &error);
settings: simplify property setter from GVariant to NMSettingConnection:interface-name The interface-name property has several deprecated aliases, like "bridge.interface-name". For backward compatibility, we keep handling them. In particular, the "missing_from_dbus_fcn" handler is set. This handles the case where GVariant only contains the deprecated form, but not "connection.interface-name". Previously, from_dbus_fcn() would check whether the deprecated form was present, and -- only if that form was invalid -- prefer it. The idea was to fail validation if the deprecated property was invalid. I think that is not necessary. Just completely ignore the deprecated property, if the new property is present. What might make sense is to check whether the deprecated and the new form are both present, that they are identical. However, I don't think that is worth the effort.
2020-02-17 17:40:17 +01:00
nmtst_assert_success (con, error);
g_assert_cmpstr (nm_connection_get_interface_name (con), ==, IFACE_NAME);
g_clear_object (&con);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
/* If vlan.interface-name is valid, but doesn't match, it will be ignored. */
libnm-core: update tests to be fully variant-based Use nmtst_variant_editor() to rewrite the libnm-core tests to work entirely in connection dictionaries, not connection hash tables.
2014-09-05 14:40:37 -04:00
NMTST_VARIANT_EDITOR (connection_dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_VLAN_SETTING_NAME,
"interface-name",
"s",
IFACE_VIRT);
);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
con = _connection_new_from_dbus (connection_dict, &error);
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
g_assert_no_error (error);
libnm-core: fix NMSettingConnection:verify() not to modify interface-name verify() used to modify interface-name of the base settings. This is discouraged, because verify() should not touch the connection. For libnm-core we can change behavior and only modify the connection in normalize(). Also, be more strict not to verify() sucessfully on invalid interface-name. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 15:02:30 +02:00
g_assert_cmpstr (nm_connection_get_interface_name (con), ==, IFACE_NAME);
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
s_con = nm_connection_get_setting_connection (con);
g_assert_cmpstr (nm_setting_connection_get_interface_name (s_con), ==, IFACE_NAME);
g_object_unref (con);
/* But removing connection.interface-name should result in vlan.connection-name
* being "promoted".
*/
libnm-core: update tests to be fully variant-based Use nmtst_variant_editor() to rewrite the libnm-core tests to work entirely in connection dictionaries, not connection hash tables.
2014-09-05 14:40:37 -04:00
NMTST_VARIANT_EDITOR (connection_dict,
NMTST_VARIANT_DROP_PROPERTY (NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_CONNECTION_INTERFACE_NAME);
);
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
con = _connection_new_from_dbus (connection_dict, &error);
libnm-core: drop :interface-name properties on virtual NMSetting types Remove the virtual :interface-name properties and their getters, and use property overrides to do backward-compat handling when serializing/deserializing. Now when constructing an NMConnection from a hash, if the virtual property is set and the NMSettingConnection property isn't, then the override for NMSettingConnection:interface-name will set that property to the value of the virtual interface-name. And when converting an NMConnection to a hash, the overrides for the virtual properties will return the value of NMSettingConnection:interface-name.
2014-08-04 19:57:20 -04:00
g_assert_no_error (error);
g_assert_cmpstr (nm_connection_get_interface_name (con), ==, IFACE_VIRT);
s_con = nm_connection_get_setting_connection (con);
g_assert_cmpstr (nm_setting_connection_get_interface_name (s_con), ==, IFACE_VIRT);
g_object_unref (con);
libnm-core: change connection hash tables to variants in API In preparation for porting to GDBus, make nm_connection_to_dbus(), etc, represent connections as GVariants of type 'a{sa{sv}}' rather than as GHashTables-of-GHashTables-of-GValues. This means we're constantly converting back and forth internally, but this is just a stepping stone on the way to the full GDBus port, and all of that code will go away again later.
2014-08-16 10:09:48 -04:00
g_variant_unref (connection_dict);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
static void
_test_connection_normalize_type_normalizable_setting (const char *type,
void (*prepare_normalizable_fcn) (NMConnection *con))
{
NMSettingConnection *s_con;
NMSetting *s_base;
GType base_type;
gs_unref_object NMConnection *con = NULL;
gs_free char *id = g_strdup_printf ("%s[%s]", G_STRFUNC, type);
base_type = nm_setting_lookup_type (type);
g_assert (base_type != G_TYPE_INVALID);
libnm: add enum for setting priorities Using plain numbers make it cumbersome to grep for setting types by priority. The only downside is, that with the enum values it is no longer obvious which value has higher or lower priority. Also, introduce NM_SETTING_PRIORITY_INVALID. This is what _nm_setting_type_get_base_type_priority() returns. For the moment it still has the same numerical value 0 as before. Later, that shall be distinct from NM_SETTING_PRIORITY_CONNECTION.
2017-06-01 13:43:52 +02:00
g_assert (_nm_setting_type_get_base_type_priority (base_type) != NM_SETTING_PRIORITY_INVALID);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
con = nmtst_create_minimal_connection (id, NULL, NULL, &s_con);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
g_object_set (s_con, NM_SETTING_CONNECTION_TYPE, type, NULL);
if (prepare_normalizable_fcn)
prepare_normalizable_fcn (con);
g_assert (!nm_connection_get_setting_by_name (con, type));
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_SETTING);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_connection_normalize (con);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
s_base = nm_connection_get_setting_by_name (con, type);
g_assert (s_base);
g_assert (G_OBJECT_TYPE (s_base) == base_type);
}
static void
_test_connection_normalize_type_unnormalizable_setting (const char *type)
{
NMSettingConnection *s_con;
GType base_type;
gs_unref_object NMConnection *con = NULL;
gs_free char *id = g_strdup_printf ("%s[%s]", G_STRFUNC, type);
base_type = nm_setting_lookup_type (type);
g_assert (base_type != G_TYPE_INVALID);
libnm: add enum for setting priorities Using plain numbers make it cumbersome to grep for setting types by priority. The only downside is, that with the enum values it is no longer obvious which value has higher or lower priority. Also, introduce NM_SETTING_PRIORITY_INVALID. This is what _nm_setting_type_get_base_type_priority() returns. For the moment it still has the same numerical value 0 as before. Later, that shall be distinct from NM_SETTING_PRIORITY_CONNECTION.
2017-06-01 13:43:52 +02:00
g_assert (_nm_setting_type_get_base_type_priority (base_type) != NM_SETTING_PRIORITY_INVALID);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
con = nmtst_create_minimal_connection (id, NULL, NULL, &s_con);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
g_object_set (s_con, NM_SETTING_CONNECTION_TYPE, type, NULL);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_SETTING);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
}
static void
_test_connection_normalize_type_normalizable_type (const char *type,
NMSetting *(*add_setting_fcn) (NMConnection *con))
{
NMSettingConnection *s_con;
NMSetting *s_base;
GType base_type;
gs_unref_object NMConnection *con = NULL;
gs_free char *id = g_strdup_printf ("%s[%s]", G_STRFUNC, type);
base_type = nm_setting_lookup_type (type);
g_assert (base_type != G_TYPE_INVALID);
libnm: add enum for setting priorities Using plain numbers make it cumbersome to grep for setting types by priority. The only downside is, that with the enum values it is no longer obvious which value has higher or lower priority. Also, introduce NM_SETTING_PRIORITY_INVALID. This is what _nm_setting_type_get_base_type_priority() returns. For the moment it still has the same numerical value 0 as before. Later, that shall be distinct from NM_SETTING_PRIORITY_CONNECTION.
2017-06-01 13:43:52 +02:00
g_assert (_nm_setting_type_get_base_type_priority (base_type) != NM_SETTING_PRIORITY_INVALID);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
con = nmtst_create_minimal_connection (id, NULL, NULL, &s_con);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
if (add_setting_fcn)
s_base = add_setting_fcn (con);
else {
s_base = NM_SETTING (g_object_new (base_type, NULL));
nm_connection_add_setting (con, s_base);
}
core/connections: pick base setting from settings the connection actually has We will need multiple base settings for Bluetooth NAP servers: bluetooth and bridge. We want to identify the device as "bluetooth" to the user, but leave the Bridge factory handle it. The "connection.type" is somewhat redundant -- let's keep it for what the user sees. And identify the actual base setting to pick the right factory by the actually present setting.
2017-05-22 20:11:40 +02:00
g_assert (!nm_setting_connection_get_connection_type (s_con));
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
g_assert (nm_connection_get_setting_by_name (con, type) == s_base);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_connection_normalize (con);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
g_assert_cmpstr (nm_connection_get_connection_type (con), ==, type);
g_assert (nm_connection_get_setting_by_name (con, type) == s_base);
}
static NMSetting *
_add_setting_fcn_adsl (NMConnection *con)
{
NMSetting *setting;
setting = g_object_new (NM_TYPE_SETTING_ADSL,
NM_SETTING_ADSL_USERNAME, "test-user",
NM_SETTING_ADSL_PROTOCOL, NM_SETTING_ADSL_PROTOCOL_PPPOA,
NM_SETTING_ADSL_ENCAPSULATION, NM_SETTING_ADSL_ENCAPSULATION_VCMUX,
NULL);
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_bluetooth (NMConnection *con)
{
NMSetting *setting;
setting = g_object_new (NM_TYPE_SETTING_BLUETOOTH,
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
NM_SETTING_BLUETOOTH_BDADDR, "11:22:33:44:55:66",
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
NM_SETTING_BLUETOOTH_TYPE, NM_SETTING_BLUETOOTH_TYPE_PANU,
NULL);
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_bond (NMConnection *con)
{
NMSetting *setting;
NMSettingConnection *s_con;
setting = g_object_new (NM_TYPE_SETTING_BOND, NULL);
nm_connection_add_setting (con, setting);
s_con = nm_connection_get_setting_connection (con);
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "test-bond",
NULL);
return setting;
}
static NMSetting *
_add_setting_fcn_bridge (NMConnection *con)
{
NMSetting *setting;
NMSettingConnection *s_con;
setting = g_object_new (NM_TYPE_SETTING_BRIDGE, NULL);
nm_connection_add_setting (con, setting);
s_con = nm_connection_get_setting_connection (con);
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "test-bridge",
NULL);
return setting;
}
static NMSetting *
_add_setting_fcn_cdma (NMConnection *con)
{
NMSetting *setting;
setting = g_object_new (NM_TYPE_SETTING_CDMA,
NM_SETTING_CDMA_NUMBER, "test-number",
NULL);
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_infiniband (NMConnection *con)
{
NMSetting *setting;
setting = g_object_new (NM_TYPE_SETTING_INFINIBAND,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "connected",
NULL);
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_olpc_mesh (NMConnection *con)
{
NMSetting *setting;
const char *ssid_data = "ssid-test";
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
GBytes *ssid;
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
ssid = g_bytes_new (ssid_data, strlen (ssid_data));
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
setting = g_object_new (NM_TYPE_SETTING_OLPC_MESH,
NM_SETTING_OLPC_MESH_SSID, ssid,
NM_SETTING_OLPC_MESH_CHANNEL, 1,
NULL);
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
g_bytes_unref (ssid);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_team (NMConnection *con)
{
NMSetting *setting;
NMSettingConnection *s_con;
setting = g_object_new (NM_TYPE_SETTING_TEAM, NULL);
nm_connection_add_setting (con, setting);
s_con = nm_connection_get_setting_connection (con);
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "test-team",
NULL);
return setting;
}
static NMSetting *
_add_setting_fcn_vlan (NMConnection *con)
{
NMSetting *setting;
setting = g_object_new (NM_TYPE_SETTING_VLAN,
NM_SETTING_VLAN_PARENT, "test-parent",
NULL);
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_vpn (NMConnection *con)
{
NMSetting *setting;
setting = g_object_new (NM_TYPE_SETTING_VPN,
NM_SETTING_VPN_SERVICE_TYPE, "test-vpn-service-type",
NULL);
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_wimax (NMConnection *con)
{
NMSetting *setting;
setting = g_object_new (NM_TYPE_SETTING_WIMAX,
NM_SETTING_WIMAX_NETWORK_NAME, "test-network",
NULL);
nm_connection_add_setting (con, setting);
return setting;
}
static NMSetting *
_add_setting_fcn_wireless (NMConnection *con)
{
NMSetting *setting;
const char *ssid_data = "ssid-test";
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
GBytes *ssid;
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
ssid = g_bytes_new (ssid_data, strlen (ssid_data));
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
setting = g_object_new (NM_TYPE_SETTING_WIRELESS,
NM_SETTING_WIRELESS_SSID, ssid,
NULL);
libnm-core: change DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES Change all DBUS_TYPE_G_UCHAR_ARRAY properties to G_TYPE_BYTES, and update corresponding APIs. Notably, this means they are now refcounted rather than being copied. Update the rest of NM for the changes. The daemon still converts SSIDs to GByteArrays internally, because changing it to use GBytes has lots of trickle-down effects. It can possibly be changed later.
2014-06-26 10:42:11 -04:00
g_bytes_unref (ssid);
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
nm_connection_add_setting (con, setting);
return setting;
}
static void
_prepare_normalizable_fcn_vlan (NMConnection *con)
{
nm_connection_add_setting (con, g_object_new (NM_TYPE_SETTING_WIRED,
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
NM_SETTING_WIRED_MAC_ADDRESS, "11:22:33:44:55:66",
libnm-core: normalize NMSettingConnection:type property nm_connection_normalize() can now detect the 'type' property based on existing base settings. It can also create a (default) base setting. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-08 18:36:02 +02:00
NULL));
}
static void
test_connection_normalize_type (void)
{
guint i;
struct {
const char *type;
gboolean normalizable;
NMSetting *(*add_setting_fcn) (NMConnection *con);
void (*prepare_normalizable_fcn) (NMConnection *con);
} types[] = {
{ NM_SETTING_GENERIC_SETTING_NAME, TRUE },
{ NM_SETTING_GSM_SETTING_NAME, TRUE },
{ NM_SETTING_WIRED_SETTING_NAME, TRUE },
{ NM_SETTING_VLAN_SETTING_NAME, TRUE, _add_setting_fcn_vlan, _prepare_normalizable_fcn_vlan },
{ NM_SETTING_ADSL_SETTING_NAME, FALSE, _add_setting_fcn_adsl },
{ NM_SETTING_BLUETOOTH_SETTING_NAME, FALSE, _add_setting_fcn_bluetooth },
{ NM_SETTING_BOND_SETTING_NAME, FALSE, _add_setting_fcn_bond },
{ NM_SETTING_BRIDGE_SETTING_NAME, FALSE, _add_setting_fcn_bridge },
{ NM_SETTING_CDMA_SETTING_NAME, FALSE, _add_setting_fcn_cdma },
{ NM_SETTING_INFINIBAND_SETTING_NAME, FALSE, _add_setting_fcn_infiniband },
{ NM_SETTING_OLPC_MESH_SETTING_NAME, FALSE, _add_setting_fcn_olpc_mesh },
{ NM_SETTING_TEAM_SETTING_NAME, FALSE, _add_setting_fcn_team },
{ NM_SETTING_VLAN_SETTING_NAME, FALSE, _add_setting_fcn_vlan },
{ NM_SETTING_VPN_SETTING_NAME, FALSE, _add_setting_fcn_vpn },
{ NM_SETTING_WIMAX_SETTING_NAME, FALSE, _add_setting_fcn_wimax },
{ NM_SETTING_WIRELESS_SETTING_NAME, FALSE, _add_setting_fcn_wireless },
{ 0 },
};
for (i = 0; types[i].type; i++) {
const char *type = types[i].type;
if (types[i].normalizable)
_test_connection_normalize_type_normalizable_setting (type, types[i].prepare_normalizable_fcn);
else
_test_connection_normalize_type_unnormalizable_setting (type);
_test_connection_normalize_type_normalizable_type (type, types[i].add_setting_fcn);
}
}
libnm-core: normalize slave-type and slave-settings of connections Some NMSettingConnection:slave-type types require a matching slave #NMSetting. Add normalization of either the 'slave-type' property or the slave-setting. Also be more strict in NMSettingConnection:verify() to enforce an existing slave-setting depending on the slave-type. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-07 17:05:10 +02:00
static void
test_connection_normalize_slave_type_1 (void)
{
gs_unref_object NMConnection *con = NULL;
NMSettingConnection *s_con;
con = nmtst_create_minimal_connection ("test_connection_normalize_slave_type_1",
"cc4cd5df-45dc-483e-b291-6b76c2338ecb",
NM_SETTING_WIRED_SETTING_NAME, &s_con);
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, "invalid-type",
NULL);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm-core: normalize slave-type and slave-settings of connections Some NMSettingConnection:slave-type types require a matching slave #NMSetting. Add normalization of either the 'slave-type' property or the slave-setting. Also be more strict in NMSettingConnection:verify() to enforce an existing slave-setting depending on the slave-type. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-07 17:05:10 +02:00
g_assert (!nm_connection_get_setting_by_name (con, NM_SETTING_BRIDGE_PORT_SETTING_NAME));
g_object_set (s_con,
NM_SETTING_CONNECTION_SLAVE_TYPE, "bridge",
NULL);
g_assert (!nm_connection_get_setting_by_name (con, NM_SETTING_BRIDGE_PORT_SETTING_NAME));
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_SETTING);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_connection_normalize (con);
libnm-core: normalize slave-type and slave-settings of connections Some NMSettingConnection:slave-type types require a matching slave #NMSetting. Add normalization of either the 'slave-type' property or the slave-setting. Also be more strict in NMSettingConnection:verify() to enforce an existing slave-setting depending on the slave-type. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-07 17:05:10 +02:00
g_assert (nm_connection_get_setting_by_name (con, NM_SETTING_BRIDGE_PORT_SETTING_NAME));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_BRIDGE_SETTING_NAME);
}
static void
test_connection_normalize_slave_type_2 (void)
{
gs_unref_object NMConnection *con = NULL;
NMSettingConnection *s_con;
con = nmtst_create_minimal_connection ("test_connection_normalize_slave_type_2",
"40bea008-ca72-439a-946b-e65f827656f9",
NM_SETTING_WIRED_SETTING_NAME, &s_con);
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, "invalid-type",
NULL);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
libnm-core: normalize slave-type and slave-settings of connections Some NMSettingConnection:slave-type types require a matching slave #NMSetting. Add normalization of either the 'slave-type' property or the slave-setting. Also be more strict in NMSettingConnection:verify() to enforce an existing slave-setting depending on the slave-type. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-07 17:05:10 +02:00
g_assert (!nm_connection_get_setting_by_name (con, NM_SETTING_BRIDGE_PORT_SETTING_NAME));
g_object_set (s_con,
NM_SETTING_CONNECTION_SLAVE_TYPE, NULL,
NULL);
nm_connection_add_setting (con, nm_setting_bridge_port_new ());
g_assert (nm_connection_get_setting_by_name (con, NM_SETTING_BRIDGE_PORT_SETTING_NAME));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NULL);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_connection_normalize (con);
libnm-core: normalize slave-type and slave-settings of connections Some NMSettingConnection:slave-type types require a matching slave #NMSetting. Add normalization of either the 'slave-type' property or the slave-setting. Also be more strict in NMSettingConnection:verify() to enforce an existing slave-setting depending on the slave-type. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-07-07 17:05:10 +02:00
g_assert (nm_connection_get_setting_by_name (con, NM_SETTING_BRIDGE_PORT_SETTING_NAME));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_BRIDGE_SETTING_NAME);
}
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
static void
test_connection_normalize_infiniband_mtu (void)
{
gs_unref_object NMConnection *con = NULL;
NMSettingInfiniband *s_infini;
libnm/infiniband: lift restriction of MTU to 2044 for IPoIB in "datagram" mode Traditionally, the MTU in "datagram" transport mode was restricted to 2044. That is no longer the case, relax that. In fact, choose a very large maximum and don't differenciate between "connected" mode (they now both use now 65520). This is only the limitation of the connection profile. Whether setting such large MTUs actually works must be determined when activating the profile. Initscripts "ifup-ib" from rdma-core package originally had a limit of 2044. This was raised to 4092 in rh#1186498. It is suggested to raise it further in bug rh#1647541. In general, kernel often does not allow setting large MTUs. And even if it allows it, it may not work because it also requires the entire network to be configured accordingly. But that means, it is generally not helpful to limit the MTU in the connection profile too strictly. Just allow large MTUs, we need to see at activation time whether the configuration works. Note also that all other setting types don't validate the range for MTU at all. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1186498 Related: https://bugzilla.redhat.com/show_bug.cgi?id=1593334 (rdma-core: raise limit from 2044 to 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1647541 (rdma-core: raise limit beyond 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1532638#c4 (rdma-core: MTU related discussion) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534869 (NetworkManager bug about this topic, but with lots of unrelated discussion. See in particular #c16) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1653494
2019-04-05 09:49:38 +02:00
guint mtu_regular = nmtst_rand_select (2044, 2045, 65520);
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
con = nmtst_create_minimal_connection ("test_connection_normalize_infiniband_mtu", NULL,
NM_SETTING_INFINIBAND_SETTING_NAME, NULL);
s_infini = nm_connection_get_setting_infiniband (con);
g_object_set (s_infini,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "connected",
NULL);
nmtst_assert_connection_verifies_and_normalizable (con);
g_object_set (s_infini,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "datagram",
libnm/infiniband: lift restriction of MTU to 2044 for IPoIB in "datagram" mode Traditionally, the MTU in "datagram" transport mode was restricted to 2044. That is no longer the case, relax that. In fact, choose a very large maximum and don't differenciate between "connected" mode (they now both use now 65520). This is only the limitation of the connection profile. Whether setting such large MTUs actually works must be determined when activating the profile. Initscripts "ifup-ib" from rdma-core package originally had a limit of 2044. This was raised to 4092 in rh#1186498. It is suggested to raise it further in bug rh#1647541. In general, kernel often does not allow setting large MTUs. And even if it allows it, it may not work because it also requires the entire network to be configured accordingly. But that means, it is generally not helpful to limit the MTU in the connection profile too strictly. Just allow large MTUs, we need to see at activation time whether the configuration works. Note also that all other setting types don't validate the range for MTU at all. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1186498 Related: https://bugzilla.redhat.com/show_bug.cgi?id=1593334 (rdma-core: raise limit from 2044 to 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1647541 (rdma-core: raise limit beyond 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1532638#c4 (rdma-core: MTU related discussion) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534869 (NetworkManager bug about this topic, but with lots of unrelated discussion. See in particular #c16) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1653494
2019-04-05 09:49:38 +02:00
NM_SETTING_INFINIBAND_MTU, (guint) mtu_regular,
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
NULL);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_assert_connection_verifies_and_normalizable (con);
nmtst_connection_normalize (con);
libnm/infiniband: lift restriction of MTU to 2044 for IPoIB in "datagram" mode Traditionally, the MTU in "datagram" transport mode was restricted to 2044. That is no longer the case, relax that. In fact, choose a very large maximum and don't differenciate between "connected" mode (they now both use now 65520). This is only the limitation of the connection profile. Whether setting such large MTUs actually works must be determined when activating the profile. Initscripts "ifup-ib" from rdma-core package originally had a limit of 2044. This was raised to 4092 in rh#1186498. It is suggested to raise it further in bug rh#1647541. In general, kernel often does not allow setting large MTUs. And even if it allows it, it may not work because it also requires the entire network to be configured accordingly. But that means, it is generally not helpful to limit the MTU in the connection profile too strictly. Just allow large MTUs, we need to see at activation time whether the configuration works. Note also that all other setting types don't validate the range for MTU at all. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1186498 Related: https://bugzilla.redhat.com/show_bug.cgi?id=1593334 (rdma-core: raise limit from 2044 to 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1647541 (rdma-core: raise limit beyond 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1532638#c4 (rdma-core: MTU related discussion) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534869 (NetworkManager bug about this topic, but with lots of unrelated discussion. See in particular #c16) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1653494
2019-04-05 09:49:38 +02:00
g_assert_cmpint (mtu_regular, ==, nm_setting_infiniband_get_mtu (s_infini));
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
g_object_set (s_infini,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "datagram",
libnm/infiniband: lift restriction of MTU to 2044 for IPoIB in "datagram" mode Traditionally, the MTU in "datagram" transport mode was restricted to 2044. That is no longer the case, relax that. In fact, choose a very large maximum and don't differenciate between "connected" mode (they now both use now 65520). This is only the limitation of the connection profile. Whether setting such large MTUs actually works must be determined when activating the profile. Initscripts "ifup-ib" from rdma-core package originally had a limit of 2044. This was raised to 4092 in rh#1186498. It is suggested to raise it further in bug rh#1647541. In general, kernel often does not allow setting large MTUs. And even if it allows it, it may not work because it also requires the entire network to be configured accordingly. But that means, it is generally not helpful to limit the MTU in the connection profile too strictly. Just allow large MTUs, we need to see at activation time whether the configuration works. Note also that all other setting types don't validate the range for MTU at all. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1186498 Related: https://bugzilla.redhat.com/show_bug.cgi?id=1593334 (rdma-core: raise limit from 2044 to 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1647541 (rdma-core: raise limit beyond 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1532638#c4 (rdma-core: MTU related discussion) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534869 (NetworkManager bug about this topic, but with lots of unrelated discussion. See in particular #c16) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1653494
2019-04-05 09:49:38 +02:00
NM_SETTING_INFINIBAND_MTU, (guint) 65521,
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
NULL);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_connection_normalize (con);
libnm/infiniband: lift restriction of MTU to 2044 for IPoIB in "datagram" mode Traditionally, the MTU in "datagram" transport mode was restricted to 2044. That is no longer the case, relax that. In fact, choose a very large maximum and don't differenciate between "connected" mode (they now both use now 65520). This is only the limitation of the connection profile. Whether setting such large MTUs actually works must be determined when activating the profile. Initscripts "ifup-ib" from rdma-core package originally had a limit of 2044. This was raised to 4092 in rh#1186498. It is suggested to raise it further in bug rh#1647541. In general, kernel often does not allow setting large MTUs. And even if it allows it, it may not work because it also requires the entire network to be configured accordingly. But that means, it is generally not helpful to limit the MTU in the connection profile too strictly. Just allow large MTUs, we need to see at activation time whether the configuration works. Note also that all other setting types don't validate the range for MTU at all. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1186498 Related: https://bugzilla.redhat.com/show_bug.cgi?id=1593334 (rdma-core: raise limit from 2044 to 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1647541 (rdma-core: raise limit beyond 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1532638#c4 (rdma-core: MTU related discussion) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534869 (NetworkManager bug about this topic, but with lots of unrelated discussion. See in particular #c16) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1653494
2019-04-05 09:49:38 +02:00
g_assert_cmpint (65520, ==, nm_setting_infiniband_get_mtu (s_infini));
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
g_object_set (s_infini,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "connected",
libnm/infiniband: lift restriction of MTU to 2044 for IPoIB in "datagram" mode Traditionally, the MTU in "datagram" transport mode was restricted to 2044. That is no longer the case, relax that. In fact, choose a very large maximum and don't differenciate between "connected" mode (they now both use now 65520). This is only the limitation of the connection profile. Whether setting such large MTUs actually works must be determined when activating the profile. Initscripts "ifup-ib" from rdma-core package originally had a limit of 2044. This was raised to 4092 in rh#1186498. It is suggested to raise it further in bug rh#1647541. In general, kernel often does not allow setting large MTUs. And even if it allows it, it may not work because it also requires the entire network to be configured accordingly. But that means, it is generally not helpful to limit the MTU in the connection profile too strictly. Just allow large MTUs, we need to see at activation time whether the configuration works. Note also that all other setting types don't validate the range for MTU at all. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1186498 Related: https://bugzilla.redhat.com/show_bug.cgi?id=1593334 (rdma-core: raise limit from 2044 to 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1647541 (rdma-core: raise limit beyond 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1532638#c4 (rdma-core: MTU related discussion) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534869 (NetworkManager bug about this topic, but with lots of unrelated discussion. See in particular #c16) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1653494
2019-04-05 09:49:38 +02:00
NM_SETTING_INFINIBAND_MTU, (guint) mtu_regular,
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
NULL);
nmtst_assert_connection_verifies_without_normalization (con);
libnm/infiniband: lift restriction of MTU to 2044 for IPoIB in "datagram" mode Traditionally, the MTU in "datagram" transport mode was restricted to 2044. That is no longer the case, relax that. In fact, choose a very large maximum and don't differenciate between "connected" mode (they now both use now 65520). This is only the limitation of the connection profile. Whether setting such large MTUs actually works must be determined when activating the profile. Initscripts "ifup-ib" from rdma-core package originally had a limit of 2044. This was raised to 4092 in rh#1186498. It is suggested to raise it further in bug rh#1647541. In general, kernel often does not allow setting large MTUs. And even if it allows it, it may not work because it also requires the entire network to be configured accordingly. But that means, it is generally not helpful to limit the MTU in the connection profile too strictly. Just allow large MTUs, we need to see at activation time whether the configuration works. Note also that all other setting types don't validate the range for MTU at all. Related: https://bugzilla.redhat.com/show_bug.cgi?id=1186498 Related: https://bugzilla.redhat.com/show_bug.cgi?id=1593334 (rdma-core: raise limit from 2044 to 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1647541 (rdma-core: raise limit beyond 4092 in ifup-ib) Related: https://bugzilla.redhat.com/show_bug.cgi?id=1532638#c4 (rdma-core: MTU related discussion) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1534869 (NetworkManager bug about this topic, but with lots of unrelated discussion. See in particular #c16) Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1653494
2019-04-05 09:49:38 +02:00
g_assert_cmpint (mtu_regular, ==, nm_setting_infiniband_get_mtu (s_infini));
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
g_object_set (s_infini,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "connected",
NM_SETTING_INFINIBAND_MTU, (guint) 65521,
NULL);
libnm-core: merge NMSetting*Error into NMConnectionError Each setting type was defining its own error type, but most of them had exactly the same three errors ("unknown", "missing property", and "invalid property"), and none of the other values was of much use programmatically anyway. So, this commit merges NMSettingError, NMSettingAdslError, etc, all into NMConnectionError. (The reason for merging into NMConnectionError rather than NMSettingError is that we also already have "NMSettingsError", for errors related to the settings service, so "NMConnectionError" is a less-confusable name for settings/connection errors than "NMSettingError".) Also, make sure that all of the affected error messages are localized, and (where appropriate) prefix them with the relevant property name. Renamed error codes: NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET Remapped error codes: NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY Dropped error codes (were previously defined but unused): NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
2014-10-20 13:52:23 -04:00
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_connection_normalize (con);
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
g_assert_cmpint (65520, ==, nm_setting_infiniband_get_mtu (s_infini));
}
libnm-core: remove gateway when never-default=yes in NMSettingIPConfig Having a gateway defined when never-default=yes causes troubles in connection matching and anyway makes no sense. If the combination is found, remove the gateway during the normalization phase. https://bugzilla.redhat.com/show_bug.cgi?id=1313091
2016-04-22 13:45:08 +02:00
static void
test_connection_normalize_gateway_never_default (void)
{
gs_unref_object NMConnection *con = NULL;
NMSettingIPConfig *s_ip4, *s_ip6;
NMIPAddress *addr;
gs_free_error GError *error = NULL;
con = nmtst_create_minimal_connection ("test1", NULL, NM_SETTING_WIRED_SETTING_NAME, NULL);
nmtst_assert_connection_verifies_and_normalizable (con);
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
g_object_set (G_OBJECT (s_ip4),
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_MANUAL,
NULL);
addr = nm_ip_address_new (AF_INET, "1.1.1.1", 24, &error);
g_assert_no_error (error);
nm_setting_ip_config_add_address (s_ip4, addr);
nm_ip_address_unref (addr);
g_object_set (s_ip4,
NM_SETTING_IP_CONFIG_GATEWAY, "1.1.1.254",
NM_SETTING_IP_CONFIG_NEVER_DEFAULT, FALSE,
NULL);
s_ip6 = (NMSettingIPConfig *) nm_setting_ip6_config_new ();
g_object_set (s_ip6,
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_AUTO,
NULL);
nm_connection_add_setting (con, (NMSetting *) s_ip4);
nm_connection_add_setting (con, (NMSetting *) s_ip6);
libnm/proxy: add proxy setting for non-slave connection during normalization And reject slave settings with proxies.
2016-10-04 16:44:39 +02:00
nm_connection_add_setting (con, nm_setting_proxy_new ());
libnm-core: remove gateway when never-default=yes in NMSettingIPConfig Having a gateway defined when never-default=yes causes troubles in connection matching and anyway makes no sense. If the combination is found, remove the gateway during the normalization phase. https://bugzilla.redhat.com/show_bug.cgi?id=1313091
2016-04-22 13:45:08 +02:00
nmtst_assert_connection_verifies_without_normalization (con);
g_assert_cmpstr ("1.1.1.254", ==, nm_setting_ip_config_get_gateway (s_ip4));
/* Now set never-default to TRUE and check that the gateway is
* removed during normalization
* */
g_object_set (s_ip4,
NM_SETTING_IP_CONFIG_NEVER_DEFAULT, TRUE,
NULL);
nmtst_assert_connection_verifies_after_normalization (con,
NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
nmtst_connection_normalize (con);
g_assert_cmpstr (NULL, ==, nm_setting_ip_config_get_gateway (s_ip4));
}
libnm-core: ip-config: normalize may-fail for disabled IP methods Since commit 7d1709d7f649 ("device: check may_fail when progressing to IP_CHECK") NM correctly checks the may-fail properties to decide whether a connection must fail after the completion of IP configuration. But for ipv4.method=disabled and ipv6.method=ignore the IP configuration is always considered failed and thus setting may-fail=no results in a connection that can never succeed. To prevent such wrong configuration, force may-fail to TRUE for those methods during connection normalization. https://bugzilla.redhat.com/show_bug.cgi?id=1334884
2016-07-01 17:05:42 +02:00
static void
test_connection_normalize_may_fail (void)
{
gs_unref_object NMConnection *con = NULL;
NMSettingIPConfig *s_ip4, *s_ip6;
con = nmtst_create_minimal_connection ("test2", NULL, NM_SETTING_WIRED_SETTING_NAME, NULL);
nmtst_assert_connection_verifies_and_normalizable (con);
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
g_object_set (G_OBJECT (s_ip4),
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_AUTO,
NM_SETTING_IP_CONFIG_MAY_FAIL, FALSE,
NULL);
s_ip6 = (NMSettingIPConfig *) nm_setting_ip6_config_new ();
g_object_set (s_ip6,
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_AUTO,
NM_SETTING_IP_CONFIG_MAY_FAIL, FALSE,
NULL);
nm_connection_add_setting (con, (NMSetting *) s_ip4);
nm_connection_add_setting (con, (NMSetting *) s_ip6);
libnm/proxy: add proxy setting for non-slave connection during normalization And reject slave settings with proxies.
2016-10-04 16:44:39 +02:00
nmtst_assert_connection_verifies_and_normalizable (con);
libnm-core: ip-config: normalize may-fail for disabled IP methods Since commit 7d1709d7f649 ("device: check may_fail when progressing to IP_CHECK") NM correctly checks the may-fail properties to decide whether a connection must fail after the completion of IP configuration. But for ipv4.method=disabled and ipv6.method=ignore the IP configuration is always considered failed and thus setting may-fail=no results in a connection that can never succeed. To prevent such wrong configuration, force may-fail to TRUE for those methods during connection normalization. https://bugzilla.redhat.com/show_bug.cgi?id=1334884
2016-07-01 17:05:42 +02:00
/* Now set method=disabled/ignore and check that may-fail becomes TRUE
* after normalization
* */
g_object_set (s_ip4,
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_DISABLED,
NULL);
g_object_set (s_ip6,
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_IGNORE,
NULL);
nmtst_assert_connection_verifies (con);
nmtst_connection_normalize (con);
g_assert_cmpint (nm_setting_ip_config_get_may_fail (s_ip4), ==, TRUE);
g_assert_cmpint (nm_setting_ip_config_get_may_fail (s_ip6), ==, TRUE);
}
libnm-core: drop extra IPs from shared connections during normalization The core only consider the first address for shared connections, don't pretend we accept multiple addresses. This change doesn't prevent supporting multiple addresses in the future. https://bugzilla.gnome.org/show_bug.cgi?id=763937
2016-09-12 18:51:00 +02:00
static void
test_connection_normalize_shared_addresses (void)
{
gs_unref_object NMConnection *con = NULL;
NMSettingIPConfig *s_ip4, *s_ip6;
NMIPAddress *addr;
gs_free_error GError *error = NULL;
con = nmtst_create_minimal_connection ("test1", NULL, NM_SETTING_WIRED_SETTING_NAME, NULL);
nmtst_assert_connection_verifies_and_normalizable (con);
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
g_object_set (G_OBJECT (s_ip4),
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_SHARED,
NULL);
addr = nm_ip_address_new (AF_INET, "1.1.1.1", 24, &error);
g_assert_no_error (error);
nm_setting_ip_config_add_address (s_ip4, addr);
nm_ip_address_unref (addr);
s_ip6 = (NMSettingIPConfig *) nm_setting_ip6_config_new ();
g_object_set (s_ip6,
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_AUTO,
NULL);
nm_connection_add_setting (con, (NMSetting *) s_ip4);
nm_connection_add_setting (con, (NMSetting *) s_ip6);
libnm/proxy: add proxy setting for non-slave connection during normalization And reject slave settings with proxies.
2016-10-04 16:44:39 +02:00
nmtst_assert_connection_verifies_and_normalizable (con);
libnm-core: drop extra IPs from shared connections during normalization The core only consider the first address for shared connections, don't pretend we accept multiple addresses. This change doesn't prevent supporting multiple addresses in the future. https://bugzilla.gnome.org/show_bug.cgi?id=763937
2016-09-12 18:51:00 +02:00
/* Now we add other addresses and check that they are
* removed during normalization
* */
addr = nm_ip_address_new (AF_INET, "2.2.2.2", 24, &error);
g_assert_no_error (error);
nm_setting_ip_config_add_address (s_ip4, addr);
nm_ip_address_unref (addr);
addr = nm_ip_address_new (AF_INET, "3.3.3.3", 24, &error);
g_assert_no_error (error);
nm_setting_ip_config_add_address (s_ip4, addr);
nm_ip_address_unref (addr);
nmtst_assert_connection_verifies_after_normalization (con,
NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
nmtst_connection_normalize (con);
g_assert_cmpuint (nm_setting_ip_config_get_num_addresses (s_ip4), ==, 1);
addr = nm_setting_ip_config_get_address (s_ip4, 0);
g_assert_cmpstr (nm_ip_address_get_address (addr), ==, "1.1.1.1");
}
libnm: fix normalizing and verifying OVS connections Normalizing can be complicated, as settings depend on each other and possibly conflict. That is, because verify() must exactly anticipate whether normalization will succeed and how the result will look like. That is because we only want to modify the connection, if we are sure that the result will verify. Hence, verify() and normalize() are strongly related. The implementation should not be spread out between NMSettingOvsInterface:verify(), NMSettingOvsPatch:verify() and _normalize_ovs_interface_type(). Also, add some unit-tests.
2017-10-27 14:30:18 +02:00
static void
test_connection_normalize_ovs_interface_type_system (gconstpointer test_data)
{
const guint TEST_CASE = GPOINTER_TO_UINT (test_data);
gs_unref_object NMConnection *con = NULL;
NMSettingConnection *s_con;
NMSettingOvsInterface *s_ovs_if;
con = nmtst_create_minimal_connection ("test_connection_normalize_ovs_interface_type_system",
NULL,
NM_SETTING_WIRED_SETTING_NAME, &s_con);
switch (TEST_CASE) {
case 1:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_SETTING);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_WIRED_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
s_ovs_if = nm_connection_get_setting_ovs_interface (con);
g_assert (s_ovs_if);
g_assert_cmpstr (nm_setting_ovs_interface_get_interface_type (s_ovs_if), ==, "system");
break;
case 2:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
s_ovs_if = NM_SETTING_OVS_INTERFACE (nm_setting_ovs_interface_new ());
nm_connection_add_setting (con, NM_SETTING (s_ovs_if));
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_WIRED_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
g_assert (s_ovs_if == nm_connection_get_setting_ovs_interface (con));
g_assert_cmpstr (nm_setting_ovs_interface_get_interface_type (s_ovs_if), ==, "system");
break;
case 3:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
s_ovs_if = NM_SETTING_OVS_INTERFACE (nm_setting_ovs_interface_new ());
nm_connection_add_setting (con, NM_SETTING (s_ovs_if));
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "system",
NULL);
nmtst_assert_connection_verifies_without_normalization (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_WIRED_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
break;
case 4:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
s_ovs_if = NM_SETTING_OVS_INTERFACE (nm_setting_ovs_interface_new ());
nm_connection_add_setting (con, NM_SETTING (s_ovs_if));
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "internal",
NULL);
/* the setting doesn't verify, because the interface-type must be "system". */
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
break;
case 5:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NULL);
s_ovs_if = NM_SETTING_OVS_INTERFACE (nm_setting_ovs_interface_new ());
nm_connection_add_setting (con, NM_SETTING (s_ovs_if));
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "system",
NULL);
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_WIRED_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
g_assert (s_con == nm_connection_get_setting_connection (con));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_OVS_PORT_SETTING_NAME);
break;
case 6:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_BRIDGE_SETTING_NAME,
NULL);
s_ovs_if = NM_SETTING_OVS_INTERFACE (nm_setting_ovs_interface_new ());
nm_connection_add_setting (con, NM_SETTING (s_ovs_if));
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "system",
NULL);
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
break;
case 7:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_BRIDGE_SETTING_NAME,
NULL);
nm_connection_add_setting (con, nm_setting_bridge_port_new ());
s_ovs_if = NM_SETTING_OVS_INTERFACE (nm_setting_ovs_interface_new ());
nm_connection_add_setting (con, NM_SETTING (s_ovs_if));
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "system",
NULL);
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
break;
default:
g_assert_not_reached ();
break;
}
}
static void
test_connection_normalize_ovs_interface_type_ovs_interface (gconstpointer test_data)
{
const guint TEST_CASE = GPOINTER_TO_UINT (test_data);
gs_unref_object NMConnection *con = NULL;
NMSettingConnection *s_con;
NMSettingOvsInterface *s_ovs_if;
NMSettingOvsPatch *s_ovs_patch;
NMSettingIP4Config *s_ip4;
NMSettingIP6Config *s_ip6;
con = nmtst_create_minimal_connection ("test_connection_normalize_ovs_interface_type_ovs_interface",
NULL,
NM_SETTING_OVS_INTERFACE_SETTING_NAME, &s_con);
s_ovs_if = nm_connection_get_setting_ovs_interface (con);
g_assert (s_ovs_if);
switch (TEST_CASE) {
case 1:
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
break;
case 2:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NULL);
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_IP4_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_PROXY_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
g_assert (s_con == nm_connection_get_setting_connection (con));
g_assert (s_ovs_if == nm_connection_get_setting_ovs_interface (con));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_OVS_PORT_SETTING_NAME);
g_assert_cmpstr (nm_setting_ovs_interface_get_interface_type (s_ovs_if), ==, "internal");
break;
case 3:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_IP4_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_PROXY_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
g_assert (s_con == nm_connection_get_setting_connection (con));
g_assert (s_ovs_if == nm_connection_get_setting_ovs_interface (con));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_OVS_PORT_SETTING_NAME);
g_assert_cmpstr (nm_setting_ovs_interface_get_interface_type (s_ovs_if), ==, "internal");
break;
case 4:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "internal",
NULL);
nmtst_assert_connection_verifies_after_normalization (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_IP4_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_PROXY_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
g_assert (s_con == nm_connection_get_setting_connection (con));
g_assert (s_ovs_if == nm_connection_get_setting_ovs_interface (con));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_OVS_PORT_SETTING_NAME);
g_assert_cmpstr (nm_setting_ovs_interface_get_interface_type (s_ovs_if), ==, "internal");
break;
case 5:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "internal",
NULL);
nm_connection_add_setting (con, nm_setting_ip4_config_new ());
nm_connection_add_setting (con, nm_setting_ip6_config_new ());
nm_connection_add_setting (con, nm_setting_proxy_new ());
s_ip4 = NM_SETTING_IP4_CONFIG (nm_connection_get_setting_ip4_config (con));
s_ip6 = NM_SETTING_IP6_CONFIG (nm_connection_get_setting_ip6_config (con));
g_object_set (s_ip4,
NM_SETTING_IP_CONFIG_METHOD, "auto",
NULL);
g_object_set (s_ip6,
NM_SETTING_IP_CONFIG_METHOD, "auto",
NULL);
nmtst_assert_connection_verifies_without_normalization (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_IP4_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_PROXY_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
break;
case 6:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "internal",
NULL);
nmtst_assert_connection_verifies_and_normalizable (con);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_IP4_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_PROXY_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME);
g_assert (s_con == nm_connection_get_setting_connection (con));
g_assert (s_ovs_if == nm_connection_get_setting_ovs_interface (con));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_OVS_PORT_SETTING_NAME);
g_assert_cmpstr (nm_setting_ovs_interface_get_interface_type (s_ovs_if), ==, "internal");
break;
case 7:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "system",
NULL);
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
break;
case 8:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "bogus",
NULL);
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_INVALID_PROPERTY);
break;
case 9:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "patch",
NULL);
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_SETTING);
break;
case 10:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "patch",
NULL);
nm_connection_add_setting (con, nm_setting_ovs_patch_new ());
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
break;
case 11:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NM_SETTING_CONNECTION_INTERFACE_NAME, "adsf",
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "patch",
NULL);
nm_connection_add_setting (con, nm_setting_ovs_patch_new ());
nmtst_assert_connection_unnormalizable (con, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_MISSING_PROPERTY);
break;
case 12:
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER, "master0",
NM_SETTING_CONNECTION_SLAVE_TYPE, NM_SETTING_OVS_PORT_SETTING_NAME,
NM_SETTING_CONNECTION_INTERFACE_NAME, "adsf",
NULL);
g_object_set (s_ovs_if,
NM_SETTING_OVS_INTERFACE_TYPE, "patch",
NULL);
s_ovs_patch = NM_SETTING_OVS_PATCH (nm_setting_ovs_patch_new ());
nm_connection_add_setting (con, NM_SETTING (s_ovs_patch));
g_object_set (s_ovs_patch,
NM_SETTING_OVS_PATCH_PEER, "1.2.3.4",
NULL);
nmtst_assert_connection_verifies_and_normalizable (con);
nmtst_connection_normalize (con);
nmtst_assert_connection_has_settings (con, NM_SETTING_CONNECTION_SETTING_NAME,
NM_SETTING_IP4_CONFIG_SETTING_NAME,
NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_PROXY_SETTING_NAME,
NM_SETTING_OVS_INTERFACE_SETTING_NAME,
NM_SETTING_OVS_PATCH_SETTING_NAME);
g_assert (s_con == nm_connection_get_setting_connection (con));
g_assert (s_ovs_if == nm_connection_get_setting_ovs_interface (con));
g_assert_cmpstr (nm_setting_connection_get_slave_type (s_con), ==, NM_SETTING_OVS_PORT_SETTING_NAME);
g_assert_cmpstr (nm_setting_ovs_interface_get_interface_type (s_ovs_if), ==, "patch");
break;
default:
g_assert_not_reached ();
}
}
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
static void
test_setting_ip4_gateway (void)
{
NMConnection *conn;
NMSettingIPConfig *s_ip4;
NMIPAddress *addr;
GVariant *conn_dict, *ip4_dict, *value;
GVariantIter iter;
GVariant *addr_var;
libnm/tests: fix build failure on big-endian machines `make check` fails: make check-TESTS make[4]: Entering directory `/builddir/build/BUILD/NetworkManager-1.1.0/libnm-core/tests' ... /core/general/test_setting_802_1x_changed_signal: OK /core/general/test_setting_ip4_gateway: ** ERROR:test-general.c:3652:test_setting_ip4_gateway: assertion failed (nm_setting_ip_config_get_gateway (s_ip4) == "192.168.1.1"): ("1.1.168.192" == "192.168.1.1") /bin/sh: line 5: 44957 Aborted (core dumped) ${dir}$tst
2014-12-19 12:46:03 +01:00
guint32 addr_vals_0[] = { htonl (0xc0a8010a), 0x00000018, htonl (0x00000000) };
guint32 addr_vals_1[] = { htonl (0xc0a8010b), 0x00000018, htonl (0xc0a80101) };
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
GVariantBuilder addrs_builder;
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
GError *error = NULL;
all: don't use static buffer for nm_utils_inet*_ntop() While nm_utils_inet*_ntop() accepts a %NULL buffer to fallback to a static buffer, don't do that. I find the possibility of using a static buffer here error prone and something that should be avoided. There is of course the downside, that in some cases it requires an additional line of code to allocate the buffer on the stack as auto-variable.
2018-11-26 16:49:51 +01:00
nmtst_assert_ip4_address (addr_vals_0[0], "192.168.1.10");
libnm/tests: fix build failure on big-endian machines `make check` fails: make check-TESTS make[4]: Entering directory `/builddir/build/BUILD/NetworkManager-1.1.0/libnm-core/tests' ... /core/general/test_setting_802_1x_changed_signal: OK /core/general/test_setting_ip4_gateway: ** ERROR:test-general.c:3652:test_setting_ip4_gateway: assertion failed (nm_setting_ip_config_get_gateway (s_ip4) == "192.168.1.1"): ("1.1.168.192" == "192.168.1.1") /bin/sh: line 5: 44957 Aborted (core dumped) ${dir}$tst
2014-12-19 12:46:03 +01:00
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
/* When serializing on the daemon side, ipv4.gateway is copied to the first
* entry of ipv4.addresses
*/
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
conn = nmtst_create_minimal_connection ("test_setting_ip4_gateway", NULL,
NM_SETTING_WIRED_SETTING_NAME, NULL);
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
g_object_set (s_ip4,
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_MANUAL,
NM_SETTING_IP_CONFIG_GATEWAY, "192.168.1.1",
NULL);
nm_connection_add_setting (conn, NM_SETTING (s_ip4));
addr = nm_ip_address_new (AF_INET, "192.168.1.10", 24, &error);
g_assert_no_error (error);
nm_setting_ip_config_add_address (s_ip4, addr);
nm_ip_address_unref (addr);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
_nm_utils_is_manager_process = TRUE;
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
conn_dict = nm_connection_to_dbus (conn, NM_CONNECTION_SERIALIZE_ALL);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
_nm_utils_is_manager_process = FALSE;
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
g_object_unref (conn);
ip4_dict = g_variant_lookup_value (conn_dict, NM_SETTING_IP4_CONFIG_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (ip4_dict != NULL);
value = g_variant_lookup_value (ip4_dict, NM_SETTING_IP_CONFIG_GATEWAY, G_VARIANT_TYPE_STRING);
g_assert (value != NULL);
g_assert_cmpstr (g_variant_get_string (value, NULL), ==, "192.168.1.1");
g_variant_unref (value);
value = g_variant_lookup_value (ip4_dict, NM_SETTING_IP_CONFIG_ADDRESSES, G_VARIANT_TYPE ("aau"));
g_assert (value != NULL);
g_variant_iter_init (&iter, value);
while (g_variant_iter_next (&iter, "@au", &addr_var)) {
const guint32 *addr_array;
gsize length;
addr_array = g_variant_get_fixed_array (addr_var, &length, sizeof (guint32));
g_assert_cmpint (length, ==, 3);
all: don't use static buffer for nm_utils_inet*_ntop() While nm_utils_inet*_ntop() accepts a %NULL buffer to fallback to a static buffer, don't do that. I find the possibility of using a static buffer here error prone and something that should be avoided. There is of course the downside, that in some cases it requires an additional line of code to allocate the buffer on the stack as auto-variable.
2018-11-26 16:49:51 +01:00
nmtst_assert_ip4_address (addr_array[2], "192.168.1.1");
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
g_variant_unref (addr_var);
}
g_variant_unref (value);
g_variant_unref (ip4_dict);
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
/* When deserializing an old-style connection, the first non-0 gateway in
* ipv4.addresses is copied to :gateway.
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
*/
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
NMTST_VARIANT_EDITOR (conn_dict,
NMTST_VARIANT_DROP_PROPERTY (NM_SETTING_IP4_CONFIG_SETTING_NAME,
NM_SETTING_IP_CONFIG_GATEWAY);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
NMTST_VARIANT_DROP_PROPERTY (NM_SETTING_IP4_CONFIG_SETTING_NAME,
"address-data");
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (conn_dict, &error);
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
g_assert_no_error (error);
s_ip4 = (NMSettingIPConfig *) nm_connection_get_setting_ip4_config (conn);
g_assert_cmpstr (nm_setting_ip_config_get_gateway (s_ip4), ==, "192.168.1.1");
g_object_unref (conn);
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
/* Try again with the gateway in the second address. */
g_variant_builder_init (&addrs_builder, G_VARIANT_TYPE ("aau"));
g_variant_builder_add (&addrs_builder, "@au",
g_variant_new_fixed_array (G_VARIANT_TYPE_UINT32,
addr_vals_0, 3, 4));
g_variant_builder_add (&addrs_builder, "@au",
g_variant_new_fixed_array (G_VARIANT_TYPE_UINT32,
addr_vals_1, 3, 4));
NMTST_VARIANT_EDITOR (conn_dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_IP4_CONFIG_SETTING_NAME,
"addresses", "aau", &addrs_builder);
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (conn_dict, &error);
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
g_assert_no_error (error);
g_variant_unref (conn_dict);
s_ip4 = (NMSettingIPConfig *) nm_connection_get_setting_ip4_config (conn);
g_assert_cmpstr (nm_setting_ip_config_get_gateway (s_ip4), ==, "192.168.1.1");
g_object_unref (conn);
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
}
static void
test_setting_ip6_gateway (void)
{
NMConnection *conn;
NMSettingIPConfig *s_ip6;
NMIPAddress *addr;
GVariant *conn_dict, *ip6_dict, *value;
GVariantIter iter;
GVariant *gateway_var;
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
GVariantBuilder addrs_builder;
guint8 addr_bytes_0[] = { 0xab, 0xcd, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x0a };
guint8 addr_bytes_1[] = { 0xab, 0xcd, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x0b };
guint8 gateway_bytes_1[] = { 0xab, 0xcd, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01 };
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
GError *error = NULL;
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
/* When serializing on the daemon side, ipv6.gateway is copied to the first
* entry of ipv6.addresses
*/
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
conn = nmtst_create_minimal_connection ("test_setting_ip6_gateway", NULL,
NM_SETTING_WIRED_SETTING_NAME, NULL);
s_ip6 = (NMSettingIPConfig *) nm_setting_ip6_config_new ();
g_object_set (s_ip6,
NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_MANUAL,
NM_SETTING_IP_CONFIG_GATEWAY, "abcd::1",
NULL);
nm_connection_add_setting (conn, NM_SETTING (s_ip6));
addr = nm_ip_address_new (AF_INET6, "abcd::10", 64, &error);
g_assert_no_error (error);
nm_setting_ip_config_add_address (s_ip6, addr);
nm_ip_address_unref (addr);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
_nm_utils_is_manager_process = TRUE;
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
conn_dict = nm_connection_to_dbus (conn, NM_CONNECTION_SERIALIZE_ALL);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
_nm_utils_is_manager_process = FALSE;
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
g_object_unref (conn);
ip6_dict = g_variant_lookup_value (conn_dict, NM_SETTING_IP6_CONFIG_SETTING_NAME, NM_VARIANT_TYPE_SETTING);
g_assert (ip6_dict != NULL);
value = g_variant_lookup_value (ip6_dict, NM_SETTING_IP_CONFIG_GATEWAY, G_VARIANT_TYPE_STRING);
g_assert (value != NULL);
g_assert_cmpstr (g_variant_get_string (value, NULL), ==, "abcd::1");
libnm/tests: fix memleaks in tests for valgrind
2015-02-06 13:31:06 +01:00
g_variant_unref (value);
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
value = g_variant_lookup_value (ip6_dict, NM_SETTING_IP_CONFIG_ADDRESSES, G_VARIANT_TYPE ("a(ayuay)"));
g_assert (value != NULL);
g_variant_iter_init (&iter, value);
while (g_variant_iter_next (&iter, "(@ayu@ay)", NULL, NULL, &gateway_var)) {
const guint8 *gateway_bytes;
gsize length;
gateway_bytes = g_variant_get_fixed_array (gateway_var, &length, 1);
g_assert_cmpint (length, ==, 16);
all: don't use static buffer for nm_utils_inet*_ntop() While nm_utils_inet*_ntop() accepts a %NULL buffer to fallback to a static buffer, don't do that. I find the possibility of using a static buffer here error prone and something that should be avoided. There is of course the downside, that in some cases it requires an additional line of code to allocate the buffer on the stack as auto-variable.
2018-11-26 16:49:51 +01:00
nmtst_assert_ip6_address ((struct in6_addr *) gateway_bytes, "abcd::1");
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
g_variant_unref (gateway_var);
}
g_variant_unref (value);
g_variant_unref (ip6_dict);
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
/* When deserializing an old-style connection, the first non-0 gateway in
* ipv6.addresses is copied to :gateway.
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
*/
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
NMTST_VARIANT_EDITOR (conn_dict,
NMTST_VARIANT_DROP_PROPERTY (NM_SETTING_IP6_CONFIG_SETTING_NAME,
NM_SETTING_IP_CONFIG_GATEWAY);
libnm-core, libnm, core: add AddressData and RouteData properties Add AddressData and RouteData properties to NMSettingIPConfig and NMIP[46]Config. These are like the existing "addresses" and "routes" properties, but using strings and containing additional attributes, like NMIPAddress and NMIPRoute. This only affects the D-Bus representations; there are no API changes to NMSettingIP{,4,6}Config or NMIP{4,6}Config as a result of this; the additional information is just added to the existing 'addresses' and 'routes' properties. NMSettingIP4Config and NMSettingIP6Config now always generate both old-style data ('addresses', 'address-labels', 'routes') and new-style data ('address-data', 'gateway', 'route-data') when serializing to D-Bus, for backward compatibility. When deserializing, they will fill in the 'addresses' and 'routes' properties from the new-style data if it is present (ignoring the old-style data), or from the old-style data if the new-style isn't present. The daemon-side NMIP4Config and NMIP6Config always emit changes for both 'Addresses'/'Routes' and 'AddressData'/'RouteData'. The libnm-side classes initially listen for changes on both properties, but start ignoring the 'Addresses' and 'Routes' properties once they know the daemon is also providing 'AddressData' and 'RouteData'.
2014-10-21 08:33:18 -04:00
NMTST_VARIANT_DROP_PROPERTY (NM_SETTING_IP6_CONFIG_SETTING_NAME,
"address-data");
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (conn_dict, &error);
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
g_assert_no_error (error);
s_ip6 = (NMSettingIPConfig *) nm_connection_get_setting_ip6_config (conn);
g_assert_cmpstr (nm_setting_ip_config_get_gateway (s_ip6), ==, "abcd::1");
g_object_unref (conn);
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
/* Try again with the gateway in the second address. */
g_variant_builder_init (&addrs_builder, G_VARIANT_TYPE ("a(ayuay)"));
g_variant_builder_add (&addrs_builder, "(@ayu@ay)",
g_variant_new_fixed_array (G_VARIANT_TYPE_BYTE,
addr_bytes_0, 16, 1),
64,
g_variant_new_fixed_array (G_VARIANT_TYPE_BYTE,
&in6addr_any, 16, 1));
g_variant_builder_add (&addrs_builder, "(@ayu@ay)",
g_variant_new_fixed_array (G_VARIANT_TYPE_BYTE,
addr_bytes_1, 16, 1),
64,
g_variant_new_fixed_array (G_VARIANT_TYPE_BYTE,
gateway_bytes_1, 16, 1));
NMTST_VARIANT_EDITOR (conn_dict,
NMTST_VARIANT_CHANGE_PROPERTY (NM_SETTING_IP6_CONFIG_SETTING_NAME,
"addresses", "a(ayuay)", &addrs_builder);
);
libnm-core/tests: refactor call to nm_simple_connection_new_from_dbus() No actual change, let's just not directly call nm_simple_connection_new_from_dbus(). Instead, add a wrapper to define in once place the flags we use for loading the connection.
2016-03-23 16:04:54 +01:00
conn = _connection_new_from_dbus (conn_dict, &error);
libnm-core: fix the rule for parsing 'gateway' out of 'addresses' (rh #1170199) We were always using the gateway field of the first address in ipv4.addresses / ipv6.addresses to set the gateway, but to be compatible with old behavior, we should actually be using the first non-0 gateway field (if the first one is 0).
2014-12-03 11:10:17 -05:00
g_assert_no_error (error);
g_variant_unref (conn_dict);
s_ip6 = (NMSettingIPConfig *) nm_connection_get_setting_ip6_config (conn);
g_assert_cmpstr (nm_setting_ip_config_get_gateway (s_ip6), ==, "abcd::1");
g_object_unref (conn);
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
}
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
typedef struct {
const char *str;
const guint8 expected[20];
const guint expected_len;
} HexItem;
libnm/test: test compare_property() to handle default values Add a test case to check whether compare/diff works correctly.
2015-05-20 17:53:25 +02:00
static void
test_setting_compare_default_strv (void)
{
gs_unref_object NMConnection *c1 = NULL, *c2 = NULL;
char **strv;
NMSettingIPConfig *s_ip2, *s_ip1;
gboolean compare;
GHashTable *out_settings = NULL;
c1 = nmtst_create_minimal_connection ("test_compare_default_strv", NULL,
NM_SETTING_WIRED_SETTING_NAME, NULL);
nmtst_assert_connection_verifies_and_normalizable (c1);
nmtst: add nmtst_assert_connection_verify() and don't normalize connection in assert It is ugly that nmtst_assert_connection_verifies_after_normalization() would normalize the argument and modify it. An assertion should not have side-effects.
2016-02-15 00:37:22 +01:00
nmtst_connection_normalize (c1);
libnm/test: test compare_property() to handle default values Add a test case to check whether compare/diff works correctly.
2015-05-20 17:53:25 +02:00
c2 = nm_simple_connection_new_clone (c1);
nmtst_assert_connection_verifies_without_normalization (c2);
nmtst_assert_connection_equals (c1, FALSE, c2, FALSE);
s_ip1 = nm_connection_get_setting_ip4_config (c1);
s_ip2 = nm_connection_get_setting_ip4_config (c2);
nm_setting_ip_config_clear_dns_options (s_ip2, FALSE);
g_object_get (G_OBJECT (s_ip2), NM_SETTING_IP_CONFIG_DNS_OPTIONS, &strv, NULL);
g_assert (!strv);
nmtst_assert_connection_equals (c1, FALSE, c2, FALSE);
nm_setting_ip_config_clear_dns_options (s_ip2, TRUE);
g_object_get (G_OBJECT (s_ip2), NM_SETTING_IP_CONFIG_DNS_OPTIONS, &strv, NULL);
g_assert (strv && !strv[0]);
g_strfreev (strv);
compare = nm_setting_diff ((NMSetting *) s_ip1, (NMSetting *) s_ip2, NM_SETTING_COMPARE_FLAG_EXACT, FALSE, &out_settings);
g_assert (!compare);
g_assert (out_settings);
g_assert (g_hash_table_contains (out_settings, NM_SETTING_IP_CONFIG_DNS_OPTIONS));
g_hash_table_unref (out_settings);
out_settings = NULL;
compare = nm_connection_diff (c1, c2, NM_SETTING_COMPARE_FLAG_EXACT, &out_settings);
g_assert (!compare);
g_assert (out_settings);
g_hash_table_unref (out_settings);
out_settings = NULL;
}
libnm: add NMSettingUser This only adds new API for a NMSettingUser. The setting class is still entirely unused. The point is getting the new API into 1.8.0 release of libnm. It's easier to backport the use of the API to a stable branch then backporting public API. https://bugzilla.gnome.org/show_bug.cgi?id=776276 https://bugzilla.redhat.com/show_bug.cgi?id=1421429
2017-03-24 12:41:04 +01:00
/*****************************************************************************/
static void
test_setting_user_data (void)
{
gs_unref_object NMSettingUser *s_user = NULL;
s_user = NM_SETTING_USER (nm_setting_user_new ());
}
/*****************************************************************************/
libnm: add NMSockAddrEndpoint API NMSockAddrEndpoint is an immutable structure that contains the endpoint string of a service. It also includes the (naive) parsing of the host and port/service parts. This will be used for the endpoint of WireGuard's peers. But since endpoints are not something specific to WireGuard, give it a general name (and purpose) independent from WireGuard. Essentially, this structure takes a string in a manner that libnm understands, and uses it for node and service arguments for getaddrinfo(). NMSockAddrEndpoint allows to have endpoints that are not parsable into a host and port part. That is useful because our settings need to be able to hold invalid values. That is for forward compatibility (server sends a new endpoint format) and for better error handling (have invalid settings that can be constructed without loss, but fail later during the NMSetting:verify() step).
2018-12-29 13:01:28 +01:00
typedef union {
struct sockaddr sa;
struct sockaddr_in in;
struct sockaddr_in6 in6;
} SockAddrUnion;
static void
_sock_addr_endpoint (const char *endpoint,
const char *host,
gint32 port)
{
nm_auto_unref_sockaddrendpoint NMSockAddrEndpoint *ep = NULL;
const char *s_endpoint;
const char *s_host;
gint32 s_port;
SockAddrUnion sockaddr = { };
g_assert (endpoint);
libnm-core/tests: fix "-Werror=logical-not-parentheses" warning in _sock_addr_endpoint() CC libnm-core/tests/libnm_core_tests_test_general-test-general.o In file included from ../shared/nm-default.h:280:0, from ../libnm-core/tests/test-general.c:24: ../libnm-core/tests/test-general.c: In function _sock_addr_endpoint: ../libnm-core/tests/test-general.c:5911:18: error: logical not is only applied to the left hand side of comparison [-Werror=logical-not-parentheses] g_assert (!host == (port == -1)); ^ ../shared/nm-utils/nm-macros-internal.h:1793:7: note: in definition of macro __NM_G_BOOLEAN_EXPR_IMPL if (expr) \ ^ /usr/include/glib-2.0/glib/gmacros.h:376:43: note: in expansion of macro _G_BOOLEAN_EXPR #define G_LIKELY(expr) (__builtin_expect (_G_BOOLEAN_EXPR((expr)), 1)) ^ /usr/include/glib-2.0/glib/gtestutils.h:116:49: note: in expansion of macro G_LIKELY if G_LIKELY (expr) ; else \ ^ ../libnm-core/tests/test-general.c:5911:2: note: in expansion of macro g_assert g_assert (!host == (port == -1)); ^ Fixes: 713e879d769f ('libnm: add NMSockAddrEndpoint API') (cherry picked from commit 1e8c08730fc92e4d6fc36b20e8189b42968d5008)
2019-04-18 09:45:35 +02:00
g_assert ((!host) == (port == -1));
libnm: add NMSockAddrEndpoint API NMSockAddrEndpoint is an immutable structure that contains the endpoint string of a service. It also includes the (naive) parsing of the host and port/service parts. This will be used for the endpoint of WireGuard's peers. But since endpoints are not something specific to WireGuard, give it a general name (and purpose) independent from WireGuard. Essentially, this structure takes a string in a manner that libnm understands, and uses it for node and service arguments for getaddrinfo(). NMSockAddrEndpoint allows to have endpoints that are not parsable into a host and port part. That is useful because our settings need to be able to hold invalid values. That is for forward compatibility (server sends a new endpoint format) and for better error handling (have invalid settings that can be constructed without loss, but fail later during the NMSetting:verify() step).
2018-12-29 13:01:28 +01:00
g_assert (port >= -1 && port <= G_MAXUINT16);
ep = nm_sock_addr_endpoint_new (endpoint);
g_assert (ep);
s_endpoint = nm_sock_addr_endpoint_get_endpoint (ep);
s_host = nm_sock_addr_endpoint_get_host (ep);
s_port = nm_sock_addr_endpoint_get_port (ep);
g_assert_cmpstr (endpoint, ==, s_endpoint);
g_assert_cmpstr (host, ==, s_host);
g_assert_cmpint (port, ==, s_port);
g_assert (!nm_sock_addr_endpoint_get_fixed_sockaddr (ep, &sockaddr));
if (endpoint[0] != ' ') {
gs_free char *endpoint2 = NULL;
/* also test with a leading space */
endpoint2 = g_strdup_printf (" %s", endpoint);
_sock_addr_endpoint (endpoint2, host, port);
}
if (endpoint[0] && endpoint[strlen (endpoint) - 1] != ' ') {
gs_free char *endpoint2 = NULL;
/* also test with a trailing space */
endpoint2 = g_strdup_printf ("%s ", endpoint);
_sock_addr_endpoint (endpoint2, host, port);
}
}
static void
_sock_addr_endpoint_fixed (const char *endpoint,
const char *host,
guint16 port,
guint scope_id)
{
nm_auto_unref_sockaddrendpoint NMSockAddrEndpoint *ep = NULL;
const char *s_endpoint;
const char *s_host;
gint32 s_port;
int addr_family;
NMIPAddr addrbin;
SockAddrUnion sockaddr = { };
g_assert (endpoint);
g_assert (host);
g_assert (port > 0);
if (!nm_utils_parse_inaddr_bin (AF_UNSPEC, host, &addr_family, &addrbin))
g_assert_not_reached ();
ep = nm_sock_addr_endpoint_new (endpoint);
g_assert (ep);
s_endpoint = nm_sock_addr_endpoint_get_endpoint (ep);
s_host = nm_sock_addr_endpoint_get_host (ep);
s_port = nm_sock_addr_endpoint_get_port (ep);
g_assert_cmpstr (endpoint, ==, s_endpoint);
g_assert_cmpstr (NULL, !=, s_host);
g_assert_cmpint (port, ==, s_port);
if (!nm_sock_addr_endpoint_get_fixed_sockaddr (ep, &sockaddr))
g_assert_not_reached ();
g_assert_cmpint (sockaddr.sa.sa_family, ==, addr_family);
if (addr_family == AF_INET) {
const SockAddrUnion s = {
.in = {
.sin_family = AF_INET,
.sin_addr = addrbin.addr4_struct,
.sin_port = htons (port),
},
};
g_assert_cmpint (sockaddr.in.sin_addr.s_addr, ==, addrbin.addr4);
g_assert_cmpint (sockaddr.in.sin_port, ==, htons (port));
g_assert (memcmp (&s, &sockaddr, sizeof (s.in)) == 0);
} else if (addr_family == AF_INET6) {
const SockAddrUnion s = {
.in6 = {
.sin6_family = AF_INET6,
.sin6_addr = addrbin.addr6,
.sin6_scope_id = scope_id,
.sin6_port = htons (port),
},
};
g_assert (memcmp (&sockaddr.in6.sin6_addr, &addrbin, sizeof (addrbin.addr6)) == 0);
g_assert_cmpint (sockaddr.in6.sin6_port, ==, htons (port));
g_assert_cmpint (sockaddr.in6.sin6_scope_id, ==, scope_id);
g_assert_cmpint (sockaddr.in6.sin6_flowinfo, ==, 0);
g_assert (memcmp (&s, &sockaddr, sizeof (s.in6)) == 0);
} else
g_assert_not_reached ();
}
static void
test_sock_addr_endpoint (void)
{
_sock_addr_endpoint ("", NULL, -1);
_sock_addr_endpoint (":", NULL, -1);
_sock_addr_endpoint ("a", NULL, -1);
_sock_addr_endpoint ("a:", NULL, -1);
_sock_addr_endpoint (":a", NULL, -1);
_sock_addr_endpoint ("[]:a", NULL, -1);
_sock_addr_endpoint ("[]a", NULL, -1);
_sock_addr_endpoint ("[]:", NULL, -1);
_sock_addr_endpoint ("[a]b", NULL, -1);
_sock_addr_endpoint ("[a:b", NULL, -1);
_sock_addr_endpoint ("[a[:b", NULL, -1);
_sock_addr_endpoint ("a:6", "a", 6);
_sock_addr_endpoint ("a:6", "a", 6);
_sock_addr_endpoint ("[a]:6", "a", 6);
_sock_addr_endpoint ("[a]:6", "a", 6);
_sock_addr_endpoint ("[a]:655", "a", 655);
_sock_addr_endpoint ("[ab]:][6", NULL, -1);
_sock_addr_endpoint ("[ab]:]:[6", NULL, -1);
_sock_addr_endpoint ("[a[]:b", NULL, -1);
_sock_addr_endpoint ("[192.169.6.x]:6", "192.169.6.x", 6);
_sock_addr_endpoint ("[192.169.6.x]:0", NULL, -1);
_sock_addr_endpoint ("192.169.6.7:0", NULL, -1);
_sock_addr_endpoint_fixed ("192.169.6.7:6", "192.169.6.7", 6, 0);
_sock_addr_endpoint_fixed ("[192.169.6.7]:6", "192.169.6.7", 6, 0);
_sock_addr_endpoint_fixed ("[a:b::]:6", "a:b::", 6, 0);
_sock_addr_endpoint_fixed ("[a:b::%7]:6", "a:b::", 6, 7);
_sock_addr_endpoint_fixed ("a:b::1%75:6", "a:b::1", 6, 75);
_sock_addr_endpoint_fixed ("a:b::1%0:64", "a:b::1", 64, 0);
}
/*****************************************************************************/
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
static void
test_hexstr2bin (void)
{
static const HexItem items[] = {
{ "aaBBCCddDD10496a", { 0xaa, 0xbb, 0xcc, 0xdd, 0xdd, 0x10, 0x49, 0x6a }, 8 },
{ "aa:bb:cc:dd:10:49:6a", { 0xaa, 0xbb, 0xcc, 0xdd, 0x10, 0x49, 0x6a }, 7 },
{ "0xccddeeff", { 0xcc, 0xdd, 0xee, 0xff }, 4 },
{ "1:2:66:77:80", { 0x01, 0x02, 0x66, 0x77, 0x80 }, 5 },
{ "e", { 0x0e }, 1 },
libnm: merge hwaddr_aton() and nm_utils_hexstr2bin() Have nm_utils_hexstr2bin() take over the allocated buffer via g_bytes_new_take().
2016-11-27 14:47:07 +01:00
{ "ef", { 0xef }, 1 },
{ "efa" },
{ "efad", { 0xef, 0xad }, 2 },
{ "ef:a", { 0xef, 0x0a }, 2 },
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
{ "aabb1199:" },
{ ":aabb1199" },
{ "aabb$$dd" },
{ "aab:ccc:ddd" },
{ "aab::ccc:ddd" },
};
guint i;
for (i = 0; i < G_N_ELEMENTS (items); i++) {
tests: minor code cleanup in tests Use nmtst_assert_success(), nm_auto() macros, and minor cleanups.
2018-08-22 21:35:33 +02:00
gs_unref_bytes GBytes *b = NULL;
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
b = nm_utils_hexstr2bin (items[i].str);
all: use nm_utils_gbytes_equal_mem()
2018-08-22 20:49:43 +02:00
if (items[i].expected_len)
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
g_assert (b);
all: use nm_utils_gbytes_equal_mem()
2018-08-22 20:49:43 +02:00
else
g_assert (!b);
g_assert (nm_utils_gbytes_equal_mem (b, items[i].expected, items[i].expected_len));
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
}
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
shared: add nm_strquote() util We already have nm_strquote_a(). That is useful, but uses alloca(), hence it is ill suited to be called from a macro, inside a loop, or from a function that should be inlined. Instead, add nm_strquote() that has the same purpose but writes to a provided string buffer.
2017-10-05 14:27:24 +02:00
static void
_do_strquote (const char *str, gsize buf_len, const char *expected)
{
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
char canary = (char) nmtst_get_rand_uint32 ();
shared: add nm_strquote() util We already have nm_strquote_a(). That is useful, but uses alloca(), hence it is ill suited to be called from a macro, inside a loop, or from a function that should be inlined. Instead, add nm_strquote() that has the same purpose but writes to a provided string buffer.
2017-10-05 14:27:24 +02:00
gs_free char *buf_full = g_malloc (buf_len + 2);
char *buf = &buf_full[1];
const char *b;
buf[-1] = canary;
buf[buf_len] = canary;
if (buf_len == 0) {
b = nm_strquote (NULL, 0, str);
g_assert (b == NULL);
g_assert (expected == NULL);
b = nm_strquote (buf, 0, str);
g_assert (b == buf);
} else {
b = nm_strquote (buf, buf_len, str);
g_assert (b == buf);
g_assert (strlen (b) < buf_len);
g_assert_cmpstr (expected, ==, b);
}
g_assert (buf[-1] == canary);
g_assert (buf[buf_len] == canary);
}
static void
test_nm_strquote (void)
{
_do_strquote (NULL, 0, NULL);
_do_strquote ("", 0, NULL);
_do_strquote ("a", 0, NULL);
_do_strquote ("ab", 0, NULL);
_do_strquote (NULL, 1, "");
_do_strquote (NULL, 2, "(");
_do_strquote (NULL, 3, "(n");
_do_strquote (NULL, 4, "(nu");
_do_strquote (NULL, 5, "(nul");
_do_strquote (NULL, 6, "(null");
_do_strquote (NULL, 7, "(null)");
_do_strquote (NULL, 8, "(null)");
_do_strquote (NULL, 100, "(null)");
_do_strquote ("", 1, "");
_do_strquote ("", 2, "^");
_do_strquote ("", 3, "\"\"");
_do_strquote ("", 4, "\"\"");
_do_strquote ("", 5, "\"\"");
_do_strquote ("", 100, "\"\"");
_do_strquote ("a", 1, "");
_do_strquote ("a", 2, "^");
_do_strquote ("a", 3, "\"^");
_do_strquote ("a", 4, "\"a\"");
_do_strquote ("a", 5, "\"a\"");
_do_strquote ("a", 6, "\"a\"");
_do_strquote ("a", 100, "\"a\"");
_do_strquote ("ab", 1, "");
_do_strquote ("ab", 2, "^");
_do_strquote ("ab", 3, "\"^");
_do_strquote ("ab", 4, "\"a^");
_do_strquote ("ab", 5, "\"ab\"");
_do_strquote ("ab", 6, "\"ab\"");
_do_strquote ("ab", 7, "\"ab\"");
_do_strquote ("ab", 100, "\"ab\"");
_do_strquote ("abc", 1, "");
_do_strquote ("abc", 2, "^");
_do_strquote ("abc", 3, "\"^");
_do_strquote ("abc", 4, "\"a^");
_do_strquote ("abc", 5, "\"ab^");
_do_strquote ("abc", 6, "\"abc\"");
_do_strquote ("abc", 7, "\"abc\"");
_do_strquote ("abc", 100, "\"abc\"");
}
/*****************************************************************************/
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
#define UUID_NS_ZERO "00000000-0000-0000-0000-000000000000"
libnm: implement variant3 UUIDs according to rfc4122 Compare the results: perl -e 'use UUID::Tiny ":std"; print(uuid_to_string(create_uuid(UUID_V3, UUID_NS_DNS, "test"))."\n");' python -c 'from uuid import *; print(uuid3(UUID("6ba7b810-9dad-11d1-80b4-00c04fd430c8"), "test"))'
2014-11-28 12:11:49 +01:00
#define UUID_NS_DNS "6ba7b810-9dad-11d1-80b4-00c04fd430c8"
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
#define UUID_NS_URL "6ba7b811-9dad-11d1-80b4-00c04fd430c8"
#define UUID_NS_OID "6ba7b812-9dad-11d1-80b4-00c04fd430c8"
#define UUID_NS_X500 "6ba7b814-9dad-11d1-80b4-00c04fd430c8"
libnm: implement variant3 UUIDs according to rfc4122 Compare the results: perl -e 'use UUID::Tiny ":std"; print(uuid_to_string(create_uuid(UUID_V3, UUID_NS_DNS, "test"))."\n");' python -c 'from uuid import *; print(uuid3(UUID("6ba7b810-9dad-11d1-80b4-00c04fd430c8"), "test"))'
2014-11-28 12:11:49 +01:00
libnm: add nm_utils_uuid_is_null() helper
2018-10-31 10:38:01 +01:00
static const NMUuid *
_uuid (const char *str)
{
static NMUuid u;
g_assert (str);
g_assert (_nm_utils_uuid_parse (str, &u));
return &u;
}
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
static void
libnm: add a type argument to nm_utils_uuid_generate_from_string() There are different types (variants) of UUIDs defined. Especially variants 3 and 5 are name based variants (rfc4122). The way we create our UUIDs in nm_utils_uuid_generate_from_string() however does not create them according to RFC and does not set the flags to indicate the variant. Modify the signature of nm_utils_uuid_generate_from_string() to accept a "uuid_type" argument, so that we later can add other algorithms without breaking API.
2014-11-27 18:40:18 +01:00
_test_uuid (int uuid_type, const char *expected_uuid, const char *str, gssize slen, gpointer type_args)
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
{
gs_free char *uuid_test = NULL;
libnm: add a type argument to nm_utils_uuid_generate_from_string() There are different types (variants) of UUIDs defined. Especially variants 3 and 5 are name based variants (rfc4122). The way we create our UUIDs in nm_utils_uuid_generate_from_string() however does not create them according to RFC and does not set the flags to indicate the variant. Modify the signature of nm_utils_uuid_generate_from_string() to accept a "uuid_type" argument, so that we later can add other algorithms without breaking API.
2014-11-27 18:40:18 +01:00
uuid_test = nm_utils_uuid_generate_from_string (str, slen, uuid_type, type_args);
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
g_assert (uuid_test);
g_assert (nm_utils_is_uuid (uuid_test));
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
if (!nm_streq (uuid_test, expected_uuid)) {
g_error ("UUID test failed: type=%d; text=%s, len=%lld, ns=%s, uuid=%s, expected=%s",
uuid_type,
str,
(long long) slen,
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
NM_IN_SET (uuid_type, NM_UTILS_UUID_TYPE_VERSION3,
NM_UTILS_UUID_TYPE_VERSION5)
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
? (((const char *) type_args) ?: "(all-zero)")
: (type_args ? "(unknown)" : "(null)"),
uuid_test,
expected_uuid);
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
}
libnm: add a type argument to nm_utils_uuid_generate_from_string() There are different types (variants) of UUIDs defined. Especially variants 3 and 5 are name based variants (rfc4122). The way we create our UUIDs in nm_utils_uuid_generate_from_string() however does not create them according to RFC and does not set the flags to indicate the variant. Modify the signature of nm_utils_uuid_generate_from_string() to accept a "uuid_type" argument, so that we later can add other algorithms without breaking API.
2014-11-27 18:40:18 +01:00
libnm: allow empty strings for nm_utils_uuid_generate_from_string() Since commit ef3de46c432743e2449612369d13eee66b22cb89 crypto_md5_hash() allows empty password. Also support empty strings for nm_utils_uuid_generate_from_string().
2014-12-04 16:49:47 +01:00
if (slen < 0) {
/* also test that passing slen==-1 yields the same result as passing strlen(str). */
libnm: add a type argument to nm_utils_uuid_generate_from_string() There are different types (variants) of UUIDs defined. Especially variants 3 and 5 are name based variants (rfc4122). The way we create our UUIDs in nm_utils_uuid_generate_from_string() however does not create them according to RFC and does not set the flags to indicate the variant. Modify the signature of nm_utils_uuid_generate_from_string() to accept a "uuid_type" argument, so that we later can add other algorithms without breaking API.
2014-11-27 18:40:18 +01:00
_test_uuid (uuid_type, expected_uuid, str, strlen (str), type_args);
libnm: allow empty strings for nm_utils_uuid_generate_from_string() Since commit ef3de46c432743e2449612369d13eee66b22cb89 crypto_md5_hash() allows empty password. Also support empty strings for nm_utils_uuid_generate_from_string().
2014-12-04 16:49:47 +01:00
} else if (str && slen == 0) {
/* also test if we accept NULL for slen==0 */
_test_uuid (uuid_type, expected_uuid, NULL, 0, type_args);
}
libnm: implement variant3 UUIDs according to rfc4122 Compare the results: perl -e 'use UUID::Tiny ":std"; print(uuid_to_string(create_uuid(UUID_V3, UUID_NS_DNS, "test"))."\n");' python -c 'from uuid import *; print(uuid3(UUID("6ba7b810-9dad-11d1-80b4-00c04fd430c8"), "test"))'
2014-11-28 12:11:49 +01:00
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
if ( NM_IN_SET (uuid_type, NM_UTILS_UUID_TYPE_VERSION3,
NM_UTILS_UUID_TYPE_VERSION5)
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
&& !type_args) {
/* For version3 and version5, a missing @type_args is equal to UUID_NS_ZERO */
_test_uuid (uuid_type, expected_uuid, str, slen, UUID_NS_ZERO);
libnm: allow empty strings for nm_utils_uuid_generate_from_string() Since commit ef3de46c432743e2449612369d13eee66b22cb89 crypto_md5_hash() allows empty password. Also support empty strings for nm_utils_uuid_generate_from_string().
2014-12-04 16:49:47 +01:00
}
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
}
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
typedef struct {
const char *uuid3;
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
const char *uuid5;
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
} ExpectedUuids;
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
static void
test_nm_utils_uuid_generate_from_string (void)
{
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
const ExpectedUuids zero_uuids[] = {
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
{ .uuid3 = "19826852-5007-3022-a72a-212f66e9fac3", .uuid5 = "b6c54489-38a0-5f50-a60a-fd8d76219cae", },
{ .uuid3 = "9153af2e-fc8e-34f3-9e8b-81f73b33d0cb", .uuid5 = "11116e73-1c03-5de6-9130-5f9925ae8ab4", },
{ .uuid3 = "2f06a3ae-d78d-30d7-b898-088a0e0b76f6", .uuid5 = "1087ebe8-1ef8-5d97-8873-735b4949004d", },
{ .uuid3 = "aca948e0-1468-3a51-9f2e-c688a484efd7", .uuid5 = "7e57d004-2b97-5e7a-b45f-5387367791cd", },
{ .uuid3 = "b74e537a-53e8-3808-9abd-58546a6542bd", .uuid5 = "1dd80df1-492c-5dc5-aec2-6bf0e104f923", },
{ .uuid3 = "1b00958a-7d76-3d08-8aba-c66c5828658c", .uuid5 = "f797f61e-a392-5acf-af25-b46057f1c8e8", },
{ .uuid3 = "7ba18f7d-c9cf-3b48-a89e-ad79243135cc", .uuid5 = "e02c9780-2fc5-5d57-b92f-4cc3a64bff16", },
{ .uuid3 = "9baf0978-1a60-35c5-9e9b-bec8d259fd4e", .uuid5 = "94167980-f909-527e-a4af-bc3155f586d3", },
{ .uuid3 = "588668c0-7631-39c7-9976-c7d414adf7ba", .uuid5 = "9e3eefda-b56e-56bd-8a3a-0b8009d4a536", },
{ .uuid3 = "8edb3613-9612-3b32-9dd7-0a01aa8ed453", .uuid5 = "9b75648e-d38c-54e8-adee-1fb295a079c9", },
{ .uuid3 = "f3b34394-63a5-3773-9014-1f8a50d765b8", .uuid5 = "dd56b598-9e74-58c3-b3e8-2c623780b8ed", },
{ .uuid3 = "0572965f-05b8-342b-b225-d5c29d449eee", .uuid5 = "5666449a-fb7e-55b7-ae9f-0552e6513a10", },
{ .uuid3 = "6f7177c3-77b0-3f42-82a8-7031e25fcccf", .uuid5 = "10b38db9-82fc-528e-9ddb-1f09b7dbf907", },
{ .uuid3 = "d1e0f845-bc1b-368c-b8c8-49ab0b9e486b", .uuid5 = "85492596-9468-5845-9c7f-d4ae999cb751", },
{ .uuid3 = "46371ea3-c8a3-34d8-b2cf-2fa90bda4378", .uuid5 = "22b1c0dd-aa5d-54a4-8768-5adfd0d112bd", },
{ .uuid3 = "f1e6b499-9b68-343b-a5c5-ece7acc49a68", .uuid5 = "9cc429f8-200e-52a3-9e3b-ef134afa1e29", },
{ .uuid3 = "9ed06458-c712-31dd-aba5-6cf79879fabe", .uuid5 = "3949f95c-5d76-5ee2-af60-8e2d8fcf649d", },
{ .uuid3 = "4ddd5cd7-bc83-36aa-909c-4e660f57c830", .uuid5 = "0e994a02-069b-58fb-b3a4-d7dc94e90fca", },
{ .uuid3 = "335fa537-0909-395d-a696-6f41827dcbeb", .uuid5 = "17db3a41-de9b-5c6b-904d-833943209b3c", },
{ .uuid3 = "dbd58444-05ad-3edd-adc7-4393ecbcb43c", .uuid5 = "1bd906f2-05f9-5ab5-a39a-4c17a188f886", },
{ .uuid3 = "a1c62d82-d13c-361b-8f4e-ca91bc2f7fc5", .uuid5 = "ce6550fd-95b7-57e4-9aa7-461522666be4", },
{ .uuid3 = "e943d83e-3f82-307f-81ed-b7a7bcd0743e", .uuid5 = "04aa09ee-b420-57ac-8a23-5d99907fb0a1", },
{ .uuid3 = "cabf46dd-9f09-375c-8f6e-f2a8cf114709", .uuid5 = "8ece2c62-0c31-5c55-b7c6-155381e3780e", },
{ .uuid3 = "19beddf3-f2fb-340f-96ac-4f394960b7a7", .uuid5 = "5762a9f9-9a21-59ab-b0d2-2cb90027ef7f", },
{ .uuid3 = "08d835c2-f4ca-394c-ba7f-2494d8b60c6c", .uuid5 = "23c8409d-4b5f-5b6a-b946-41e49bad6c78", },
{ .uuid3 = "3b8c6847-5331-35bf-9cd9-ced50e53cd7c", .uuid5 = "e8e396be-95d5-5569-8edc-e0b64c2b7613", },
{ .uuid3 = "e601f160-484b-3254-8f3b-0a25c7203d8a", .uuid5 = "bc8b3cbc-ad5b-5808-a1b0-e0f7a1ad68a3", },
{ .uuid3 = "e5e492ed-5349-379d-b7de-a370a51e44a3", .uuid5 = "62c5ed3f-9afa-59ad-874f-a9dd8afc69d4", },
{ .uuid3 = "c40111f6-fe97-305e-bfce-7db730c3d2ec", .uuid5 = "66877a72-7243-59ed-b9e3-b5023b6da9c2", },
{ .uuid3 = "21e18ea8-95c2-362b-9ca9-25d6a0ff2dff", .uuid5 = "49a49eee-7e86-5d66-837a-8a8810cb5562", },
{ .uuid3 = "adab623b-1343-307f-80d8-58d005376ad9", .uuid5 = "e4a2a7ed-3bf3-53cf-a2bb-154dbb39a38c", },
{ .uuid3 = "67e9fc7c-dafe-356d-ac1a-a63ce3f44813", .uuid5 = "50cacfc9-f5d2-52dd-897c-a25a0927b816", },
{ .uuid3 = "36cc7f20-126c-3e40-94e7-737ac7486547", .uuid5 = "ca629991-3f2b-5e86-9bb7-37a335f7d809", },
{ .uuid3 = "fe282996-ac5e-3d13-b478-5def30007a8e", .uuid5 = "c1adf8a7-f72a-58ae-82d5-d18807f12e2e", },
{ .uuid3 = "3bfe339c-05ae-3233-a1a5-ebf1ead589db", .uuid5 = "6120c3cd-24e1-5ce4-987b-f8bfee2e4633", },
{ .uuid3 = "d1d90bc7-da4a-3cd7-a7c8-a1a89765d8ee", .uuid5 = "433d6a26-c319-5fcf-9a30-5ec6ad59d109", },
{ .uuid3 = "10b88a02-0102-359b-81e9-7e3b0ff7d25e", .uuid5 = "77d228d9-1b96-59e2-a07e-a8fdd4f62884", },
{ .uuid3 = "7da5e4f2-6df0-3aca-a1b0-b7f8b1340e1d", .uuid5 = "698259bf-a32b-5e00-9ec6-88b12278c4ad", },
{ .uuid3 = "cbe24d98-ca20-3058-86b6-24a6b36ceff0", .uuid5 = "60dbca63-704f-5666-9f64-f4e1a630c4aa", },
{ .uuid3 = "04d84e6a-b793-3993-afbf-bae7cfc42b49", .uuid5 = "79d63ec0-a39d-557d-8299-f4c97acfadc3", },
{ .uuid3 = "fdd157d8-a537-350a-9cc9-1930e8666c63", .uuid5 = "7df7f75e-a146-5a76-828b-bac052db312b", },
{ .uuid3 = "0bea36bb-24a7-3ee6-a98d-116433c14cd4", .uuid5 = "2bcca2e9-2879-53e3-b09d-cbbfd58771b2", },
{ .uuid3 = "52b040a4-1b84-32d2-b758-f82386f7e0f0", .uuid5 = "cb7bdca3-e9f7-50cd-b72e-73cb9ff24f62", },
{ .uuid3 = "0f0a4e26-e034-3021-acf2-4e886af43092", .uuid5 = "8e428e2b-5da3-5368-b760-5ca07ccbd819", },
{ .uuid3 = "819d3cd1-afe5-3e4a-9f0c-945e25d09879", .uuid5 = "f340ef4d-139c-567a-b0fc-7c495336674e", },
{ .uuid3 = "e7df1a3b-c9f8-3e5a-88d6-ba72b2a0f27b", .uuid5 = "7e3f5fd2-3c93-58d6-9f35-6e0192445b11", },
{ .uuid3 = "0854bedf-74ba-3f2b-b823-dc2c90d27c76", .uuid5 = "bc112b6b-c5de-5ee9-b816-808792743a20", },
{ .uuid3 = "a1b8c3ba-f821-32ef-a3fd-b97b3855efa8", .uuid5 = "47f8f82d-9fcd-553c-90c5-3f3cb3ad00ad", },
{ .uuid3 = "9458f819-079b-3033-9430-ba10f576c067", .uuid5 = "bee5c091-5f01-51fa-86bb-e9488fd3b4da", },
{ .uuid3 = "8e1f240a-e386-3e00-866a-6f9da1e3503f", .uuid5 = "8ea92cea-d741-566f-a44a-d51e65b4c5e4", },
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
};
const ExpectedUuids dns_uuids[] = {
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
{ .uuid3 = "4385125b-dd1e-3025-880f-3311517cc8d5", .uuid5 = "6af613b6-569c-5c22-9c37-2ed93f31d3af", },
{ .uuid3 = "afd0b036-625a-3aa8-b639-9dc8c8fff0ff", .uuid5 = "b04965e6-a9bb-591f-8f8a-1adcb2c8dc39", },
{ .uuid3 = "9c45c2f1-1761-3daa-ad31-1ff8703ae846", .uuid5 = "4b166dbe-d99d-5091-abdd-95b83330ed3a", },
{ .uuid3 = "15e0ba07-10e4-3d7f-aaff-c00fed873c88", .uuid5 = "98123fde-012f-5ff3-8b50-881449dac91a", },
{ .uuid3 = "bc27b4db-bc0f-34f9-ae8e-4b72f2d51b60", .uuid5 = "6ed955c6-506a-5343-9be4-2c0afae02eef", },
{ .uuid3 = "7586bfed-b8b8-3bb3-9c95-09a4a79dc0f7", .uuid5 = "c8691da2-158a-5ed6-8537-0e6f140801f2", },
{ .uuid3 = "881430b6-8d28-3175-b87d-e81f2f5978c6", .uuid5 = "a6c4fc8f-6950-51de-a9ae-2c519c465071", },
{ .uuid3 = "24075675-98ae-354e-89ca-0126a9ad36e3", .uuid5 = "a9f96b98-dd44-5216-ab0d-dbfc6b262edf", },
{ .uuid3 = "2c269ea4-dbfd-32dd-9bd7-a5c22677d18b", .uuid5 = "e99caacd-6c45-5906-bd9f-b79e62f25963", },
{ .uuid3 = "44eb0948-118f-3f28-87e4-f61c8f889aba", .uuid5 = "e4d80b30-151e-51b5-9f4f-18a3b82718e6", },
{ .uuid3 = "fc72beeb-f790-36ee-a73d-33888c9d8880", .uuid5 = "0159d6c7-973f-5e7a-a9a0-d195d0ea6fe2", },
{ .uuid3 = "1e46afa2-6176-3cd3-9750-3015846723df", .uuid5 = "7fef88f7-411d-5669-b42d-bf5fc7f9b58b", },
{ .uuid3 = "0042b01d-95bd-343f-bd9f-3186bfd63508", .uuid5 = "52524d6e-10dc-5261-aa36-8b2efcbaa5f0", },
{ .uuid3 = "115ff52f-d605-3b4b-98fe-c0ea57f4930c", .uuid5 = "91c274f2-9a0d-5ce6-ac3d-7529f452df21", },
{ .uuid3 = "ed0221e8-ac7d-393b-821d-25183567885b", .uuid5 = "0ff1e264-520d-543a-87dd-181a491e667e", },
{ .uuid3 = "508ef333-85a6-314c-bcf3-17ddc32b2216", .uuid5 = "23986425-d3a5-5e13-8bab-299745777a8d", },
{ .uuid3 = "a4715ee0-524a-37cc-beb2-a0b5030757b7", .uuid5 = "c15b38c9-9a3e-543c-a703-dd742f25b4d5", },
{ .uuid3 = "d1c72756-aaec-3470-a2f2-97415f44d72f", .uuid5 = "db680066-c83d-5ed7-89a4-1d79466ea62d", },
{ .uuid3 = "7aec2f01-586e-3d53-b8f3-6cf7e6b649a4", .uuid5 = "cadb7952-2bba-5609-88d4-8e47ec4e7920", },
{ .uuid3 = "3d234b88-8d6f-319a-91ea-edb6059fc825", .uuid5 = "35140057-a2a4-5adb-a500-46f8ed8b66a9", },
{ .uuid3 = "d2568554-93ec-30c7-9e15-f383be19e5bb", .uuid5 = "66e549b7-01e2-5d07-98d5-430f74d8d3b2", },
{ .uuid3 = "800e59a7-dd0f-3114-8e58-ab7e213895ca", .uuid5 = "292c8e99-2378-55aa-83d8-350e0ac3f1cc", },
{ .uuid3 = "3b7d03f0-e067-3d72-84f4-e410ac36ef57", .uuid5 = "0e3b230a-0509-55d8-96a0-9875f387a2be", },
{ .uuid3 = "8762be68-de95-391a-94a0-c5fd0446e037", .uuid5 = "4c507660-a83b-55c0-9b2b-83eccb07723d", },
{ .uuid3 = "2bd8b4c9-01af-3cd0-aced-94ee6e2004b8", .uuid5 = "a1b9b633-da11-58be-b1a9-5cfa2848f186", },
{ .uuid3 = "a627d6a4-394a-33f5-b68e-22bfb6488d01", .uuid5 = "c2708a8b-120a-56f5-a30d-990048af87cc", },
{ .uuid3 = "6a592510-17d9-3925-b321-4a8d4927f8d0", .uuid5 = "e7263999-68b6-5a23-b530-af25b7efd632", },
{ .uuid3 = "9ee72491-59c4-333c-bb93-fe733a842fdb", .uuid5 = "ce1ae2d5-3454-5952-97ff-36ff935bcfe9", },
{ .uuid3 = "2591c62c-0a9d-3c28-97bc-fa0401556a3c", .uuid5 = "33677b87-bc8d-5ff6-9a25-fe60225e4bf0", },
{ .uuid3 = "7912be1e-4562-373b-92e2-3d6d2123bc8e", .uuid5 = "ed2305ae-e8f9-5387-b860-3d80ae6c02f7", },
{ .uuid3 = "09370cda-89a4-3a48-b592-9c0486e0d5e4", .uuid5 = "604ed872-ae2d-5d91-8e3e-572f3a3aaaa5", },
{ .uuid3 = "de5980d3-a137-373c-850b-ca3e5f100779", .uuid5 = "8f8173d9-2f8d-5636-a693-24d9f79ba651", },
{ .uuid3 = "9441501d-f633-365a-8955-9df443edc762", .uuid5 = "36eb8d4d-b854-51f1-9fdf-3735964225d5", },
{ .uuid3 = "434ada18-13ce-3c08-8b40-a1a1ae030569", .uuid5 = "3493b6ca-f84b-56a9-97cc-c0bd1c46c4c0", },
{ .uuid3 = "a13b6160-bd23-3710-a150-41d800dd30b4", .uuid5 = "f413ea13-fcd9-5b44-9d22-1fa1f7b063a5", },
{ .uuid3 = "73a67c12-c5f0-3288-ad6a-c78aea0917b0", .uuid5 = "f468d924-d23b-56c2-b90f-3d1cf4b45337", },
{ .uuid3 = "a126ee4f-a222-357d-b71b-7d3f226c559f", .uuid5 = "8828c9d6-ed76-5c09-bf64-ba9e9cd90896", },
{ .uuid3 = "48f4f36b-b015-3137-9b6e-351bb175c7f7", .uuid5 = "facb7618-55ca-5c30-9cba-fd567b6c0611", },
{ .uuid3 = "3fe8f6a3-fe4a-3487-89d6-dd06c6ad02e3", .uuid5 = "96f3de0e-6412-5434-b406-67ef3352ab85", },
{ .uuid3 = "d68fa2d4-adc9-3b20-ac77-42585cd1d59f", .uuid5 = "9ebacb89-40ab-52b3-93a2-9054611d8f55", },
{ .uuid3 = "819f86a3-31d5-3e72-a83e-142c3a3e4832", .uuid5 = "681046ff-9129-5ade-b11c-769864e02184", },
{ .uuid3 = "9957b433-ddc8-3113-a3e6-5512cf13dab1", .uuid5 = "c13d0b5d-1ca3-57b6-a23f-8586bca44928", },
{ .uuid3 = "5aab6e0c-b7d3-379c-92e3-2bfbb5572511", .uuid5 = "7c411b5e-9d3f-50b5-9c28-62096e41c4ed", },
{ .uuid3 = "11c8ff30-3a7d-3547-80a7-d61b8abeeda8", .uuid5 = "f825aafe-6696-5121-b263-6b2c408b7f43", },
{ .uuid3 = "98799b9f-1c5e-30b3-930f-e412b862cbe4", .uuid5 = "f2b4caea-61c3-5bed-8ce7-d8b9d16e129e", },
{ .uuid3 = "9bdf2544-31d8-3555-94b0-6a749118a996", .uuid5 = "3593855a-6557-5736-8cab-172c6987f949", },
{ .uuid3 = "ddcfb9b3-e990-3985-9021-546a2711e7e5", .uuid5 = "36392431-d554-5385-b876-7bc6e1cb26b3", },
{ .uuid3 = "190d7a78-1484-3136-80a6-40f28852785c", .uuid5 = "7e645493-0898-5501-8155-e8578b4f5224", },
{ .uuid3 = "6ed693e4-7dc0-3210-856b-a6eb4cc73e13", .uuid5 = "14dc6a81-0491-5683-baaf-7582a61c5798", },
{ .uuid3 = "b6a14b21-e73a-3ce2-9076-a804c434f5c6", .uuid5 = "883e0a9c-e3b3-5f9c-8073-2913cbbb99ec", },
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
};
char i_str[30];
guint i;
libnm: allow empty strings for nm_utils_uuid_generate_from_string() Since commit ef3de46c432743e2449612369d13eee66b22cb89 crypto_md5_hash() allows empty password. Also support empty strings for nm_utils_uuid_generate_from_string().
2014-12-04 16:49:47 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_LEGACY, "d41d8cd9-8f00-b204-e980-0998ecf8427e", "", -1, NULL);
libnm: add a type argument to nm_utils_uuid_generate_from_string() There are different types (variants) of UUIDs defined. Especially variants 3 and 5 are name based variants (rfc4122). The way we create our UUIDs in nm_utils_uuid_generate_from_string() however does not create them according to RFC and does not set the flags to indicate the variant. Modify the signature of nm_utils_uuid_generate_from_string() to accept a "uuid_type" argument, so that we later can add other algorithms without breaking API.
2014-11-27 18:40:18 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_LEGACY, "0cc175b9-c0f1-b6a8-31c3-99e269772661", "a", -1, NULL);
_test_uuid (NM_UTILS_UUID_TYPE_LEGACY, "098f6bcd-4621-d373-cade-4e832627b4f6", "test", -1, NULL);
libnm: allow empty strings for nm_utils_uuid_generate_from_string() Since commit ef3de46c432743e2449612369d13eee66b22cb89 crypto_md5_hash() allows empty password. Also support empty strings for nm_utils_uuid_generate_from_string().
2014-12-04 16:49:47 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_LEGACY, "70350f60-27bc-e371-3f6b-76473084309b", "a\0b", 3, NULL);
libnm: add a type argument to nm_utils_uuid_generate_from_string() There are different types (variants) of UUIDs defined. Especially variants 3 and 5 are name based variants (rfc4122). The way we create our UUIDs in nm_utils_uuid_generate_from_string() however does not create them according to RFC and does not set the flags to indicate the variant. Modify the signature of nm_utils_uuid_generate_from_string() to accept a "uuid_type" argument, so that we later can add other algorithms without breaking API.
2014-11-27 18:40:18 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_LEGACY, "59c0547b-7fe2-1c15-2cce-e328e8bf6742", "/etc/NetworkManager/system-connections/em1", -1, NULL);
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
libnm/trivial: rename uuid type VARIANT3 to VERSION3 In RFC 4122, this is called "version 3", not "variant 3". While for UUIDs there is also a concept of "variants", that is something else. Fix naming.
2018-10-31 07:32:23 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "4ae71336-e44b-39bf-b9d2-752e234818a5", "", -1, NULL);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "0531103a-d8fc-3dd4-b972-d98e4750994e", "a", -1, NULL);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "96e17d7a-ac89-38cf-95e1-bf5098da34e1", "test", -1, NULL);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "8156568e-4ae6-3f34-a93e-18e2c6cbbf78", "a\0b", 3, NULL);
libnm: implement variant3 UUIDs according to rfc4122 Compare the results: perl -e 'use UUID::Tiny ":std"; print(uuid_to_string(create_uuid(UUID_V3, UUID_NS_DNS, "test"))."\n");' python -c 'from uuid import *; print(uuid3(UUID("6ba7b810-9dad-11d1-80b4-00c04fd430c8"), "test"))'
2014-11-28 12:11:49 +01:00
libnm/trivial: rename uuid type VARIANT3 to VERSION3 In RFC 4122, this is called "version 3", not "variant 3". While for UUIDs there is also a concept of "variants", that is something else. Fix naming.
2018-10-31 07:32:23 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "c87ee674-4ddc-3efe-a74e-dfe25da5d7b3", "", -1, UUID_NS_DNS);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "4c104dd0-4821-30d5-9ce3-0e7a1f8b7c0d", "a", -1, UUID_NS_DNS);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "45a113ac-c7f2-30b0-90a5-a399ab912716", "test", -1, UUID_NS_DNS);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "002a0ada-f547-375a-bab5-896a11d1927e", "a\0b", 3, UUID_NS_DNS);
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "9a75f5f2-195e-31a9-9d07-8c18b5d3b285", "test123", -1, UUID_NS_DNS);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "ec794efe-a384-3b11-a0b6-ec8995bc6acc", "x", -1, UUID_NS_DNS);
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "a7650b9f-f19f-5300-8a13-91160ea8de2c", "a\0b", 3, NULL);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "4f3f2898-69e3-5a0d-820a-c4e87987dbce", "a", -1, UUID_NS_DNS);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "05b16a01-46c6-56dd-bd6e-c6dfb4a1427a", "x", -1, UUID_NS_DNS);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "c9ed566a-6b79-5d3a-b2b7-96a936b48cf3", "test123", -1, UUID_NS_DNS);
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
for (i = 0; i < G_N_ELEMENTS (zero_uuids); i++) {
nm_sprintf_buf (i_str, "%u", i),
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, zero_uuids[i].uuid3, i_str, -1, NULL);
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, zero_uuids[i].uuid5, i_str, -1, NULL);
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
}
for (i = 0; i < G_N_ELEMENTS (dns_uuids); i++) {
nm_sprintf_buf (i_str, "%u", i),
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, dns_uuids[i].uuid3, i_str, -1, UUID_NS_DNS);
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, dns_uuids[i].uuid5, i_str, -1, UUID_NS_DNS);
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
}
/* examples from cpython unit tests: */
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "6fa459ea-ee8a-3ca4-894e-db77e160355e", "python.org", -1, UUID_NS_DNS);
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "886313e1-3b8a-5372-9b90-0c9aee199e5d", "python.org", -1, UUID_NS_DNS);
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "9fe8e8c4-aaa8-32a9-a55c-4535a88b748d", "http://python.org/", -1, UUID_NS_URL);
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "4c565f0d-3f5a-5890-b41b-20cf47701c5e", "http://python.org/", -1, UUID_NS_URL);
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "dd1a1cef-13d5-368a-ad82-eca71acd4cd1", "1.3.6.1", -1, UUID_NS_OID);
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "1447fa61-5277-5fef-a9b3-fbc6e44f4af3", "1.3.6.1", -1, UUID_NS_OID);
libnm/tests: add more tests for generating UUIDs The expected values are checked with python's uuid module and OSSP uuid.
2018-10-31 08:25:28 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION3, "658d3002-db6b-3040-a1d1-8ddd7d189a4d", "c=ca", -1, UUID_NS_X500);
libnm: add support for SHA1 based version 5 UUIDs The entire point of using version 3/5 UUIDs is to generate stable UUIDs based on a string. It's usually important that we don't change the UUID generation algorithm later on. Since we didn't have a version 5 implementation, we would always resort to the MD5 based version 3. Version 5 is recommended by RFC 4122: o Choose either MD5 [4] or SHA-1 [8] as the hash algorithm; If backward compatibility is not an issue, SHA-1 is preferred. Add a version 5 implementation so we can use it in the future. All test values are generated with python's uuid module or OSSP uuid.
2018-10-31 08:27:41 +01:00
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "cc957dd1-a972-5349-98cd-874190002798", "c=ca", -1, UUID_NS_X500);
_test_uuid (NM_UTILS_UUID_TYPE_VERSION5, "74738ff5-5367-5958-9aee-98fffdcd1876", "www.example.org", -1, UUID_NS_DNS);
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm: move _nm_utils_uuid_generate_from_strings() from src/ to libnm-core/
2015-02-22 20:15:52 +01:00
static void
__test_uuid (const char *expected_uuid, const char *str, gssize slen, char *uuid_test)
{
g_assert (uuid_test);
g_assert (nm_utils_is_uuid (uuid_test));
if (strcmp (uuid_test, expected_uuid)) {
g_error ("UUID test failed (1): text=%s, len=%lld, expected=%s, uuid_test=%s",
str, (long long) slen, expected_uuid, uuid_test);
}
g_free (uuid_test);
libnm/trivial: rename uuid type VARIANT3 to VERSION3 In RFC 4122, this is called "version 3", not "variant 3". While for UUIDs there is also a concept of "variants", that is something else. Fix naming.
2018-10-31 07:32:23 +01:00
uuid_test = nm_utils_uuid_generate_from_string (str, slen, NM_UTILS_UUID_TYPE_VERSION3, NM_UTILS_UUID_NS);
libnm: move _nm_utils_uuid_generate_from_strings() from src/ to libnm-core/
2015-02-22 20:15:52 +01:00
g_assert (uuid_test);
g_assert (nm_utils_is_uuid (uuid_test));
if (strcmp (uuid_test, expected_uuid)) {
g_error ("UUID test failed (2): text=%s; len=%lld, expected=%s, uuid2=%s",
str, (long long) slen, expected_uuid, uuid_test);
}
g_free (uuid_test);
}
#define _test_uuid(expected_uuid, str, strlen, ...) __test_uuid (expected_uuid, str, strlen, _nm_utils_uuid_generate_from_strings(__VA_ARGS__, NULL))
static void
test_nm_utils_uuid_generate_from_strings (void)
{
libnm: avoid "-Wmissing-braces" warning for intializing NMUuid
2019-02-08 13:34:37 +01:00
const NMUuid uuid0 = { };
libnm: add nm_utils_uuid_is_null() helper
2018-10-31 10:38:01 +01:00
g_assert_cmpmem (&uuid0, sizeof (uuid0), _uuid ("00000000-0000-0000-0000-000000000000"), 16);
g_assert (nm_utils_uuid_is_null (NULL));
g_assert (nm_utils_uuid_is_null (&uuid0));
g_assert ( nm_utils_uuid_is_null (_uuid ("00000000-0000-0000-0000-000000000000")));
g_assert (!nm_utils_uuid_is_null (_uuid ("10000000-0000-0000-0000-000000000000")));
libnm: move _nm_utils_uuid_generate_from_strings() from src/ to libnm-core/
2015-02-22 20:15:52 +01:00
_test_uuid ("b07c334a-399b-32de-8d50-58e4e08f98e3", "", 0, NULL);
_test_uuid ("b8a426cb-bcb5-30a3-bd8f-6786fea72df9", "\0", 1, "");
_test_uuid ("12a4a982-7aae-39e1-951e-41aeb1250959", "a\0", 2, "a");
_test_uuid ("69e22c7e-f89f-3a43-b239-1cb52ed8db69", "aa\0", 3, "aa");
_test_uuid ("59829fd3-5ad5-3d90-a7b0-4911747e4088", "\0\0", 2, "", "");
_test_uuid ("01ad0e06-6c50-3384-8d86-ddab81421425", "a\0\0", 3, "a", "");
_test_uuid ("e1ed8647-9ed3-3ec8-8c6d-e8204524d71d", "aa\0\0", 4, "aa", "");
_test_uuid ("fb1c7cd6-275c-3489-9382-83b900da8af0", "\0a\0", 3, "", "a");
_test_uuid ("5d79494e-c4ba-31a6-80a2-d6016ccd7e17", "a\0a\0", 4, "a", "a");
_test_uuid ("fd698d86-1b60-3ebe-855f-7aada9950a8d", "aa\0a\0", 5, "aa", "a");
_test_uuid ("8c573b48-0f01-30ba-bb94-c5f59f4fe517", "\0aa\0", 4, "", "aa");
_test_uuid ("2bdd3d46-eb83-3c53-a41b-a724d04b5544", "a\0aa\0", 5, "a", "aa");
_test_uuid ("13d4b780-07c1-3ba7-b449-81c4844ef039", "aa\0aa\0", 6, "aa", "aa");
_test_uuid ("dd265bf7-c05a-3037-9939-b9629858a477", "a\0b\0", 4, "a", "b");
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
libnm: move _nm_utils_ascii_str_to_int64() from src/ to libnm-core/
2015-02-22 11:54:03 +01:00
static void
test_nm_utils_ascii_str_to_int64_check (const char *str, guint base, gint64 min,
gint64 max, gint64 fallback, int exp_errno,
gint64 exp_val)
{
gint64 v;
errno = 1;
v = _nm_utils_ascii_str_to_int64 (str, base, min, max, fallback);
g_assert_cmpint (errno, ==, exp_errno);
g_assert_cmpint (v, ==, exp_val);
}
static void
test_nm_utils_ascii_str_to_int64_do (const char *str, guint base, gint64 min,
gint64 max, gint64 fallback, int exp_errno,
gint64 exp_val)
{
const char *sign = "";
const char *val;
static const char *whitespaces[] = {
"",
" ",
"\r\n\t",
" \r\n\t ",
" \r\n\t \t\r\n\t",
NULL,
};
static const char *nulls[] = {
"",
"0",
"00",
"0000",
"0000000000000000",
"00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
"00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
NULL,
};
const char **ws_pre, **ws_post, **null;
guint i;
if (str == NULL || exp_errno != 0) {
test_nm_utils_ascii_str_to_int64_check (str, base, min, max, fallback, exp_errno, exp_val);
return;
}
if (strncmp (str, "-", 1) == 0)
sign = "-";
val = str + strlen (sign);
for (ws_pre = whitespaces; *ws_pre; ws_pre++) {
for (ws_post = whitespaces; *ws_post; ws_post++) {
for (null = nulls; *null; null++) {
for (i = 0; ; i++) {
char *s;
const char *str_base = "";
if (base == 16) {
if (i == 1)
str_base = "0x";
else if (i > 1)
break;
} else if (base == 8) {
if (i == 1)
str_base = "0";
else if (i > 1)
break;
} else if (base == 0) {
if (i > 0)
break;
/* with base==0, a leading zero would be interpreted as octal. Only test without *null */
if ((*null)[0])
break;
} else {
if (i > 0)
break;
}
s = g_strdup_printf ("%s%s%s%s%s%s", *ws_pre, sign, str_base, *null, val, *ws_post);
test_nm_utils_ascii_str_to_int64_check (s, base, min, max, fallback, exp_errno, exp_val);
g_free (s);
}
}
}
}
}
static void
test_nm_utils_ascii_str_to_int64 (void)
{
test_nm_utils_ascii_str_to_int64_do (NULL, 10, 0, 10000, -1, EINVAL, -1);
test_nm_utils_ascii_str_to_int64_do ("", 10, 0, 10000, -1, EINVAL, -1);
test_nm_utils_ascii_str_to_int64_do ("1x", 10, 0, 10000, -1, EINVAL, -1);
test_nm_utils_ascii_str_to_int64_do ("4711", 10, 0, 10000, -1, 0, 4711);
test_nm_utils_ascii_str_to_int64_do ("10000", 10, 0, 10000, -1, 0, 10000);
test_nm_utils_ascii_str_to_int64_do ("10001", 10, 0, 10000, -1, ERANGE, -1);
test_nm_utils_ascii_str_to_int64_do ("FF", 16, 0, 10000, -1, 0, 255);
test_nm_utils_ascii_str_to_int64_do ("FF", 10, 0, 10000, -2, EINVAL, -2);
test_nm_utils_ascii_str_to_int64_do ("9223372036854775807", 10, 0, G_MAXINT64, -2, 0, G_MAXINT64);
test_nm_utils_ascii_str_to_int64_do ("7FFFFFFFFFFFFFFF", 16, 0, G_MAXINT64, -2, 0, G_MAXINT64);
test_nm_utils_ascii_str_to_int64_do ("9223372036854775808", 10, 0, G_MAXINT64, -2, ERANGE, -2);
test_nm_utils_ascii_str_to_int64_do ("-9223372036854775808", 10, G_MININT64, 0, -2, 0, G_MININT64);
test_nm_utils_ascii_str_to_int64_do ("-9223372036854775808", 10, G_MININT64+1, 0, -2, ERANGE, -2);
test_nm_utils_ascii_str_to_int64_do ("-9223372036854775809", 10, G_MININT64, 0, -2, ERANGE, -2);
test_nm_utils_ascii_str_to_int64_do ("1.0", 10, 1, 1, -1, EINVAL, -1);
test_nm_utils_ascii_str_to_int64_do ("1x0", 16, -10, 10, -100, EINVAL, -100);
test_nm_utils_ascii_str_to_int64_do ("0", 16, -10, 10, -100, 0, 0);
test_nm_utils_ascii_str_to_int64_do ("10001111", 2, -1000, 1000, -100000, 0, 0x8F);
test_nm_utils_ascii_str_to_int64_do ("-10001111", 2, -1000, 1000, -100000, 0, -0x8F);
test_nm_utils_ascii_str_to_int64_do ("1111111", 2, G_MININT64, G_MAXINT64, -1, 0, 0x7F);
test_nm_utils_ascii_str_to_int64_do ("111111111111111", 2, G_MININT64, G_MAXINT64, -1, 0, 0x7FFF);
test_nm_utils_ascii_str_to_int64_do ("11111111111111111111111111111111111111111111111", 2, G_MININT64, G_MAXINT64, -1, 0, 0x7FFFFFFFFFFF);
test_nm_utils_ascii_str_to_int64_do ("111111111111111111111111111111111111111111111111111111111111111", 2, G_MININT64, G_MAXINT64, -1, 0, 0x7FFFFFFFFFFFFFFF);
test_nm_utils_ascii_str_to_int64_do ("100000000000000000000000000000000000000000000000000000000000000", 2, G_MININT64, G_MAXINT64, -1, 0, 0x4000000000000000);
test_nm_utils_ascii_str_to_int64_do ("1000000000000000000000000000000000000000000000000000000000000000", 2, G_MININT64, G_MAXINT64, -1, ERANGE, -1);
test_nm_utils_ascii_str_to_int64_do ("-100000000000000000000000000000000000000000000000000000000000000", 2, G_MININT64, G_MAXINT64, -1, 0, -0x4000000000000000);
test_nm_utils_ascii_str_to_int64_do ("111111111111111111111111111111111111111111111111111111111111111", 2, G_MININT64, G_MAXINT64, -1, 0, 0x7FFFFFFFFFFFFFFF);
test_nm_utils_ascii_str_to_int64_do ("-100000000000000000000000000000000000000000000000000000000000000", 2, G_MININT64, G_MAXINT64, -1, 0, -0x4000000000000000);
test_nm_utils_ascii_str_to_int64_do ("0x70", 10, G_MININT64, G_MAXINT64, -1, EINVAL, -1);
test_nm_utils_ascii_str_to_int64_do ("4711", 0, G_MININT64, G_MAXINT64, -1, 0, 4711);
test_nm_utils_ascii_str_to_int64_do ("04711", 0, G_MININT64, G_MAXINT64, -1, 0, 04711);
test_nm_utils_ascii_str_to_int64_do ("0x4711", 0, G_MININT64, G_MAXINT64, -1, 0, 0x4711);
test_nm_utils_ascii_str_to_int64_do ("080", 0, G_MININT64, G_MAXINT64, -1, EINVAL, -1);
test_nm_utils_ascii_str_to_int64_do ("070", 0, G_MININT64, G_MAXINT64, -1, 0, 7*8);
test_nm_utils_ascii_str_to_int64_do ("0x70", 0, G_MININT64, G_MAXINT64, -1, 0, 0x70);
libnm/tests: explicilty check behavior of _nm_utils_ascii_str_to_int64() with leading zeros
2019-11-08 07:47:29 +01:00
g_assert_cmpint (21, ==, _nm_utils_ascii_str_to_int64 ("025", 0, 0, 1000, -1));
g_assert_cmpint (21, ==, _nm_utils_ascii_str_to_int64 ("0025", 0, 0, 1000, -1));
g_assert_cmpint (25, ==, _nm_utils_ascii_str_to_int64 ("025", 10, 0, 1000, -1));
g_assert_cmpint (25, ==, _nm_utils_ascii_str_to_int64 ("0025", 10, 0, 1000, -1));
libnm: move _nm_utils_ascii_str_to_int64() from src/ to libnm-core/
2015-02-22 11:54:03 +01:00
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm: move _nm_utils_ascii_str_to_int64() from src/ to libnm-core/
2015-02-22 11:54:03 +01:00
libnm: add NMUtilsStrStrDictKey utility When having a hash-of-hashes where each hash is indexed by a name, (such as GKeyFile), you can either implement it as a hash-of-hashes or define your own version of indexes that pack both levels of names into one key. This is an implementation of such a key. Use it as: GHashTable *hash = g_hash_table_new_full (_nm_utils_strstrdictkey_hash, _nm_utils_strstrdictkey_equal, g_free, _destroy_value); and create keys via: NMUtilsStrStrDictKey *k = _nm_utils_strstrdictkey_create (s1, s2); For lookup you can use static strings (note that the static string might increase the size of the binary): g_hash_table_contains (hash, _nm_utils_strstrdictkey_static ("outer", "inner"))
2015-05-22 18:52:22 +02:00
static void
libnm/test: use (void) for empty parentheses in function prototype Fixes: ec92ecedae28c9b7e78c2de5bc378fc7afc9eaf0
2015-08-05 16:23:00 +02:00
test_nm_utils_strstrdictkey (void)
libnm: add NMUtilsStrStrDictKey utility When having a hash-of-hashes where each hash is indexed by a name, (such as GKeyFile), you can either implement it as a hash-of-hashes or define your own version of indexes that pack both levels of names into one key. This is an implementation of such a key. Use it as: GHashTable *hash = g_hash_table_new_full (_nm_utils_strstrdictkey_hash, _nm_utils_strstrdictkey_equal, g_free, _destroy_value); and create keys via: NMUtilsStrStrDictKey *k = _nm_utils_strstrdictkey_create (s1, s2); For lookup you can use static strings (note that the static string might increase the size of the binary): g_hash_table_contains (hash, _nm_utils_strstrdictkey_static ("outer", "inner"))
2015-05-22 18:52:22 +02:00
{
#define _VALUES_STATIC(_v1, _v2) { .v1 = _v1, .v2 = _v2, .v_static = _nm_utils_strstrdictkey_static (_v1, _v2), }
const struct {
const char *v1;
const char *v2;
NMUtilsStrStrDictKey *v_static;
} *val1, *val2, values[] = {
{ NULL, NULL },
{ "", NULL },
{ NULL, "" },
{ "a", NULL },
{ NULL, "a" },
_VALUES_STATIC ("", ""),
_VALUES_STATIC ("a", ""),
_VALUES_STATIC ("", "a"),
_VALUES_STATIC ("a", "b"),
};
guint i, j;
for (i = 0; i < G_N_ELEMENTS (values); i++) {
gs_free NMUtilsStrStrDictKey *key1 = NULL;
val1 = &values[i];
key1 = _nm_utils_strstrdictkey_create (val1->v1, val1->v2);
if (val1->v_static) {
g_assert (_nm_utils_strstrdictkey_equal (key1, val1->v_static));
g_assert (_nm_utils_strstrdictkey_equal (val1->v_static, key1));
g_assert_cmpint (_nm_utils_strstrdictkey_hash (key1), ==, _nm_utils_strstrdictkey_hash (val1->v_static));
}
for (j = 0; j < G_N_ELEMENTS (values); j++) {
gs_free NMUtilsStrStrDictKey *key2 = NULL;
val2 = &values[j];
key2 = _nm_utils_strstrdictkey_create (val2->v1, val2->v2);
if (i != j) {
g_assert (!_nm_utils_strstrdictkey_equal (key1, key2));
g_assert (!_nm_utils_strstrdictkey_equal (key2, key1));
}
}
}
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm: add NMUtilsStrStrDictKey utility When having a hash-of-hashes where each hash is indexed by a name, (such as GKeyFile), you can either implement it as a hash-of-hashes or define your own version of indexes that pack both levels of names into one key. This is an implementation of such a key. Use it as: GHashTable *hash = g_hash_table_new_full (_nm_utils_strstrdictkey_hash, _nm_utils_strstrdictkey_equal, g_free, _destroy_value); and create keys via: NMUtilsStrStrDictKey *k = _nm_utils_strstrdictkey_create (s1, s2); For lookup you can use static strings (note that the static string might increase the size of the binary): g_hash_table_contains (hash, _nm_utils_strstrdictkey_static ("outer", "inner"))
2015-05-22 18:52:22 +02:00
shared: add NM_PTRARRAY_LEN() utility macro I used to use g_strv_length ((char **) p) instead, but that feels ugly because it g_strv_length() is not designed to operate on arbitrary pointer arrays.
2017-04-06 09:47:05 +02:00
static guint
_g_strv_length (gconstpointer arr)
{
return arr ? g_strv_length ((char **) arr) : 0;
}
static void
test_nm_ptrarray_len (void)
{
#define _PTRARRAY_cmp(len, arr) \
G_STMT_START { \
g_assert_cmpint (len, ==, NM_PTRARRAY_LEN (arr)); \
g_assert_cmpint (len, ==, _g_strv_length (arr)); \
} G_STMT_END
#define _PTRARRAY_LEN0(T) \
G_STMT_START { \
T **vnull = NULL; \
T *const*vnull1 = NULL; \
T *const*const vnull2 = NULL; \
T *v0[] = { NULL }; \
T *const*v01 = v0; \
T *const*const v02 = v0; \
T **const v03 = v0; \
\
_PTRARRAY_cmp (0, vnull); \
_PTRARRAY_cmp (0, vnull1); \
_PTRARRAY_cmp (0, vnull2); \
_PTRARRAY_cmp (0, v0); \
_PTRARRAY_cmp (0, v01); \
_PTRARRAY_cmp (0, v02); \
_PTRARRAY_cmp (0, v03); \
} G_STMT_END
_PTRARRAY_LEN0 (char);
_PTRARRAY_LEN0 (const char);
_PTRARRAY_LEN0 (int);
_PTRARRAY_LEN0 (const int);
_PTRARRAY_LEN0 (void *);
_PTRARRAY_LEN0 (void);
_PTRARRAY_LEN0 (const void);
#define _PTRARRAY_LENn(T) \
G_STMT_START { \
T x[5] = { 0 }; \
\
T *v1[] = { &x[0], NULL }; \
T *const*v11 = v1; \
T *const*const v12 = v1; \
T **const v13 = v1; \
\
T *v2[] = { &x[0], &x[1], NULL }; \
T *const*v21 = v2; \
T *const*const v22 = v2; \
T **const v23 = v2; \
\
_PTRARRAY_cmp (1, v1); \
_PTRARRAY_cmp (1, v11); \
_PTRARRAY_cmp (1, v12); \
_PTRARRAY_cmp (1, v13); \
\
_PTRARRAY_cmp (2, v2); \
_PTRARRAY_cmp (2, v21); \
_PTRARRAY_cmp (2, v22); \
_PTRARRAY_cmp (2, v23); \
} G_STMT_END
_PTRARRAY_LENn (char);
_PTRARRAY_LENn (const char);
_PTRARRAY_LENn (int);
_PTRARRAY_LENn (const int);
_PTRARRAY_LENn (void *);
}
/*****************************************************************************/
libnm-core/test: add tests for DNS option utilities
2015-04-01 11:47:42 +02:00
static void
libnm: rename DNSOptionDesc and @dns_option_descs to have an "nm"-prefix Declarations in header files should always have an "nm" related prefix. Fixes: 019943bb5d9efea52392610b207dececafa6fdb3
2015-05-20 12:31:10 +02:00
test_nm_utils_dns_option_validate_do (char *option, gboolean ipv6, const NMUtilsDNSOptionDesc *descs,
libnm-core/test: add tests for DNS option utilities
2015-04-01 11:47:42 +02:00
gboolean exp_result, char *exp_name, gboolean exp_value)
{
char *name;
long value = 0;
gboolean result;
result = _nm_utils_dns_option_validate (option, &name, &value, ipv6, descs);
g_assert (result == exp_result);
g_assert_cmpstr (name, ==, exp_name);
g_assert (value == exp_value);
g_free (name);
}
libnm: rename DNSOptionDesc and @dns_option_descs to have an "nm"-prefix Declarations in header files should always have an "nm" related prefix. Fixes: 019943bb5d9efea52392610b207dececafa6fdb3
2015-05-20 12:31:10 +02:00
static const NMUtilsDNSOptionDesc opt_descs[] = {
libnm-core/test: add tests for DNS option utilities
2015-04-01 11:47:42 +02:00
/* name num ipv6 */
{ "opt1", FALSE, FALSE },
{ "opt2", TRUE, FALSE },
{ "opt3", FALSE, TRUE },
{ "opt4", TRUE, TRUE },
{ NULL, FALSE, FALSE }
};
static void
test_nm_utils_dns_option_validate (void)
{
/* opt ipv6 descs result name value */
test_nm_utils_dns_option_validate_do ("", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do (":", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do (":1", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do (":val", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt", FALSE, NULL, TRUE, "opt", -1);
test_nm_utils_dns_option_validate_do ("opt:", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt:12", FALSE, NULL, TRUE, "opt", 12);
test_nm_utils_dns_option_validate_do ("opt:12 ", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt:val", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt:2val", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt:2:3", FALSE, NULL, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt-6", FALSE, NULL, TRUE, "opt-6", -1);
test_nm_utils_dns_option_validate_do ("opt1", FALSE, opt_descs, TRUE, "opt1", -1);
test_nm_utils_dns_option_validate_do ("opt1", TRUE, opt_descs, TRUE, "opt1", -1);
test_nm_utils_dns_option_validate_do ("opt1:3", FALSE, opt_descs, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt2", FALSE, opt_descs, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt2:5", FALSE, opt_descs, TRUE, "opt2", 5);
test_nm_utils_dns_option_validate_do ("opt3", FALSE, opt_descs, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt3", TRUE, opt_descs, TRUE, "opt3", -1);
test_nm_utils_dns_option_validate_do ("opt4", FALSE, opt_descs, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt4", TRUE, opt_descs, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt4:40", FALSE, opt_descs, FALSE, NULL, -1);
test_nm_utils_dns_option_validate_do ("opt4:40", TRUE, opt_descs, TRUE, "opt4", 40);
}
static void
test_nm_utils_dns_option_find_idx (void)
{
GPtrArray *options;
options = g_ptr_array_new ();
g_ptr_array_add (options, "debug");
g_ptr_array_add (options, "timeout:5");
g_ptr_array_add (options, "edns0");
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, "debug"), ==, 0);
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, "debug:1"), ==, 0);
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, "timeout"), ==, 1);
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, "timeout:5"), ==, 1);
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, "timeout:2"), ==, 1);
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, "edns0"), ==, 2);
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, "rotate"), ==, -1);
g_assert_cmpint (_nm_utils_dns_option_find_idx (options, ""), ==, -1);
g_ptr_array_free (options, TRUE);
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm-core/test: add tests for DNS option utilities
2015-04-01 11:47:42 +02:00
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
static void
_json_config_check_valid (const char *conf, gboolean expected)
{
libnm/tests: fix memleak in test test_nm_utils_check_valid_json() Fixes: 82f8a54854f65d2bf648329d1d9b411de04959cb
2016-06-02 10:01:33 +02:00
gs_free_error GError *error = NULL;
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
gboolean res;
libnm: expose nm_utils_is_json_object() utility function Since we possibly already link against libjansson, we can also expose some helper utils which allows nmcli to do basic validation of JSON without requiring to duplicate the effort of using libjansson. Also, tighten up the cecks to ensure that we have a JSON object at hand. We are really interested in that and not of arrays or literals.
2016-09-23 10:30:50 +02:00
res = nm_utils_is_json_object (conf, &error);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
g_assert_cmpint (res, ==, expected);
g_assert (res || error);
}
static void
test_nm_utils_check_valid_json (void)
{
_json_config_check_valid (NULL, FALSE);
_json_config_check_valid ("", FALSE);
Revert "Makefile: rework team compilation flags" I don't think we should do this. - renamining/dropping configure options is still an annoyance, because it requires to different ./configure options depending on the version. The rename from --enable-teamctl to --enable-team might be theoretically nice, but more annoying then helpful. - There is no strict dependency between --enable-team and --enable-json-validation. At most, one could argue that when enabling the team plugin (--enable-teamctl), then libnm must also be build with --enable-json-validation. But in fact, the team plugin will happily work with a libnm that doesn't link against libjansson. That is --enable-teamctl --disable-json-validation will work in practice just fine. On the other hand, libnm is a client library to create connection profiles, fully supporting team profiles also makes sense if the actual plugin is not installed (or build). Thus, --disable-teamctl --enable-json-validation certainly makes sense. At this point, one might ask whether libnm is even still complete without libjansson. Maybe libnm should *require* --enable-json-validation. But that is not what the patch was doing, and it would also need some careful consideration before doing so. This reverts commit 9d5cd7eae8edc8c558d26f04ffd163effafe57f9.
2017-12-08 08:56:46 +01:00
#if WITH_JSON_VALIDATION
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_json_config_check_valid ("{ }", TRUE);
_json_config_check_valid ("{ \"a\" : 1 }", TRUE);
_json_config_check_valid ("{ \"a\" : }", FALSE);
#else
/* Without JSON library everything except empty string is considered valid */
_json_config_check_valid ("{ }", TRUE);
tests: fix tests without libjansson support (--enable-json-validation=no) (cherry picked from commit a5acd0bdc63449d9b499b62479f612c216cf24c0)
2017-01-17 23:44:03 +01:00
_json_config_check_valid ("{'%!-a1} ", TRUE);
_json_config_check_valid (" {'%!-a1}", TRUE);
_json_config_check_valid ("{'%!-a1", FALSE);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
#endif
}
static void
_team_config_equal_check (const char *conf1,
const char *conf2,
gboolean port_config,
gboolean expected)
{
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
nm_auto_free_team_setting NMTeamSetting *team_a = NULL;
nm_auto_free_team_setting NMTeamSetting *team_b = NULL;
gboolean is_same;
if (nmtst_get_rand_bool ())
tests: replace NMTST_SWAP() by new NM_SWAP() macro NMTST_SWAP() used memcpy() for copying the value, while NM_SWAP() uses a temporary variable with typeof(). I think the latter is preferable. Also, the macro is essentially doing the same thing.
2020-03-23 08:04:33 +01:00
NM_SWAP (conf1, conf2);
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
libnm/tests: check for identical team config in _team_config_equal_check()
2019-05-12 11:16:55 +02:00
if (!nm_streq0 (conf1, conf2)) {
_team_config_equal_check (conf1, conf1, port_config, TRUE);
_team_config_equal_check (conf2, conf2, port_config, TRUE);
}
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
team_a = nm_team_setting_new (port_config, conf1);
team_b = nm_team_setting_new (port_config, conf2);
is_same = (nm_team_setting_cmp (team_a, team_b, TRUE) == 0);
g_assert_cmpint (is_same, ==, expected);
if (nm_streq0 (conf1, conf2)) {
g_assert_cmpint (nm_team_setting_cmp (team_a, team_b, FALSE), ==, 0);
g_assert (expected);
} else
g_assert_cmpint (nm_team_setting_cmp (team_a, team_b, FALSE), !=, 0);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
}
static void
test_nm_utils_team_config_equal (void)
{
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
_team_config_equal_check ("",
"",
TRUE,
TRUE);
_team_config_equal_check ("",
" ",
TRUE,
TRUE);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_team_config_equal_check ("{}",
"{ }",
TRUE,
TRUE);
_team_config_equal_check ("{}",
"{",
TRUE,
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
TRUE);
_team_config_equal_check ("{ \"a\": 1 }",
"{ \"a\": 1 }",
TRUE,
TRUE);
_team_config_equal_check ("{ \"a\": 1 }",
"{ \"a\": 1 }",
TRUE,
TRUE);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
/* team config */
_team_config_equal_check ("{ }",
"{ \"runner\" : { \"name\" : \"random\"} }",
FALSE,
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
!WITH_JSON_VALIDATION);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_team_config_equal_check ("{ \"runner\" : { \"name\" : \"roundrobin\"} }",
"{ \"runner\" : { \"name\" : \"random\"} }",
FALSE,
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
!WITH_JSON_VALIDATION);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_team_config_equal_check ("{ \"runner\" : { \"name\" : \"random\"} }",
"{ \"runner\" : { \"name\" : \"random\"} }",
FALSE,
TRUE);
libnm-core: normalize "tx_hash" when comparing team config teamd adds the "tx_hash" property for "lacp" and "loadbalance" runners when not present. Do the same so that our original configuration matches with the one reported by teamd. https://bugzilla.redhat.com/show_bug.cgi?id=1497333
2017-10-09 20:55:38 +02:00
_team_config_equal_check ("{ \"runner\" : { \"name\" : \"loadbalance\"} }",
"{ \"runner\" : { \"name\" : \"loadbalance\"} }",
FALSE,
TRUE);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_team_config_equal_check ("{ \"runner\" : { \"name\" : \"random\"}, \"ports\" : { \"eth0\" : {} } }",
"{ \"runner\" : { \"name\" : \"random\"}, \"ports\" : { \"eth1\" : {} } }",
FALSE,
TRUE);
libnm-core: normalize "tx_hash" when comparing team config teamd adds the "tx_hash" property for "lacp" and "loadbalance" runners when not present. Do the same so that our original configuration matches with the one reported by teamd. https://bugzilla.redhat.com/show_bug.cgi?id=1497333
2017-10-09 20:55:38 +02:00
_team_config_equal_check ("{ \"runner\" : { \"name\" : \"lacp\"} }",
"{ \"runner\" : { \"name\" : \"lacp\", \"tx_hash\" : [ \"eth\", \"ipv4\", \"ipv6\" ] } }",
FALSE,
libnm/team: fix handling default values and stricter validate team config For each artifical team property we need to track whether it was explicitly set (i.e., present in JSON/GVariant or set by the user via NMSettingTeam/NMSettingTeamPort API). -- As a plus, libnm is now no longer concerned with the underling default values that teamd uses. For example, the effective default value for "notify_peers.count" depends on the selected runner. But libnm does not need to care, it only cares wheher the property is set in JSON or not. This also means that the default (e.g. as interesting to `nmcli -o con show $PROFILE`) is independent from other properties (like the runner). Also change the default value for the GObject properties of NMSettingTeam and NMSettingTeamPort to indicate the "unset" value. For most properties, the default value is a special value that is not a valid configuration itself. For some properties the default value is itself a valid value, namely, "runner.active", "runner.fast_rate", "port.sticky" and "port.prio". As far as NMTeamSetting is concerned, it distinguishes between unset value and set value (including the default value). That means, when it parses a JSON or GVariant, it will remember whether the property was present or not. When using API of NMSettingTeam/NMSettingTeamPort to set a property to the default value, it marks the property as unset. For example, setting NM_SETTING_TEAM_RUNNER_ACTIVE to TRUE (the default), means that the value will not be serialized to JSON/GVariant. For the above 4 properties (where the default value is itself a valid value) this is a limitation of libnm API, as it does not allow to explicitly set '"runner": { "active": true }'. See SET_FIELD_MODE_SET_UNLESS_DEFAULT, Note that changing the default value for properties of NMSetting is problematic, because it changes behavior for how settings are parsed from keyfile/GVariant. For team settings that's not the case, because if a JSON "config" is present, all other properties are ignore. Also, we serialize properties to JSON/GVariant depending on whether it's marked as present, and not whether the value is set to the default (_nm_team_settings_property_to_dbus()). -- While at it, sticter validate the settings. Note that if a setting is initialized from JSON, the strict validation is not not performed. That means, such a setting will always validate, regardless whether the values in JSON are invalid according to libnm. Only when using the extended properties, strict validation is turned on. Note that libnm serializes the properties to GVariant both as JSON "config" and extended properties. Since when parsing a setting from GVariant will prefer the "config" (if present), in most cases also validation is performed. Likewise, settings plugins (keyfile, ifcfg-rh) only persist the JSON config to disk. When loading a setting from file, strict validation is also not performed. The stricter validation only happens if as last operation one of the artificial properties was set, or if the setting was created from a GVariant that has no "config" field. -- This is a (another) change in behavior.
2019-05-31 10:48:47 +02:00
!WITH_JSON_VALIDATION);
libnm-core: normalize "tx_hash" when comparing team config teamd adds the "tx_hash" property for "lacp" and "loadbalance" runners when not present. Do the same so that our original configuration matches with the one reported by teamd. https://bugzilla.redhat.com/show_bug.cgi?id=1497333
2017-10-09 20:55:38 +02:00
_team_config_equal_check ("{ \"runner\" : { \"name\" : \"roundrobin\"} }",
"{ \"runner\" : { \"name\" : \"roundrobin\", \"tx_hash\" : [ \"eth\", \"ipv4\", \"ipv6\" ] } }",
FALSE,
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
!WITH_JSON_VALIDATION);
libnm-core: normalize "tx_hash" when comparing team config teamd adds the "tx_hash" property for "lacp" and "loadbalance" runners when not present. Do the same so that our original configuration matches with the one reported by teamd. https://bugzilla.redhat.com/show_bug.cgi?id=1497333
2017-10-09 20:55:38 +02:00
_team_config_equal_check ("{ \"runner\" : { \"name\" : \"lacp\"} }",
"{ \"runner\" : { \"name\" : \"lacp\", \"tx_hash\" : [ \"eth\" ] } }",
FALSE,
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
!WITH_JSON_VALIDATION);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
/* team port config */
_team_config_equal_check ("{ }",
"{ \"link_watch\" : { \"name\" : \"ethtool\"} }",
TRUE,
libnm/team: fix handling default values and stricter validate team config For each artifical team property we need to track whether it was explicitly set (i.e., present in JSON/GVariant or set by the user via NMSettingTeam/NMSettingTeamPort API). -- As a plus, libnm is now no longer concerned with the underling default values that teamd uses. For example, the effective default value for "notify_peers.count" depends on the selected runner. But libnm does not need to care, it only cares wheher the property is set in JSON or not. This also means that the default (e.g. as interesting to `nmcli -o con show $PROFILE`) is independent from other properties (like the runner). Also change the default value for the GObject properties of NMSettingTeam and NMSettingTeamPort to indicate the "unset" value. For most properties, the default value is a special value that is not a valid configuration itself. For some properties the default value is itself a valid value, namely, "runner.active", "runner.fast_rate", "port.sticky" and "port.prio". As far as NMTeamSetting is concerned, it distinguishes between unset value and set value (including the default value). That means, when it parses a JSON or GVariant, it will remember whether the property was present or not. When using API of NMSettingTeam/NMSettingTeamPort to set a property to the default value, it marks the property as unset. For example, setting NM_SETTING_TEAM_RUNNER_ACTIVE to TRUE (the default), means that the value will not be serialized to JSON/GVariant. For the above 4 properties (where the default value is itself a valid value) this is a limitation of libnm API, as it does not allow to explicitly set '"runner": { "active": true }'. See SET_FIELD_MODE_SET_UNLESS_DEFAULT, Note that changing the default value for properties of NMSetting is problematic, because it changes behavior for how settings are parsed from keyfile/GVariant. For team settings that's not the case, because if a JSON "config" is present, all other properties are ignore. Also, we serialize properties to JSON/GVariant depending on whether it's marked as present, and not whether the value is set to the default (_nm_team_settings_property_to_dbus()). -- While at it, sticter validate the settings. Note that if a setting is initialized from JSON, the strict validation is not not performed. That means, such a setting will always validate, regardless whether the values in JSON are invalid according to libnm. Only when using the extended properties, strict validation is turned on. Note that libnm serializes the properties to GVariant both as JSON "config" and extended properties. Since when parsing a setting from GVariant will prefer the "config" (if present), in most cases also validation is performed. Likewise, settings plugins (keyfile, ifcfg-rh) only persist the JSON config to disk. When loading a setting from file, strict validation is also not performed. The stricter validation only happens if as last operation one of the artificial properties was set, or if the setting was created from a GVariant that has no "config" field. -- This is a (another) change in behavior.
2019-05-31 10:48:47 +02:00
!WITH_JSON_VALIDATION);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_team_config_equal_check ("{ }",
"{ \"link_watch\" : { \"name\" : \"arp_ping\"} }",
TRUE,
libnm: rework team handling of JSON config Completely refactor the team/JSON handling in libnm's NMSettingTeam and NMSettingTeamPort. - team handling was added as rh#1398925. The goal is to have a more convenient way to set properties than constructing JSON. This requires libnm to implement the hard task of parsing JSON (and exposing well-understood properties) and generating JSON (based on these "artificial" properties). But not only libnm. In particular nmcli and the D-Bus API must make this "simpler" API accessible. - since NMSettingTeam and NMSettingTeamPort are conceptually the same, add "libnm-core/nm-team-utils.h" and NMTeamSetting that tries to handle the similar code side-by-sdie. The setting classes now just delegate for everything to NMTeamSetting. - Previously, there was a very fuzzy understanding of the provided JSON config. Tighten that up, when setting a JSON config it regenerates/parses all other properties and tries to make the best of it. When modifying any abstraction property, the entire JSON config gets regenerated. In particular, don't try to merge existing JSON config with the new fields. If the user uses the abstraction API, then the entire JSON gets replaced. For example note that nm_setting_team_add_link_watcher() would not be reflected in the JSON config (a bug). That only accidentally worked because client would serializing the changed link watcher to GVariant/D-Bus, then NetworkManager would set it via g_object_set(), which would renerate the JSON, and finally persist it to disk. But as far as libnm is concerned, nm_setting_team_add_link_watcher() would bring the settings instance in an inconsistent state where JSON and the link watcher property disagree. Setting any property must immediately update both the JSON and the abstraction API. - when constucting a team setting from D-Bus, we would previously parse both "config" and abstraction properties. That is wrong. Since our settings plugins only support JSON, all information must be present in the JSON config anyway. So, when "config" is present, only the JSON must be parsed. In the best case, the other information is redudant and contributes nothing. In the worse case, they information differs (which might happen if the client version differs from the server version). As the settings plugin only supports JSON, it's wrong to consider redundant, differing information from D-Bus. - we now only convert string to JSON or back when needed. Previously, setting a property resulted in parsing several JSON multiple times (per property). All operations should now scale well and be reasonably efficient. - also the property-changed signals are now handled correctly. Since NMTeamSetting knows the current state of all attributes, it can emit the exact property changed signals for what changed. - we no longer use libjansson to generate the JSON. JSON is supposed to be a machine readable exchange format, hence a major goal is to be easily handled by applications. While parsing JSON is not so trivial, writing a well-known set of values to JSON is. The advantage is that when you build libnm without libjansson support, then we still can convert the artificial properties to JSON. - Requiring libjansson in libnm is a burden, because most of the time it is not needed (as most users don't create team configurations). With this change we only require it to parse the team settings (no longer to write them). It should be reasonably simple to use a more minimalistic JSON parser that is sufficient for us, so that we can get rid of the libjansson dependency (for libnm). This also avoids the pain that we have due to the symbol collision of libjansson and libjson-glib. https://bugzilla.redhat.com/show_bug.cgi?id=1691619
2019-05-06 12:36:41 +02:00
TRUE);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_team_config_equal_check ("{ \"link_watch\" : { \"name\" : \"ethtool\"} }",
"{ \"link_watch\" : { \"name\" : \"arp_ping\"} }",
TRUE,
libnm/team: fix handling default values and stricter validate team config For each artifical team property we need to track whether it was explicitly set (i.e., present in JSON/GVariant or set by the user via NMSettingTeam/NMSettingTeamPort API). -- As a plus, libnm is now no longer concerned with the underling default values that teamd uses. For example, the effective default value for "notify_peers.count" depends on the selected runner. But libnm does not need to care, it only cares wheher the property is set in JSON or not. This also means that the default (e.g. as interesting to `nmcli -o con show $PROFILE`) is independent from other properties (like the runner). Also change the default value for the GObject properties of NMSettingTeam and NMSettingTeamPort to indicate the "unset" value. For most properties, the default value is a special value that is not a valid configuration itself. For some properties the default value is itself a valid value, namely, "runner.active", "runner.fast_rate", "port.sticky" and "port.prio". As far as NMTeamSetting is concerned, it distinguishes between unset value and set value (including the default value). That means, when it parses a JSON or GVariant, it will remember whether the property was present or not. When using API of NMSettingTeam/NMSettingTeamPort to set a property to the default value, it marks the property as unset. For example, setting NM_SETTING_TEAM_RUNNER_ACTIVE to TRUE (the default), means that the value will not be serialized to JSON/GVariant. For the above 4 properties (where the default value is itself a valid value) this is a limitation of libnm API, as it does not allow to explicitly set '"runner": { "active": true }'. See SET_FIELD_MODE_SET_UNLESS_DEFAULT, Note that changing the default value for properties of NMSetting is problematic, because it changes behavior for how settings are parsed from keyfile/GVariant. For team settings that's not the case, because if a JSON "config" is present, all other properties are ignore. Also, we serialize properties to JSON/GVariant depending on whether it's marked as present, and not whether the value is set to the default (_nm_team_settings_property_to_dbus()). -- While at it, sticter validate the settings. Note that if a setting is initialized from JSON, the strict validation is not not performed. That means, such a setting will always validate, regardless whether the values in JSON are invalid according to libnm. Only when using the extended properties, strict validation is turned on. Note that libnm serializes the properties to GVariant both as JSON "config" and extended properties. Since when parsing a setting from GVariant will prefer the "config" (if present), in most cases also validation is performed. Likewise, settings plugins (keyfile, ifcfg-rh) only persist the JSON config to disk. When loading a setting from file, strict validation is also not performed. The stricter validation only happens if as last operation one of the artificial properties was set, or if the setting was created from a GVariant that has no "config" field. -- This is a (another) change in behavior.
2019-05-31 10:48:47 +02:00
!WITH_JSON_VALIDATION);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
_team_config_equal_check ("{ \"link_watch\" : { \"name\" : \"arp_ping\"} }",
"{ \"link_watch\" : { \"name\" : \"arp_ping\"} }",
TRUE,
TRUE);
_team_config_equal_check ("{ \"link_watch\" : { \"name\" : \"arp_ping\"}, \"ports\" : { \"eth0\" : {} } }",
"{ \"link_watch\" : { \"name\" : \"arp_ping\"}, \"ports\" : { \"eth1\" : {} } }",
TRUE,
TRUE);
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
enum TEST_IS_POWER_OF_TWP_ENUM_SIGNED {
_DUMMY_1 = -1,
};
enum TEST_IS_POWER_OF_TWP_ENUM_UNSIGNED {
_DUMMY_2,
};
enum TEST_IS_POWER_OF_TWP_ENUM_SIGNED_64 {
_DUMMY_3 = (1LL << 40),
};
enum TEST_IS_POWER_OF_TWP_ENUM_UNSIGNED_64 {
_DUMMY_4a = -1,
_DUMMY_4b = (1LL << 40),
};
#define test_nm_utils_is_power_of_two_do(type, x, expect) \
G_STMT_START { \
typeof (x) x1 = (x); \
type x2 = (type) x1; \
\
shared: refactor nm_utils_is_power_of_two() to return false for 0 Returning TRUE for zero makes no sense. Obviously, zero is not a power of two. Also, the function is used to check whether a number has only one bit (flag) set, so, an alternative name would be "has-one-bit-set", which also should return FALSE for zero. All callers didn't really care for the previous meaning "has-at-most-one-bit-set". This also avoids the issue of checking (x >= 0), which causes -Wtype-limits warnings for unsigned types. Which was avoided by doing (x == 0 || x > 0), which caused -Wlogical-op warning, which then was avoided (x == 0 || (x > 0 && 1)). Just don't.
2017-05-22 13:28:35 +02:00
g_assert_cmpint (expect, ==, nm_utils_is_power_of_two (x1)); \
if ( ((typeof (x1)) x2) == x1 \
&& ((typeof (x2)) x1) == x2 \
&& x2 > 0) { \
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
/* x2 equals @x, and is positive. Compare to @expect */ \
g_assert_cmpint (expect, ==, nm_utils_is_power_of_two (x2)); \
shared: refactor nm_utils_is_power_of_two() to return false for 0 Returning TRUE for zero makes no sense. Obviously, zero is not a power of two. Also, the function is used to check whether a number has only one bit (flag) set, so, an alternative name would be "has-one-bit-set", which also should return FALSE for zero. All callers didn't really care for the previous meaning "has-at-most-one-bit-set". This also avoids the issue of checking (x >= 0), which causes -Wtype-limits warnings for unsigned types. Which was avoided by doing (x == 0 || x > 0), which caused -Wlogical-op warning, which then was avoided (x == 0 || (x > 0 && 1)). Just don't.
2017-05-22 13:28:35 +02:00
} else if (!(x2 > 0)) { \
/* a non positive value is always FALSE. */ \
g_assert_cmpint (FALSE, ==, nm_utils_is_power_of_two (x2)); \
} \
if (x2) { \
x2 = -x2; \
if (!(x2 > 0)) { \
/* for negative values, we return FALSE. */ \
g_assert_cmpint (FALSE, ==, nm_utils_is_power_of_two (x2)); \
} \
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
} \
} G_STMT_END
static void
all: fix additional compiler warningis about function declarations warning: function declaration isn’t a prototype [-Wstrict-prototypes] In C function() and function(void) are two different prototypes (as opposed to C++). function() accepts an arbitrary number of arguments function(void) accepts zero arguments
2015-06-19 13:06:37 +02:00
test_nm_utils_is_power_of_two (void)
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
{
guint64 xyes, xno;
all: don't use gchar/gshort/gint/glong but C types We commonly don't use the glib typedefs for char/short/int/long, but their C types directly. $ git grep '\<g\(char\|short\|int\|long\|float\|double\)\>' | wc -l 587 $ git grep '\<\(char\|short\|int\|long\|float\|double\)\>' | wc -l 21114 One could argue that using the glib typedefs is preferable in public API (of our glib based libnm library) or where it clearly is related to glib, like during g_object_set (obj, PROPERTY, (gint) value, NULL); However, that argument does not seem strong, because in practice we don't follow that argument today, and seldomly use the glib typedefs. Also, the style guide for this would be hard to formalize, because "using them where clearly related to a glib" is a very loose suggestion. Also note that glib typedefs will always just be typedefs of the underlying C types. There is no danger of glib changing the meaning of these typedefs (because that would be a major API break of glib). A simple style guide is instead: don't use these typedefs. No manual actions, I only ran the bash script: FILES=($(git ls-files '*.[hc]')) sed -i \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\>\( [^ ]\)/\1\2/g' \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\> /\1 /g' \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\>/\1/g' \ "${FILES[@]}"
2018-07-11 07:40:19 +02:00
int i, j;
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
GRand *rand = nmtst_get_rand ();
int numbits;
for (i = -1; i < 64; i++) {
/* find a (positive) x which is a power of two. */
if (i == -1)
xyes = 0;
else {
test: fix undefined behavior shifting signed integer in test
2017-04-17 13:19:36 +02:00
xyes = (((guint64) 1) << i);
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
g_assert (xyes != 0);
}
xno = xyes;
if (xyes != 0) {
again:
/* Find another @xno, that is not a power of two. Do that,
* by randomly setting bits. */
numbits = g_rand_int_range (rand, 1, 65);
while (xno != ~((guint64) 0) && numbits > 0) {
test: fix undefined behavior shifting signed integer in test
2017-04-17 13:19:36 +02:00
guint64 v = (((guint64) 1) << g_rand_int_range (rand, 0, 64));
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
if ((xno | v) != xno) {
xno |= v;
--numbits;
}
}
if (xno == xyes)
goto again;
}
for (j = 0; j < 2; j++) {
gboolean expect = j == 0;
guint64 x = expect ? xyes : xno;
shared: refactor nm_utils_is_power_of_two() to return false for 0 Returning TRUE for zero makes no sense. Obviously, zero is not a power of two. Also, the function is used to check whether a number has only one bit (flag) set, so, an alternative name would be "has-one-bit-set", which also should return FALSE for zero. All callers didn't really care for the previous meaning "has-at-most-one-bit-set". This also avoids the issue of checking (x >= 0), which causes -Wtype-limits warnings for unsigned types. Which was avoided by doing (x == 0 || x > 0), which caused -Wlogical-op warning, which then was avoided (x == 0 || (x > 0 && 1)). Just don't.
2017-05-22 13:28:35 +02:00
if (expect && xyes == 0)
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
continue;
/* check if @x is as @expect, when casted to a certain data type. */
test_nm_utils_is_power_of_two_do (gint8, x, expect);
test_nm_utils_is_power_of_two_do (guint8, x, expect);
test_nm_utils_is_power_of_two_do (gint16, x, expect);
test_nm_utils_is_power_of_two_do (guint16, x, expect);
test_nm_utils_is_power_of_two_do (gint32, x, expect);
test_nm_utils_is_power_of_two_do (guint32, x, expect);
test_nm_utils_is_power_of_two_do (gint64, x, expect);
test_nm_utils_is_power_of_two_do (guint64, x, expect);
test_nm_utils_is_power_of_two_do (char, x, expect);
test_nm_utils_is_power_of_two_do (unsigned char, x, expect);
test_nm_utils_is_power_of_two_do (signed char, x, expect);
test_nm_utils_is_power_of_two_do (enum TEST_IS_POWER_OF_TWP_ENUM_SIGNED, x, expect);
test_nm_utils_is_power_of_two_do (enum TEST_IS_POWER_OF_TWP_ENUM_UNSIGNED, x, expect);
test_nm_utils_is_power_of_two_do (enum TEST_IS_POWER_OF_TWP_ENUM_SIGNED_64, x, expect);
test_nm_utils_is_power_of_two_do (enum TEST_IS_POWER_OF_TWP_ENUM_UNSIGNED_64, x, expect);
}
}
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
static int
_test_find_binary_search_cmp (gconstpointer a, gconstpointer b, gpointer dummy)
{
int ia, ib;
ia = GPOINTER_TO_INT (a);
ib = GPOINTER_TO_INT (b);
if (ia == ib)
return 0;
if (ia < ib)
return -1;
return 1;
}
static void
_test_find_binary_search_do (const int *array, gsize len)
{
gsize i;
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
gssize idx, idx_first, idx_last;
core: const arguments for _nm_utils_ptrarray_find_*() functions
2016-09-23 15:03:41 +02:00
gs_free gconstpointer *parray = g_new (gconstpointer, len);
const int NEEDLE = 0;
gconstpointer pneedle = GINT_TO_POINTER (NEEDLE);
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
gssize expected_result;
for (i = 0; i < len; i++)
parray[i] = GINT_TO_POINTER (array[i]);
expected_result = _nm_utils_ptrarray_find_first (parray, len, pneedle);
shared: move nm_utils_ptrarray_find_binary_search() to shared utils
2018-08-01 11:03:02 +02:00
idx = nm_utils_ptrarray_find_binary_search (parray, len, pneedle, _test_find_binary_search_cmp, NULL, &idx_first, &idx_last);
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
if (expected_result >= 0) {
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
g_assert_cmpint (expected_result, ==, idx);
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
} else {
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
gssize idx2 = ~idx;
g_assert_cmpint (idx, <, 0);
g_assert (idx2 >= 0);
g_assert (idx2 <= len);
g_assert (idx2 - 1 < 0 || _test_find_binary_search_cmp (parray[idx2 - 1], pneedle, NULL) < 0);
g_assert (idx2 >= len || _test_find_binary_search_cmp (parray[idx2], pneedle, NULL) > 0);
}
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
g_assert_cmpint (idx, ==, idx_first);
g_assert_cmpint (idx, ==, idx_last);
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
for (i = 0; i < len; i++) {
int cmp;
cmp = _test_find_binary_search_cmp (parray[i], pneedle, NULL);
if (cmp == 0) {
g_assert (pneedle == parray[i]);
g_assert (idx >= 0);
g_assert (i == idx);
} else {
g_assert (pneedle != parray[i]);
if (cmp < 0) {
if (idx < 0)
g_assert (i < ~idx);
else
g_assert (i < idx);
} else {
if (idx < 0)
g_assert (i >= ~idx);
else
g_assert (i >= idx);
}
}
}
}
core: add _nm_utils_array_find_binary_search() Also add nm_cmp_uint32_p_with_data(). Will be used later.
2016-09-23 13:37:22 +02:00
static void
_test_find_binary_search_do_uint32 (const int *int_array, gsize len)
{
gssize idx;
const int OFFSET = 100;
const int NEEDLE = 0 + OFFSET;
gssize expected_result = -1;
core: don't use variable length arrays Let's compile with -Wvla, so that G_STATIC_ASSERT() is really static. But then we cannot use variable length arrays.
2017-12-11 18:21:34 +01:00
guint32 array[30];
g_assert (len <= G_N_ELEMENTS (array));
core: add _nm_utils_array_find_binary_search() Also add nm_cmp_uint32_p_with_data(). Will be used later.
2016-09-23 13:37:22 +02:00
/* the test data has negative values. Shift them... */
for (idx = 0; idx < len; idx++) {
int v = int_array[idx];
g_assert (v > -OFFSET);
g_assert (v < OFFSET);
g_assert (idx == 0 || v > int_array[idx - 1]);
array[idx] = (guint32) (int_array[idx] + OFFSET);
if (array[idx] == NEEDLE)
expected_result = idx;
}
shared: move nm_utils_array_find_binary_search() to shared utils
2018-07-28 12:26:37 +02:00
idx = nm_utils_array_find_binary_search (array,
sizeof (guint32),
len,
&NEEDLE,
nm_cmp_uint32_p_with_data,
NULL);
core: add _nm_utils_array_find_binary_search() Also add nm_cmp_uint32_p_with_data(). Will be used later.
2016-09-23 13:37:22 +02:00
if (expected_result >= 0)
g_assert_cmpint (expected_result, ==, idx);
else {
gssize idx2 = ~idx;
g_assert_cmpint (idx, <, 0);
g_assert (idx2 >= 0);
g_assert (idx2 <= len);
g_assert (idx2 - 1 < 0 || array[idx2 - 1] < NEEDLE);
g_assert (idx2 >= len || array[idx2] > NEEDLE);
}
}
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
#define test_find_binary_search_do(...) \
G_STMT_START { \
core: const arguments for _nm_utils_ptrarray_find_*() functions
2016-09-23 15:03:41 +02:00
const int _array[] = { __VA_ARGS__ }; \
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
_test_find_binary_search_do (_array, G_N_ELEMENTS (_array)); \
core: add _nm_utils_array_find_binary_search() Also add nm_cmp_uint32_p_with_data(). Will be used later.
2016-09-23 13:37:22 +02:00
_test_find_binary_search_do_uint32 (_array, G_N_ELEMENTS (_array)); \
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
} G_STMT_END
static void
test_nm_utils_ptrarray_find_binary_search (void)
{
test_find_binary_search_do ( 0);
test_find_binary_search_do ( -1, 0);
test_find_binary_search_do ( -2, -1, 0);
test_find_binary_search_do (-3, -2, -1, 0);
test_find_binary_search_do ( 0, 1);
test_find_binary_search_do ( 0, 1, 2);
test_find_binary_search_do ( -1, 0, 1, 2);
test_find_binary_search_do ( -2, -1, 0, 1, 2);
test_find_binary_search_do (-3, -2, -1, 0, 1, 2);
test_find_binary_search_do (-3, -2, -1, 0, 1, 2);
test_find_binary_search_do (-3, -2, -1, 0, 1, 2, 3);
test_find_binary_search_do (-3, -2, -1, 0, 1, 2, 3, 4);
test_find_binary_search_do ( -1);
test_find_binary_search_do ( -2, -1);
test_find_binary_search_do (-3, -2, -1);
test_find_binary_search_do ( 1);
test_find_binary_search_do ( 1, 2);
test_find_binary_search_do ( -1, 1, 2);
test_find_binary_search_do ( -2, -1, 1, 2);
test_find_binary_search_do (-3, -2, -1, 1, 2);
test_find_binary_search_do (-3, -2, -1, 1, 2);
test_find_binary_search_do (-3, -2, -1, 1, 2, 3);
test_find_binary_search_do (-3, -2, -1, 1, 2, 3, 4);
}
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
/*****************************************************************************/
#define BIN_SEARCH_W_DUPS_LEN 100
#define BIN_SEARCH_W_DUPS_JITTER 10
static int
_test_bin_search2_cmp (gconstpointer pa,
gconstpointer pb,
gpointer user_data)
{
int a = GPOINTER_TO_INT (pa);
int b = GPOINTER_TO_INT (pb);
g_assert (a >= 0 && a <= BIN_SEARCH_W_DUPS_LEN + BIN_SEARCH_W_DUPS_JITTER);
g_assert (b >= 0 && b <= BIN_SEARCH_W_DUPS_LEN + BIN_SEARCH_W_DUPS_JITTER);
NM_CMP_DIRECT (a, b);
return 0;
}
static int
_test_bin_search2_cmp_p (gconstpointer pa,
gconstpointer pb,
gpointer user_data)
{
return _test_bin_search2_cmp (*((gpointer *) pa), *((gpointer *) pb), NULL);
}
static void
test_nm_utils_ptrarray_find_binary_search_with_duplicates (void)
{
gssize idx, idx2, idx_first2, idx_first, idx_last;
int i_test, i_len, i;
gssize j;
gconstpointer arr[BIN_SEARCH_W_DUPS_LEN];
const int N_TEST = 10;
for (i_test = 0; i_test < N_TEST; i_test++) {
for (i_len = 0; i_len < BIN_SEARCH_W_DUPS_LEN; i_len++) {
/* fill with random numbers... surely there are some duplicates
* there... or maybe even there are none... */
for (i = 0; i < i_len; i++)
tests/trivial: rename nmtst_get_rand_int() to nmtst_get_rand_uint32() nmtst_get_rand_int() was originally named that way, because it calls g_rand_int(). But I think if a function returns an uint32, it should also be named that way. Rename.
2019-05-30 08:33:14 +02:00
arr[i] = GINT_TO_POINTER (nmtst_get_rand_uint32 () % (i_len + BIN_SEARCH_W_DUPS_JITTER));
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
g_qsort_with_data (arr,
i_len,
sizeof (gpointer),
_test_bin_search2_cmp_p,
NULL);
for (i = 0; i < i_len + BIN_SEARCH_W_DUPS_JITTER; i++) {
gconstpointer p = GINT_TO_POINTER (i);
shared: move nm_utils_ptrarray_find_binary_search() to shared utils
2018-08-01 11:03:02 +02:00
idx = nm_utils_ptrarray_find_binary_search (arr, i_len, p, _test_bin_search2_cmp, NULL, &idx_first, &idx_last);
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
idx_first2 = _nm_utils_ptrarray_find_first (arr, i_len, p);
shared: move nm_utils_array_find_binary_search() to shared utils
2018-07-28 12:26:37 +02:00
idx2 = nm_utils_array_find_binary_search (arr, sizeof (gpointer), i_len, &p, _test_bin_search2_cmp_p, NULL);
utils: extend binary-search to return the first/last index binary-search can find an index of a matching entry in a sorted list. However, if the list contains multiple entries that compare equal, it can be interesting to find the first/last entry. For example, if you want to append new items after the last. Extend binary search to optionally continue the binary search to determine the range that compares equal.
2017-12-07 15:20:42 +01:00
g_assert_cmpint (idx, ==, idx2);
if (idx_first2 < 0) {
g_assert_cmpint (idx, <, 0);
g_assert_cmpint (idx, ==, idx_first);
g_assert_cmpint (idx, ==, idx_last);
idx = ~idx;
g_assert_cmpint (idx, >=, 0);
g_assert_cmpint (idx, <=, i_len);
if (i_len == 0)
g_assert_cmpint (idx, ==, 0);
else {
g_assert (idx == i_len || GPOINTER_TO_INT (arr[idx]) > i);
g_assert (idx == 0 || GPOINTER_TO_INT (arr[idx - 1]) < i);
}
} else {
g_assert_cmpint (idx_first, ==, idx_first2);
g_assert_cmpint (idx_first, >=, 0);
g_assert_cmpint (idx_last, <, i_len);
g_assert_cmpint (idx_first, <=, idx_last);
g_assert_cmpint (idx, >=, idx_first);
g_assert_cmpint (idx, <=, idx_last);
for (j = idx_first; j < idx_last; j++)
g_assert (GPOINTER_TO_INT (arr[j]) == i);
g_assert (idx_first == 0 || GPOINTER_TO_INT (arr[idx_first - 1]) < i);
g_assert (idx_last == i_len - 1 || GPOINTER_TO_INT (arr[idx_last + 1]) > i);
}
}
}
}
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
static void
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
_test_nm_utils_enum_to_str_do_full (GType type,
int flags,
const char *exp_str,
const NMUtilsEnumValueInfo *value_infos)
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
{
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
gs_free char *str = NULL;
int flags2;
gs_free char *err_token = NULL;
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
gboolean result;
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
g_assert (exp_str);
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
str = _nm_utils_enum_to_str_full (type, flags, ", ", value_infos);
g_assert_cmpstr (str, ==, exp_str);
if (!value_infos) {
gs_free char *str2 = NULL;
str2 = nm_utils_enum_to_str (type, flags);
g_assert_cmpstr (str2, ==, exp_str);
}
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
result = _nm_utils_enum_from_str_full (type, str, &flags2, &err_token, value_infos);
g_assert (result == TRUE);
g_assert_cmpint (flags2, ==, flags);
g_assert_cmpstr (err_token, ==, NULL);
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
}
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
#define _test_nm_utils_enum_to_str_do(...) _test_nm_utils_enum_to_str_do_full (__VA_ARGS__, NULL)
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
static void
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
_test_nm_utils_enum_from_str_do_full (GType type,
const char *str,
gboolean exp_result,
int exp_flags,
const char *exp_err_token,
const NMUtilsEnumValueInfo *value_infos)
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
{
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
int flags;
gs_free char *err_token = NULL;
gboolean result;
libnm: add _nm_utils_ptrarray_find_binary_search() helper
2015-06-23 15:05:25 +02:00
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
result = _nm_utils_enum_from_str_full (type, str, &flags, &err_token, value_infos);
g_assert (result == exp_result);
g_assert_cmpint (flags, ==, exp_flags);
g_assert_cmpstr (err_token, ==, exp_err_token);
if (!value_infos) {
int flags2;
gs_free char *err_token2 = NULL;
gboolean result2;
result2 = nm_utils_enum_from_str (type, str, &flags2, &err_token2);
g_assert (result2 == exp_result);
g_assert_cmpint (flags2, ==, exp_flags);
g_assert_cmpstr (err_token2, ==, exp_err_token);
}
if (result) {
int flags2;
gs_free char *str2 = NULL;
gs_free char *err_token2 = NULL;
str2 = _nm_utils_enum_to_str_full (type, flags, ", ", value_infos);
g_assert (str2);
result = _nm_utils_enum_from_str_full (type, str2, &flags2, &err_token2, value_infos);
g_assert (result == TRUE);
g_assert_cmpint (flags2, ==, flags);
g_assert_cmpstr (err_token, ==, NULL);
}
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
}
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
#define _test_nm_utils_enum_from_str_do(...) _test_nm_utils_enum_from_str_do_full(__VA_ARGS__, NULL)
libnm-core: add nm_utils_enum_get_values() Add function nm_utils_enum_get_values() which returns a string array containing the enum values. It can be used, for example, to build a list of allowed values for user.
2015-09-06 13:11:11 +02:00
static void
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
_test_nm_utils_enum_get_values_do (GType type, int from, int to, const char *exp_str)
libnm-core: add nm_utils_enum_get_values() Add function nm_utils_enum_get_values() which returns a string array containing the enum values. It can be used, for example, to build a list of allowed values for user.
2015-09-06 13:11:11 +02:00
{
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
gs_free const char **strv = NULL;
gs_free char *str = NULL;
g_assert (exp_str);
libnm-core: add nm_utils_enum_get_values() Add function nm_utils_enum_get_values() which returns a string array containing the enum values. It can be used, for example, to build a list of allowed values for user.
2015-09-06 13:11:11 +02:00
strv = nm_utils_enum_get_values (type, from, to);
g_assert (strv);
str = g_strjoinv (",", (char **) strv);
g_assert_cmpstr (str, ==, exp_str);
}
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
static void
test_nm_utils_enum (void)
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
{
GType bool_enum = nm_test_general_bool_enum_get_type();
GType meta_flags = nm_test_general_meta_flags_get_type();
GType color_flags = nm_test_general_color_flags_get_type();
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
static const NMUtilsEnumValueInfo color_value_infos[] = {
{
.nick = "nick-4d",
.value = 0x4D,
},
{
.nick = "nick-5",
.value = 5,
},
{
.nick = "nick-red",
.value = NM_TEST_GENERAL_COLOR_FLAGS_RED,
},
libnm/tests: fix crash in tests Fixes: daf4ba43da702e182df3c90b969c1d8fd762185b
2018-05-29 14:33:21 +02:00
{ 0 },
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
};
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
_test_nm_utils_enum_to_str_do (bool_enum, NM_TEST_GENERAL_BOOL_ENUM_YES, "yes");
_test_nm_utils_enum_to_str_do (bool_enum, NM_TEST_GENERAL_BOOL_ENUM_UNKNOWN, "unknown");
_test_nm_utils_enum_to_str_do (bool_enum, NM_TEST_GENERAL_BOOL_ENUM_INVALID, "4");
_test_nm_utils_enum_to_str_do (bool_enum, NM_TEST_GENERAL_BOOL_ENUM_67, "67");
_test_nm_utils_enum_to_str_do (bool_enum, NM_TEST_GENERAL_BOOL_ENUM_46, "64");
_test_nm_utils_enum_to_str_do (meta_flags, NM_TEST_GENERAL_META_FLAGS_NONE, "none");
_test_nm_utils_enum_to_str_do (meta_flags, NM_TEST_GENERAL_META_FLAGS_BAZ, "baz");
_test_nm_utils_enum_to_str_do (meta_flags, NM_TEST_GENERAL_META_FLAGS_FOO |
NM_TEST_GENERAL_META_FLAGS_BAR |
NM_TEST_GENERAL_META_FLAGS_BAZ, "foo, bar, baz");
_test_nm_utils_enum_to_str_do (meta_flags, 0xFF, "foo, bar, baz, 0xf8");
_test_nm_utils_enum_to_str_do (meta_flags, NM_TEST_GENERAL_META_FLAGS_0x8, "0x8");
_test_nm_utils_enum_to_str_do (meta_flags, NM_TEST_GENERAL_META_FLAGS_0x4, "0x10");
_test_nm_utils_enum_to_str_do (color_flags, NM_TEST_GENERAL_COLOR_FLAGS_RED, "red");
_test_nm_utils_enum_to_str_do (color_flags, NM_TEST_GENERAL_COLOR_FLAGS_WHITE, "0x1");
_test_nm_utils_enum_to_str_do (color_flags, NM_TEST_GENERAL_COLOR_FLAGS_RED |
NM_TEST_GENERAL_COLOR_FLAGS_GREEN, "red, green");
_test_nm_utils_enum_to_str_do_full (color_flags,
NM_TEST_GENERAL_COLOR_FLAGS_RED
| NM_TEST_GENERAL_COLOR_FLAGS_GREEN,
"nick-red, green",
color_value_infos);
_test_nm_utils_enum_to_str_do_full (color_flags,
0x4D
| NM_TEST_GENERAL_COLOR_FLAGS_RED
| NM_TEST_GENERAL_COLOR_FLAGS_GREEN,
"nick-4d",
color_value_infos);
_test_nm_utils_enum_to_str_do_full (color_flags,
5
| NM_TEST_GENERAL_COLOR_FLAGS_GREEN,
"nick-5, green",
color_value_infos);
_test_nm_utils_enum_from_str_do (bool_enum, "", FALSE, 0, NULL);
_test_nm_utils_enum_from_str_do (bool_enum, " ", FALSE, 0, NULL);
_test_nm_utils_enum_from_str_do (bool_enum, "invalid", FALSE, 0, "invalid");
_test_nm_utils_enum_from_str_do (bool_enum, "yes", TRUE, NM_TEST_GENERAL_BOOL_ENUM_YES, NULL);
_test_nm_utils_enum_from_str_do (bool_enum, "no", TRUE, NM_TEST_GENERAL_BOOL_ENUM_NO, NULL);
_test_nm_utils_enum_from_str_do (bool_enum, "yes,no", FALSE, 0, "yes,no");
_test_nm_utils_enum_from_str_do (meta_flags, "", TRUE, 0, NULL);
_test_nm_utils_enum_from_str_do (meta_flags, " ", TRUE, 0, NULL);
_test_nm_utils_enum_from_str_do (meta_flags, "foo", TRUE, NM_TEST_GENERAL_META_FLAGS_FOO, NULL);
_test_nm_utils_enum_from_str_do (meta_flags, "foo,baz", TRUE, NM_TEST_GENERAL_META_FLAGS_FOO |
nmcli: wake-on-lan property set/get fixes - accept a numeric value (decimal or hex (0x prefix)) - display a numeric value of the property in addition to the strings - add/accept spaces between string names to behave similar to other flags' properties.
2015-09-14 14:23:50 +02:00
NM_TEST_GENERAL_META_FLAGS_BAZ, NULL);
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
_test_nm_utils_enum_from_str_do (meta_flags, "foo, baz", TRUE, NM_TEST_GENERAL_META_FLAGS_FOO |
NM_TEST_GENERAL_META_FLAGS_BAZ, NULL);
_test_nm_utils_enum_from_str_do (meta_flags, "foo,,bar", TRUE, NM_TEST_GENERAL_META_FLAGS_FOO |
NM_TEST_GENERAL_META_FLAGS_BAR, NULL);
_test_nm_utils_enum_from_str_do (meta_flags, "foo,baz,quux,bar", FALSE, 0, "quux");
_test_nm_utils_enum_from_str_do (meta_flags, "foo,0x6", TRUE, NM_TEST_GENERAL_META_FLAGS_FOO | 0x6, NULL);
_test_nm_utils_enum_from_str_do (meta_flags, "0x30,0x08,foo", TRUE, 0x39, NULL);
_test_nm_utils_enum_from_str_do (color_flags, "green", TRUE, NM_TEST_GENERAL_COLOR_FLAGS_GREEN, NULL);
_test_nm_utils_enum_from_str_do (color_flags, "blue,red", TRUE, NM_TEST_GENERAL_COLOR_FLAGS_BLUE |
NM_TEST_GENERAL_COLOR_FLAGS_RED, NULL);
_test_nm_utils_enum_from_str_do (color_flags, "blue,white", FALSE, 0, "white");
_test_nm_utils_enum_from_str_do_full (color_flags,
"nick-red",
TRUE,
NM_TEST_GENERAL_COLOR_FLAGS_RED,
NULL,
color_value_infos);
_test_nm_utils_enum_from_str_do_full (color_flags,
"0x4D",
TRUE,
0x4D,
NULL,
color_value_infos);
_test_nm_utils_enum_from_str_do_full (color_flags,
"green,nick-4d",
TRUE,
0x4D
| NM_TEST_GENERAL_COLOR_FLAGS_GREEN,
NULL,
color_value_infos);
_test_nm_utils_enum_from_str_do_full (color_flags,
"nick-4d,nick-red,nick-5,green,nick-red",
TRUE,
shared: fix parsing aliases for flags in _nm_utils_enum_from_str_full() Otherwise, the last alias overwrites previous values. Fixes: b9fa0e0a195e9cf7363b563cca7ffec4b5016afd
2018-05-29 11:26:11 +02:00
0x4D
| NM_TEST_GENERAL_COLOR_FLAGS_GREEN,
shared/tests: extend tests for nm_utils_enum_from_str()
2018-05-29 12:07:53 +02:00
NULL,
color_value_infos);
_test_nm_utils_enum_get_values_do (bool_enum, 0, G_MAXINT, "no,yes,maybe,unknown,67,64");
_test_nm_utils_enum_get_values_do (bool_enum, NM_TEST_GENERAL_BOOL_ENUM_YES,
NM_TEST_GENERAL_BOOL_ENUM_MAYBE, "yes,maybe");
_test_nm_utils_enum_get_values_do (meta_flags, 0, G_MAXINT, "none,foo,bar,baz,0x8,0x10");
_test_nm_utils_enum_get_values_do (color_flags, 0, G_MAXINT, "blue,red,green");
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
static void
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
_do_test_utils_str_utf8safe_unescape (const char *str, const char *expected, gsize expected_len)
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
{
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
gsize l;
const char *s;
gs_free gpointer buf_free_1 = NULL;
gs_free char *str_free_1 = NULL;
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
shared: support stripping whitespace from nm_utils_buf_utf8safe_unescape() When parsing user input if is often convenient to allow stripping whitespace. Especially with escaped strings, the user could still escape the whitespace, if the space should be taken literally. Add support for that to nm_utils_buf_utf8safe_unescape(). Note that this is not the same as calling g_strstrip() before/after unescape. That is, because nm_utils_buf_utf8safe_unescape() correctly preserves escaped whitespace. If you call g_strstrip() before/after the unescape, you don't know whether the whitespace is escaped.
2020-05-04 09:52:16 +02:00
s = nm_utils_buf_utf8safe_unescape (str, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE, &l, &buf_free_1);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert_cmpint (expected_len, ==, l);
g_assert_cmpstr (s, ==, expected);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
if (str == NULL) {
g_assert (!s);
g_assert (!buf_free_1);
g_assert_cmpint (l, ==, 0);
} else {
g_assert (s);
if (!strchr (str, '\\')) {
g_assert (!buf_free_1);
g_assert (s == str);
g_assert_cmpint (l, ==, strlen (str));
} else {
g_assert (buf_free_1);
g_assert (s == buf_free_1);
g_assert (memcmp (s, expected, expected_len) == 0);
}
}
if ( expected
&& l == strlen (expected)) {
/* there are no embeeded NULs. Check that nm_utils_str_utf8safe_unescape() yields the same result. */
shared: support stripping whitespace from nm_utils_buf_utf8safe_unescape() When parsing user input if is often convenient to allow stripping whitespace. Especially with escaped strings, the user could still escape the whitespace, if the space should be taken literally. Add support for that to nm_utils_buf_utf8safe_unescape(). Note that this is not the same as calling g_strstrip() before/after unescape. That is, because nm_utils_buf_utf8safe_unescape() correctly preserves escaped whitespace. If you call g_strstrip() before/after the unescape, you don't know whether the whitespace is escaped.
2020-05-04 09:52:16 +02:00
s = nm_utils_str_utf8safe_unescape (str, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE, &str_free_1);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert_cmpstr (s, ==, expected);
if (strchr (str, '\\')) {
g_assert (str_free_1 != str);
g_assert (s == str_free_1);
} else
g_assert (s == str);
}
}
#define do_test_utils_str_utf8safe_unescape(str, expected) \
_do_test_utils_str_utf8safe_unescape (""str"", expected, NM_STRLEN (expected))
static void
_do_test_utils_str_utf8safe (const char *str, gsize str_len, const char *expected, NMUtilsStrUtf8SafeFlags flags)
{
const char *str_safe;
const char *buf_safe;
const char *s;
gs_free char *str_free_1 = NULL;
gs_free char *str_free_2 = NULL;
gs_free char *str_free_3 = NULL;
gs_free char *str_free_4 = NULL;
gs_free char *str_free_5 = NULL;
gs_free char *str_free_6 = NULL;
gs_free char *str_free_7 = NULL;
gs_free char *str_free_8 = NULL;
gboolean str_has_nul = FALSE;
shared: add NM_UTILS_STR_UTF8_SAFE_FLAG_SECRET flag The new flag tells that as we re-allocate data buffers during escaping, we bzero the memory to avoid leaking secrets.
2018-09-09 12:19:07 +02:00
#define RND_FLAG ((nmtst_get_rand_bool ()) ? NM_UTILS_STR_UTF8_SAFE_FLAG_NONE : NM_UTILS_STR_UTF8_SAFE_FLAG_SECRET)
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
shared: add NM_UTILS_STR_UTF8_SAFE_FLAG_SECRET flag The new flag tells that as we re-allocate data buffers during escaping, we bzero the memory to avoid leaking secrets.
2018-09-09 12:19:07 +02:00
buf_safe = nm_utils_buf_utf8safe_escape (str, str_len, flags | RND_FLAG, &str_free_1);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
shared: add NM_UTILS_STR_UTF8_SAFE_FLAG_SECRET flag The new flag tells that as we re-allocate data buffers during escaping, we bzero the memory to avoid leaking secrets.
2018-09-09 12:19:07 +02:00
str_safe = nm_utils_str_utf8safe_escape (str, flags | RND_FLAG, &str_free_2);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
if (str_len == 0) {
g_assert (buf_safe == NULL);
g_assert (str_free_1 == NULL);
g_assert (str_safe == str);
g_assert (str == NULL || str[0] == '\0');
g_assert (str_free_2 == NULL);
} else if (str_len == strlen (str)) {
g_assert (buf_safe);
g_assert_cmpstr (buf_safe, ==, str_safe);
/* nm_utils_buf_utf8safe_escape() can only return a pointer equal to the input string,
* if and only if str_len is negative. Otherwise, the input str won't be NUL terminated
* and cannot be returned. */
g_assert (buf_safe != str);
g_assert (buf_safe == str_free_1);
} else
str_has_nul = TRUE;
shared: add NM_UTILS_STR_UTF8_SAFE_FLAG_SECRET flag The new flag tells that as we re-allocate data buffers during escaping, we bzero the memory to avoid leaking secrets.
2018-09-09 12:19:07 +02:00
str_free_3 = nm_utils_str_utf8safe_escape_cp (str, flags | RND_FLAG);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert_cmpstr (str_free_3, ==, str_safe);
g_assert ((!str && !str_free_3) || (str != str_free_3));
if (str_len > 0)
_do_test_utils_str_utf8safe_unescape (buf_safe, str, str_len);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
if (expected == NULL) {
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert (!str_has_nul);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
g_assert (str_safe == str);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert (!str_free_2);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
if (str) {
g_assert (!strchr (str, '\\'));
g_assert (g_utf8_validate (str, -1, NULL));
}
shared: support stripping whitespace from nm_utils_buf_utf8safe_unescape() When parsing user input if is often convenient to allow stripping whitespace. Especially with escaped strings, the user could still escape the whitespace, if the space should be taken literally. Add support for that to nm_utils_buf_utf8safe_unescape(). Note that this is not the same as calling g_strstrip() before/after unescape. That is, because nm_utils_buf_utf8safe_unescape() correctly preserves escaped whitespace. If you call g_strstrip() before/after the unescape, you don't know whether the whitespace is escaped.
2020-05-04 09:52:16 +02:00
g_assert (str == nm_utils_str_utf8safe_unescape (str_safe, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE, &str_free_4));
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert (!str_free_4);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
shared: support stripping whitespace from nm_utils_buf_utf8safe_unescape() When parsing user input if is often convenient to allow stripping whitespace. Especially with escaped strings, the user could still escape the whitespace, if the space should be taken literally. Add support for that to nm_utils_buf_utf8safe_unescape(). Note that this is not the same as calling g_strstrip() before/after unescape. That is, because nm_utils_buf_utf8safe_unescape() correctly preserves escaped whitespace. If you call g_strstrip() before/after the unescape, you don't know whether the whitespace is escaped.
2020-05-04 09:52:16 +02:00
str_free_5 = nm_utils_str_utf8safe_unescape_cp (str_safe, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
if (str) {
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert (str_free_5 != str);
g_assert_cmpstr (str_free_5, ==, str);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
} else
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert (!str_free_5);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
return;
}
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
if (!str_has_nul) {
g_assert (str);
g_assert (str_safe != str);
g_assert (str_safe == str_free_2);
g_assert ( strchr (str, '\\')
|| !g_utf8_validate (str, -1, NULL)
|| ( NM_FLAGS_HAS (flags, NM_UTILS_STR_UTF8_SAFE_FLAG_ESCAPE_NON_ASCII)
&& NM_STRCHAR_ANY (str, ch, (guchar) ch >= 127))
|| ( NM_FLAGS_HAS (flags, NM_UTILS_STR_UTF8_SAFE_FLAG_ESCAPE_CTRL)
&& NM_STRCHAR_ANY (str, ch, (guchar) ch < ' ')));
g_assert (g_utf8_validate (str_safe, -1, NULL));
str_free_6 = g_strcompress (str_safe);
g_assert_cmpstr (str, ==, str_free_6);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
shared: support stripping whitespace from nm_utils_buf_utf8safe_unescape() When parsing user input if is often convenient to allow stripping whitespace. Especially with escaped strings, the user could still escape the whitespace, if the space should be taken literally. Add support for that to nm_utils_buf_utf8safe_unescape(). Note that this is not the same as calling g_strstrip() before/after unescape. That is, because nm_utils_buf_utf8safe_unescape() correctly preserves escaped whitespace. If you call g_strstrip() before/after the unescape, you don't know whether the whitespace is escaped.
2020-05-04 09:52:16 +02:00
str_free_7 = nm_utils_str_utf8safe_unescape_cp (str_safe, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert (str_free_7 != str);
g_assert_cmpstr (str_free_7, ==, str);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
shared: support stripping whitespace from nm_utils_buf_utf8safe_unescape() When parsing user input if is often convenient to allow stripping whitespace. Especially with escaped strings, the user could still escape the whitespace, if the space should be taken literally. Add support for that to nm_utils_buf_utf8safe_unescape(). Note that this is not the same as calling g_strstrip() before/after unescape. That is, because nm_utils_buf_utf8safe_unescape() correctly preserves escaped whitespace. If you call g_strstrip() before/after the unescape, you don't know whether the whitespace is escaped.
2020-05-04 09:52:16 +02:00
s = nm_utils_str_utf8safe_unescape (str_safe, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE, &str_free_8);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert (str_free_8 != str);
g_assert (s == str_free_8);
g_assert_cmpstr (str_free_8, ==, str);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
g_assert_cmpstr (str_safe, ==, expected);
return;
}
g_assert_cmpstr (buf_safe, ==, expected);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
}
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
#define do_test_utils_str_utf8safe(str, expected, flags) \
_do_test_utils_str_utf8safe (""str"", NM_STRLEN (str), expected, flags)
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
static void
test_utils_str_utf8safe (void)
{
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
_do_test_utils_str_utf8safe (NULL, 0, NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
do_test_utils_str_utf8safe ("", NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
do_test_utils_str_utf8safe ("\\", "\\\\", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\\a", "\\\\a", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
do_test_utils_str_utf8safe ("\314", "\\314", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\314\315x\315\315x", "\\314\\315x\\315\\315x", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\314\315xx", "\\314\\315xx", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\314xx", "\\314xx", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\xa0", "\\240", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\xe2\x91\xa0", NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\xe2\xe2\x91\xa0", "\\342\xe2\x91\xa0", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\xe2\xe2\x91\xa0\xa0", "\\342\xe2\x91\xa0\\240", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("a", NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("ab", NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("ab\314", "ab\\314", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("ab\314adsf", "ab\\314adsf", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("abadsf", NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("abäb", NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("x\xa0", "x\\240", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("Ä\304ab\\äb", "Ä\\304ab\\\\äb", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("Äab\\äb", "Äab\\\\äb", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("ÄÄab\\äb", "ÄÄab\\\\äb", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("㈞abä㈞b", NULL, NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("abäb", "ab\\303\\244b", NM_UTILS_STR_UTF8_SAFE_FLAG_ESCAPE_NON_ASCII);
do_test_utils_str_utf8safe ("ab\ab", "ab\\007b", NM_UTILS_STR_UTF8_SAFE_FLAG_ESCAPE_CTRL);
shared: add nm_utils_buf_utf8safe_escape() util We already have nm_utils_str_utf8safe_escape() to convert a NUL termianted string to an UTF-8 string. nm_utils_str_utf8safe_escape() operates under the assumption, that the input strig is already valid UTF-8 and returns the input string verbatim. That way, in the common expected cases, the string just looks like a regular UTF-8 string. However, in case there are invalid UTF-8 sequences (or a backslash escape characters), the function will use backslash escaping to encode the input string as a valid UTF-8 sequence. Note that the escaped sequence, can be reverted to the original non-UTF-8 string via unescape. An example, where this is useful are file names or interface names. Which are not in a defined encoding, but NUL terminated and commonly ASCII or UTF-8 encoded. Extend this, to also handle not NUL terminated buffers. The same applies, except that the process cannot be reverted via g_strcompress() -- because the NUL character cannot be unescaped. This will be useful to escape a Wi-Fi SSID. Commonly we expect the SSID to be in UTF-8/ASCII encoding and we want to print it verbatim. Only if that is not the case, we fallback to backslash escaping. However, the orginal value can be fully recovered via unescape(). The difference between an SSID and a filename is, that the former can contain '\0' bytes.
2018-08-12 11:43:25 +02:00
do_test_utils_str_utf8safe ("\0", "\\000", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\0a\0", "\\000a\\000", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\\\0", "\\\\\\000", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\n\0", "\n\\000", NM_UTILS_STR_UTF8_SAFE_FLAG_NONE);
do_test_utils_str_utf8safe ("\n\0", "\\012\\000", NM_UTILS_STR_UTF8_SAFE_FLAG_ESCAPE_CTRL);
do_test_utils_str_utf8safe_unescape ("\n\\0", "\n\0");
do_test_utils_str_utf8safe_unescape ("\n\\01", "\n\01");
do_test_utils_str_utf8safe_unescape ("\n\\012", "\n\012");
do_test_utils_str_utf8safe_unescape ("\n\\.", "\n.");
do_test_utils_str_utf8safe_unescape ("\\n\\.3\\r", "\n.3\r");
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
}
/*****************************************************************************/
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
static int
_test_nm_in_set_get (int *call_counter, gboolean allow_called, int value)
{
g_assert (call_counter);
*call_counter += 1;
if (!allow_called)
g_assert_not_reached ();
return value;
}
static void
_test_nm_in_set_assert (int *call_counter, int expected)
{
g_assert (call_counter);
g_assert_cmpint (expected, ==, *call_counter);
*call_counter = 0;
}
static void
test_nm_in_set (void)
{
int call_counter = 0;
#define G(x) _test_nm_in_set_get (&call_counter, TRUE, x)
#define N(x) _test_nm_in_set_get (&call_counter, FALSE, x)
#define _ASSERT(expected, expr) \
G_STMT_START { \
_test_nm_in_set_assert (&call_counter, 0); \
g_assert (expr); \
_test_nm_in_set_assert (&call_counter, (expected)); \
} G_STMT_END
_ASSERT (1, !NM_IN_SET (-1, G( 1)));
_ASSERT (1, NM_IN_SET (-1, G(-1)));
_ASSERT (2, !NM_IN_SET (-1, G( 1), G( 2)));
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (1, NM_IN_SET (-1, G(-1), N( 2)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
_ASSERT (2, NM_IN_SET (-1, G( 1), G(-1)));
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (1, NM_IN_SET (-1, G(-1), N(-1)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
_ASSERT (3, !NM_IN_SET (-1, G( 1), G( 2), G( 3)));
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (1, NM_IN_SET (-1, G(-1), N( 2), N( 3)));
_ASSERT (2, NM_IN_SET (-1, G( 1), G(-1), N( 3)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
_ASSERT (3, NM_IN_SET (-1, G( 1), G( 2), G(-1)));
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (2, NM_IN_SET (-1, G( 1), G(-1), N(-1)));
_ASSERT (1, NM_IN_SET (-1, G(-1), N( 2), N(-1)));
_ASSERT (1, NM_IN_SET (-1, G(-1), N(-1), N( 3)));
_ASSERT (1, NM_IN_SET (-1, G(-1), N(-1), N(-1)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
_ASSERT (4, !NM_IN_SET (-1, G( 1), G( 2), G( 3), G( 4)));
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (1, NM_IN_SET (-1, G(-1), N( 2), N( 3), N( 4)));
_ASSERT (2, NM_IN_SET (-1, G( 1), G(-1), N( 3), N( 4)));
_ASSERT (3, NM_IN_SET (-1, G( 1), G( 2), G(-1), N( 4)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
_ASSERT (4, NM_IN_SET (-1, G( 1), G( 2), G( 3), G(-1)));
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (4, NM_IN_SET (-1, G( 1), G( 2), G( 3), G(-1), G( 5)));
macros: simplify NM_IN_SET() macro Also fix NM_IN_SET_SE() to guaranteed return 0 or 1.
2016-02-11 17:24:33 +01:00
_ASSERT (5, NM_IN_SET (-1, G( 1), G( 2), G( 3), G( 4), G(-1)));
_ASSERT (6, NM_IN_SET (-1, G( 1), G( 2), G( 3), G( 4), G( 5), G( -1)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (1, !NM_IN_SET_SE (-1, G( 1)));
_ASSERT (1, NM_IN_SET_SE (-1, G(-1)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (2, !NM_IN_SET_SE (-1, G( 1), G( 2)));
_ASSERT (2, NM_IN_SET_SE (-1, G(-1), G( 2)));
_ASSERT (2, NM_IN_SET_SE (-1, G( 1), G(-1)));
_ASSERT (2, NM_IN_SET_SE (-1, G(-1), G(-1)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (3, !NM_IN_SET_SE (-1, G( 1), G( 2), G( 3)));
_ASSERT (3, NM_IN_SET_SE (-1, G(-1), G( 2), G( 3)));
_ASSERT (3, NM_IN_SET_SE (-1, G( 1), G(-1), G( 3)));
_ASSERT (3, NM_IN_SET_SE (-1, G( 1), G( 2), G(-1)));
_ASSERT (3, NM_IN_SET_SE (-1, G( 1), G(-1), G(-1)));
_ASSERT (3, NM_IN_SET_SE (-1, G(-1), G( 2), G(-1)));
_ASSERT (3, NM_IN_SET_SE (-1, G(-1), G(-1), G( 3)));
_ASSERT (3, NM_IN_SET_SE (-1, G(-1), G(-1), G(-1)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (4, !NM_IN_SET_SE (-1, G( 1), G( 2), G( 3), G( 4)));
_ASSERT (4, NM_IN_SET_SE (-1, G(-1), G( 2), G( 3), G( 4)));
_ASSERT (4, NM_IN_SET_SE (-1, G( 1), G(-1), G( 3), G( 4)));
_ASSERT (4, NM_IN_SET_SE (-1, G( 1), G( 2), G(-1), G( 4)));
_ASSERT (4, NM_IN_SET_SE (-1, G( 1), G( 2), G( 3), G(-1)));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
macros: use short-circuit evaluation in NM_IN_SET() All current users of NM_IN_SET() would rather use short-circuit evalation (or don't care). It seems that doing it by default seems favorable. The only downside is, that this might have somewhat unexpected behavior to a user who expects a regular function (which would evaluate always all arguments). Fixes: 7860ef959ae4b244017b9dc491e9c0a6024279dd
2015-08-12 13:04:12 +02:00
_ASSERT (5, NM_IN_SET_SE (-1, G( 1), G( 2), G( 3), G(-1), G( 5)));
macros: simplify NM_IN_SET() macro Also fix NM_IN_SET_SE() to guaranteed return 0 or 1.
2016-02-11 17:24:33 +01:00
_ASSERT (6, NM_IN_SET_SE (-1, G( 1), G( 2), G( 3), G( 4), G( 5), G(-1)));
macros: simplify NM_IN_SET() and NM_IN_STRSET() macros and support up to 16 arguments.
2016-09-22 14:14:22 +02:00
libnm-core/tests: avoid -Wstring-compare in unit test for NM_IN_SET() Clang 10 doesn't like NM_IN_SET() with strings and is right about that: ../libnm-core/tests/test-general.c:7763:9: error: result of comparison against a string literal is unspecified (use an explicit string comparison function instead) [-Werror,-Wstring-compare] (void) NM_IN_SET ("a", "1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16"); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ However, NM_IN_STRSET() should work.
2020-02-21 15:15:12 +01:00
g_assert (!NM_IN_SET (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16));
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
#undef G
#undef N
#undef _ASSERT
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
macros: add NM_IN_STRSET() Add macro similar to NM_IN_SET() that checks for C strings. NULL values are allowed and handled as one would expect.
2016-02-11 16:53:06 +01:00
static const char *
_test_nm_in_set_getstr (int *call_counter, gboolean allow_called, const char *value)
{
g_assert (call_counter);
*call_counter += 1;
if (!allow_called)
g_assert_not_reached ();
return value;
}
static void
test_nm_in_strset (void)
{
int call_counter = 0;
#define G(x) _test_nm_in_set_getstr (&call_counter, TRUE, x)
#define N(x) _test_nm_in_set_getstr (&call_counter, FALSE, x)
#define _ASSERT(expected, expr) \
G_STMT_START { \
_test_nm_in_set_assert (&call_counter, 0); \
g_assert (expr); \
_test_nm_in_set_assert (&call_counter, (expected)); \
} G_STMT_END
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL)));
_ASSERT (1, !NM_IN_STRSET ("a", G(NULL)));
_ASSERT (1, !NM_IN_STRSET (NULL, G("a")));
_ASSERT (1, NM_IN_STRSET_SE (NULL, G(NULL)));
_ASSERT (1, !NM_IN_STRSET_SE ("a", G(NULL)));
_ASSERT (1, !NM_IN_STRSET_SE (NULL, G("a")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N(NULL)));
_ASSERT (2, !NM_IN_STRSET ("a", G(NULL), G(NULL)));
_ASSERT (2, NM_IN_STRSET (NULL, G("a"), G(NULL)));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("a")));
_ASSERT (2, NM_IN_STRSET ("a", G(NULL), G("a")));
_ASSERT (2, !NM_IN_STRSET (NULL, G("a"), G("a")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("b")));
_ASSERT (2, !NM_IN_STRSET ("a", G(NULL), G("b")));
_ASSERT (2, !NM_IN_STRSET (NULL, G("a"), G("b")));
_ASSERT (2, NM_IN_STRSET_SE (NULL, G(NULL), G(NULL)));
_ASSERT (2, !NM_IN_STRSET_SE ("a", G(NULL), G(NULL)));
_ASSERT (2, NM_IN_STRSET_SE (NULL, G("a"), G(NULL)));
_ASSERT (2, NM_IN_STRSET_SE (NULL, G(NULL), G("a")));
_ASSERT (2, NM_IN_STRSET_SE ("a", G(NULL), G("a")));
_ASSERT (2, !NM_IN_STRSET_SE (NULL, G("a"), G("a")));
_ASSERT (2, NM_IN_STRSET_SE (NULL, G(NULL), G("b")));
_ASSERT (2, !NM_IN_STRSET_SE ("a", G(NULL), G("b")));
_ASSERT (2, !NM_IN_STRSET_SE (NULL, G("a"), G("b")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N(NULL), N(NULL)));
_ASSERT (3, !NM_IN_STRSET ("a", G(NULL), G(NULL), G(NULL)));
_ASSERT (2, NM_IN_STRSET (NULL, G("a"), G(NULL), N(NULL)));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("a"), N(NULL)));
_ASSERT (2, NM_IN_STRSET ("a", G(NULL), G("a"), N(NULL)));
_ASSERT (3, NM_IN_STRSET (NULL, G("a"), G("a"), G(NULL)));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("b"), N(NULL)));
_ASSERT (3, !NM_IN_STRSET ("a", G(NULL), G("b"), G(NULL)));
_ASSERT (3, NM_IN_STRSET (NULL, G("a"), G("b"), G(NULL)));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N(NULL), N("a")));
_ASSERT (3, NM_IN_STRSET ("a", G(NULL), G(NULL), G("a")));
_ASSERT (2, NM_IN_STRSET (NULL, G("a"), G(NULL), N("a")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("a"), N("a")));
_ASSERT (2, NM_IN_STRSET ("a", G(NULL), G("a"), N("a")));
_ASSERT (3, !NM_IN_STRSET (NULL, G("a"), G("a"), G("a")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("b"), N("a")));
_ASSERT (3, NM_IN_STRSET ("a", G(NULL), G("b"), G("a")));
_ASSERT (3, !NM_IN_STRSET (NULL, G("a"), G("b"), G("a")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N(NULL), N("b")));
_ASSERT (3, !NM_IN_STRSET ("a", G(NULL), G(NULL), G("b")));
_ASSERT (2, NM_IN_STRSET (NULL, G("a"), G(NULL), N("b")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("a"), N("b")));
_ASSERT (2, NM_IN_STRSET ("a", G(NULL), G("a"), N("b")));
_ASSERT (3, !NM_IN_STRSET (NULL, G("a"), G("a"), G("b")));
_ASSERT (1, NM_IN_STRSET (NULL, G(NULL), N("b"), N("b")));
_ASSERT (3, !NM_IN_STRSET ("a", G(NULL), G("b"), G("b")));
_ASSERT (3, !NM_IN_STRSET (NULL, G("a"), G("b"), G("b")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G(NULL), G(NULL)));
_ASSERT (3, !NM_IN_STRSET_SE ("a", G(NULL), G(NULL), G(NULL)));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G("a"), G(NULL), G(NULL)));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G("a"), G(NULL)));
_ASSERT (3, NM_IN_STRSET_SE ("a", G(NULL), G("a"), G(NULL)));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G("a"), G("a"), G(NULL)));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G("b"), G(NULL)));
_ASSERT (3, !NM_IN_STRSET_SE ("a", G(NULL), G("b"), G(NULL)));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G("a"), G("b"), G(NULL)));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G(NULL), G("a")));
_ASSERT (3, NM_IN_STRSET_SE ("a", G(NULL), G(NULL), G("a")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G("a"), G(NULL), G("a")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G("a"), G("a")));
_ASSERT (3, NM_IN_STRSET_SE ("a", G(NULL), G("a"), G("a")));
_ASSERT (3, !NM_IN_STRSET_SE (NULL, G("a"), G("a"), G("a")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G("b"), G("a")));
_ASSERT (3, NM_IN_STRSET_SE ("a", G(NULL), G("b"), G("a")));
_ASSERT (3, !NM_IN_STRSET_SE (NULL, G("a"), G("b"), G("a")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G(NULL), G("b")));
_ASSERT (3, !NM_IN_STRSET_SE ("a", G(NULL), G(NULL), G("b")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G("a"), G(NULL), G("b")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G("a"), G("b")));
_ASSERT (3, NM_IN_STRSET_SE ("a", G(NULL), G("a"), G("b")));
_ASSERT (3, !NM_IN_STRSET_SE (NULL, G("a"), G("a"), G("b")));
_ASSERT (3, NM_IN_STRSET_SE (NULL, G(NULL), G("b"), G("b")));
_ASSERT (3, !NM_IN_STRSET_SE ("a", G(NULL), G("b"), G("b")));
_ASSERT (3, !NM_IN_STRSET_SE (NULL, G("a"), G("b"), G("b")));
_ASSERT (3, NM_IN_STRSET ("a", G(NULL), G("b"), G("a"), N("a")));
_ASSERT (4, NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("a")));
_ASSERT (4, !NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("d")));
_ASSERT (4, NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("a"), N("a")));
_ASSERT (5, NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("d"), G("a")));
_ASSERT (5, !NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("d"), G("e")));
_ASSERT (5, NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("d"), G("a"), N("a")));
_ASSERT (6, NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("d"), G("e"), G("a")));
_ASSERT (6, !NM_IN_STRSET ("a", G(NULL), G("b"), G("c"), G("d"), G("e"), G("f")));
macros: simplify NM_IN_SET() and NM_IN_STRSET() macros and support up to 16 arguments.
2016-09-22 14:14:22 +02:00
libnm-core/tests: avoid -Wstring-compare in unit test for NM_IN_SET() Clang 10 doesn't like NM_IN_SET() with strings and is right about that: ../libnm-core/tests/test-general.c:7763:9: error: result of comparison against a string literal is unspecified (use an explicit string comparison function instead) [-Werror,-Wstring-compare] (void) NM_IN_SET ("a", "1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16"); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ However, NM_IN_STRSET() should work.
2020-02-21 15:15:12 +01:00
g_assert (!NM_IN_STRSET (NULL, "1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16"));
g_assert (!NM_IN_STRSET ("_", "1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16"));
g_assert ( NM_IN_STRSET ("10", "1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16"));
macros: add NM_IN_STRSET() Add macro similar to NM_IN_SET() that checks for C strings. NULL values are allowed and handled as one would expect.
2016-02-11 16:53:06 +01:00
#undef G
#undef N
#undef _ASSERT
}
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
static void
test_route_attributes_parse (void)
{
GHashTable *ht;
GError *error = NULL;
GVariant *variant;
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
ht = nm_utils_parse_variant_attributes ("mtu=1400 src=1.2.3.4 cwnd=14",
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
' ', '=', FALSE,
nm_ip_route_get_variant_attribute_spec (),
&error);
g_assert_no_error (error);
g_assert (ht);
g_hash_table_unref (ht);
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
ht = nm_utils_parse_variant_attributes ("mtu=1400 src=1.2.3.4 cwnd=14 \\",
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
' ', '=', FALSE,
nm_ip_route_get_variant_attribute_spec (),
&error);
g_assert_error (error, NM_CONNECTION_ERROR, NM_CONNECTION_ERROR_FAILED);
g_assert (!ht);
g_clear_error (&error);
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
ht = nm_utils_parse_variant_attributes ("mtu.1400 src.1\\.2\\.3\\.4 ",
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
' ', '.', FALSE,
nm_ip_route_get_variant_attribute_spec (),
&error);
g_assert (ht);
g_assert_no_error (error);
variant = g_hash_table_lookup (ht, NM_IP_ROUTE_ATTRIBUTE_MTU);
g_assert (variant);
g_assert (g_variant_is_of_type (variant, G_VARIANT_TYPE_UINT32));
g_assert_cmpuint (g_variant_get_uint32 (variant), ==, 1400);
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
variant = g_hash_table_lookup (ht, NM_IP_ROUTE_ATTRIBUTE_SRC);
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
g_assert (variant);
g_assert (g_variant_is_of_type (variant, G_VARIANT_TYPE_STRING));
g_assert_cmpstr (g_variant_get_string (variant, NULL), ==, "1.2.3.4");
g_hash_table_unref (ht);
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
ht = nm_utils_parse_variant_attributes ("from:fd01\\:\\:42\\/64/initrwnd:21",
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
'/', ':', FALSE,
nm_ip_route_get_variant_attribute_spec (),
&error);
g_assert (ht);
g_assert_no_error (error);
variant = g_hash_table_lookup (ht, NM_IP_ROUTE_ATTRIBUTE_INITRWND);
g_assert (variant);
g_assert (g_variant_is_of_type (variant, G_VARIANT_TYPE_UINT32));
g_assert_cmpuint (g_variant_get_uint32 (variant), ==, 21);
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
variant = g_hash_table_lookup (ht, NM_IP_ROUTE_ATTRIBUTE_FROM);
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
g_assert (variant);
g_assert (g_variant_is_of_type (variant, G_VARIANT_TYPE_STRING));
g_assert_cmpstr (g_variant_get_string (variant, NULL), ==, "fd01::42/64");
g_hash_table_unref (ht);
}
static void
test_route_attributes_format (void)
{
gs_unref_hashtable GHashTable *ht = NULL;
char *str;
all: use nm_str_hash() instead of g_str_hash() We also do this for libnm and libnm-core, where it causes visible changes in behavior. But if somebody would rely on the hashing implementation for hash tables, it would be seriously flawed.
2017-11-15 16:06:43 +01:00
ht = g_hash_table_new_full (nm_str_hash, g_str_equal,
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
NULL, (GDestroyNotify) g_variant_unref);
str = nm_utils_format_variant_attributes (NULL, ' ', '=');
g_assert_cmpstr (str, ==, NULL);
str = nm_utils_format_variant_attributes (ht, ' ', '=');
g_assert_cmpstr (str, ==, NULL);
g_hash_table_insert (ht, NM_IP_ROUTE_ATTRIBUTE_MTU, g_variant_new_uint32 (5000));
g_hash_table_insert (ht, NM_IP_ROUTE_ATTRIBUTE_INITRWND, g_variant_new_uint32 (20));
g_hash_table_insert (ht, NM_IP_ROUTE_ATTRIBUTE_LOCK_MTU, g_variant_new_boolean (TRUE));
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
g_hash_table_insert (ht, NM_IP_ROUTE_ATTRIBUTE_SRC, g_variant_new_string ("aaaa:bbbb::1"));
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
str = nm_utils_format_variant_attributes (ht, ' ', '=');
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
g_assert_cmpstr (str, ==, "initrwnd=20 lock-mtu=true mtu=5000 src=aaaa:bbbb::1");
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
g_hash_table_remove_all (ht);
g_free (str);
g_hash_table_insert (ht, NM_IP_ROUTE_ATTRIBUTE_WINDOW, g_variant_new_uint32 (30000));
g_hash_table_insert (ht, NM_IP_ROUTE_ATTRIBUTE_INITCWND, g_variant_new_uint32 (21));
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
g_hash_table_insert (ht, NM_IP_ROUTE_ATTRIBUTE_FROM, g_variant_new_string ("aaaa:bbbb:cccc:dddd::/64"));
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
str = nm_utils_format_variant_attributes (ht, '/', ':');
core,libnm-core: use same route attribute names of iproute2 Users are probably more familiar with iproute2 route option names than kernel ones. Fixes: 54e58eb96bbfcd26d31ddba2e98ff2c59335a02a
2017-03-15 09:29:47 +01:00
g_assert_cmpstr (str, ==, "from:aaaa\\:bbbb\\:cccc\\:dddd\\:\\:\\/64/initcwnd:21/window:30000");
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
g_hash_table_remove_all (ht);
g_free (str);
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
macros: add NM_IN_STRSET() Add macro similar to NM_IN_SET() that checks for C strings. NULL values are allowed and handled as one would expect.
2016-02-11 16:53:06 +01:00
libnm: lookup route attributes attribute spec via binary search
2019-11-27 13:13:48 +01:00
static void
test_variant_attribute_spec (void)
{
const NMVariantAttributeSpec *const *const specs_list[] = {
nm_ip_route_get_variant_attribute_spec (),
};
int i_specs;
for (i_specs = 0; i_specs < G_N_ELEMENTS (specs_list); i_specs++) {
const NMVariantAttributeSpec *const *const specs = specs_list[i_specs];
gsize len;
gsize i;
g_assert (specs);
len = NM_PTRARRAY_LEN (specs);
g_assert_cmpint (len, >, 0u);
_nmtst_variant_attribute_spec_assert_sorted (specs, len);
for (i = 0; i < len; i++)
g_assert (specs[i] == _nm_variant_attribute_spec_find_binary_search (specs, len, specs[i]->name));
g_assert (!_nm_variant_attribute_spec_find_binary_search (specs, len, "bogus"));
}
}
/*****************************************************************************/
shared: add test for NM_SET_OUT()
2016-09-26 16:59:30 +02:00
static gboolean
all: don't use gchar/gshort/gint/glong but C types We commonly don't use the glib typedefs for char/short/int/long, but their C types directly. $ git grep '\<g\(char\|short\|int\|long\|float\|double\)\>' | wc -l 587 $ git grep '\<\(char\|short\|int\|long\|float\|double\)\>' | wc -l 21114 One could argue that using the glib typedefs is preferable in public API (of our glib based libnm library) or where it clearly is related to glib, like during g_object_set (obj, PROPERTY, (gint) value, NULL); However, that argument does not seem strong, because in practice we don't follow that argument today, and seldomly use the glib typedefs. Also, the style guide for this would be hard to formalize, because "using them where clearly related to a glib" is a very loose suggestion. Also note that glib typedefs will always just be typedefs of the underlying C types. There is no danger of glib changing the meaning of these typedefs (because that would be a major API break of glib). A simple style guide is instead: don't use these typedefs. No manual actions, I only ran the bash script: FILES=($(git ls-files '*.[hc]')) sed -i \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\>\( [^ ]\)/\1\2/g' \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\> /\1 /g' \ -e 's/\<g\(char\|short\|int\|long\|float\|double\)\>/\1/g' \ "${FILES[@]}"
2018-07-11 07:40:19 +02:00
do_test_nm_set_out_called (int *call_count)
shared: add test for NM_SET_OUT()
2016-09-26 16:59:30 +02:00
{
(*call_count)++;
return TRUE;
}
static void
test_nm_set_out (void)
{
gboolean val;
gboolean *p_val;
int call_count;
/* NM_SET_OUT() has an unexpected non-function like behavior
* wrt. side-effects of the value argument. Test it */
p_val = &val;
call_count = 0;
NM_SET_OUT (p_val, do_test_nm_set_out_called (&call_count));
g_assert_cmpint (call_count, ==, 1);
p_val = NULL;
call_count = 0;
NM_SET_OUT (p_val, do_test_nm_set_out_called (&call_count));
g_assert_cmpint (call_count, ==, 0);
shared: re-define _G_BOOLEAN_EXPR() to allow nesting g_assert() g_assert() uses G_LIKELY(), which in turn uses _G_BOOLEAN_EXPR(). As glib's version of _G_BOOLEAN_EXPR() uses a local variable _g_boolean_var_, we cannot nest a G_LIKELY() inside a G_LIKELY(), or inside a g_assert(), or a g_assert() inside a g_assert(). Workaround that, by redefining the macro. I already encountered this problem before, when having a nm_assert() inside a ({...}) block, inside a g_assert(). Then I just avoided that combination, but this situation is quite easy to encounter.
2016-10-11 12:23:02 +02:00
/* test that we successfully re-defined _G_BOOLEAN_EXPR() */
#define _T1(a) \
({ \
libnm/tests: fix bug in test Fixes: 6b904a51ee3e6dcb8926d963d901e6689d4f9fad
2016-10-11 14:08:36 +02:00
g_assert (a > 2); \
shared: re-define _G_BOOLEAN_EXPR() to allow nesting g_assert() g_assert() uses G_LIKELY(), which in turn uses _G_BOOLEAN_EXPR(). As glib's version of _G_BOOLEAN_EXPR() uses a local variable _g_boolean_var_, we cannot nest a G_LIKELY() inside a G_LIKELY(), or inside a g_assert(), or a g_assert() inside a g_assert(). Workaround that, by redefining the macro. I already encountered this problem before, when having a nm_assert() inside a ({...}) block, inside a g_assert(). Then I just avoided that combination, but this situation is quite easy to encounter.
2016-10-11 12:23:02 +02:00
a; \
})
g_assert (_T1 (3) > 1);
#undef _T1
shared: add test for NM_SET_OUT()
2016-09-26 16:59:30 +02:00
}
all: modify line separator comments to be 80 chars wide sed 's#^/\*\{5\}\*\+/$#/*****************************************************************************/#' $(git grep -l '\*\{5\}' | grep '\.[hc]$') -i
2016-10-02 18:22:50 +02:00
/*****************************************************************************/
shared: add test for NM_SET_OUT()
2016-09-26 16:59:30 +02:00
shared: minor cleanup of nm_utils_get_start_time_for_pid()
2018-05-25 15:38:21 +02:00
static void
test_get_start_time_for_pid (void)
{
guint64 x_start_time;
char x_state;
pid_t x_ppid;
x_start_time = nm_utils_get_start_time_for_pid (getpid (), &x_state, &x_ppid);
g_assert (x_start_time > 0);
g_assert (x_ppid == getppid ());
g_assert (!NM_IN_SET (x_state, '\0', ' '));
}
/*****************************************************************************/
shared: support zero arguments for NM_NARG() macro It relies on the GCC extension ##__VA_ARGS__, but we do that on various places already. Also add a test.
2018-07-10 08:45:39 +02:00
static void
test_nm_va_args_macros (void)
{
#define GET_NARG_1(...) \
NM_NARG (__VA_ARGS__)
g_assert_cmpint ( 0, ==, GET_NARG_1 ());
g_assert_cmpint ( 1, ==, GET_NARG_1 (x));
g_assert_cmpint ( 2, ==, GET_NARG_1 ( , ));
g_assert_cmpint ( 2, ==, GET_NARG_1 ( , x));
g_assert_cmpint ( 2, ==, GET_NARG_1 (x, ));
g_assert_cmpint ( 2, ==, GET_NARG_1 (x, x));
g_assert_cmpint ( 3, ==, GET_NARG_1 ( , , ));
g_assert_cmpint ( 3, ==, GET_NARG_1 ( , , x));
g_assert_cmpint ( 3, ==, GET_NARG_1 ( , x, ));
g_assert_cmpint ( 3, ==, GET_NARG_1 ( , x, x));
g_assert_cmpint ( 3, ==, GET_NARG_1 (x, , ));
g_assert_cmpint ( 3, ==, GET_NARG_1 (x, , x));
g_assert_cmpint ( 3, ==, GET_NARG_1 (x, x, ));
g_assert_cmpint ( 3, ==, GET_NARG_1 (x, x, x));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , , , ));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , , , x));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , , x, ));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , , x, x));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , x, , ));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , x, , x));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , x, x, ));
g_assert_cmpint ( 4, ==, GET_NARG_1 ( , x, x, x));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, , , ));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, , , x));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, , x, ));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, , x, x));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, x, , ));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, x, , x));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, x, x, ));
g_assert_cmpint ( 4, ==, GET_NARG_1 (x, x, x, x));
g_assert_cmpint ( 5, ==, GET_NARG_1 (x, x, x, x, x));
g_assert_cmpint ( 6, ==, GET_NARG_1 (x, x, x, x, x, x));
g_assert_cmpint ( 7, ==, GET_NARG_1 (x, x, x, x, x, x, x));
g_assert_cmpint ( 8, ==, GET_NARG_1 (x, x, x, x, x, x, x, x));
g_assert_cmpint ( 9, ==, GET_NARG_1 (x, x, x, x, x, x, x, x, x));
g_assert_cmpint (10, ==, GET_NARG_1 (x, x, x, x, x, x, x, x, x, x));
G_STATIC_ASSERT_EXPR (0 == GET_NARG_1 ());
G_STATIC_ASSERT_EXPR (1 == GET_NARG_1 (x));
G_STATIC_ASSERT_EXPR (2 == GET_NARG_1 (x, x));
}
/*****************************************************************************/
libnm, cli, ifcfg-rh: add NMSettingEthtool setting Note that in NetworkManager API (D-Bus, libnm, and nmcli), the features are called "feature-xyz". The "feature-" prefix is used, because NMSettingEthtool possibly will gain support for options that are not only -K|--offload|--features, for example -C|--coalesce. The "xzy" suffix is either how ethtool utility calls the feature ("tso", "rx"). Or, if ethtool utility specifies no alias for that feature, it's the name from kernel's ETH_SS_FEATURES ("tx-tcp6-segmentation"). If possible, we prefer ethtool utility's naming. Also note, how the features "feature-sg", "feature-tso", and "feature-tx" actually refer to multiple underlying kernel features at once. This too follows what ethtool utility does. The functionality is not yet implemented server-side.
2018-07-16 23:37:55 +02:00
static void
test_ethtool_offload (void)
{
const NMEthtoolData *d;
g_assert_cmpint (nm_ethtool_id_get_by_name ("invalid"), ==, NM_ETHTOOL_ID_UNKNOWN);
g_assert_cmpint (nm_ethtool_id_get_by_name ("feature-rx"), ==, NM_ETHTOOL_ID_FEATURE_RX);
d = nm_ethtool_data_get_by_optname (NM_ETHTOOL_OPTNAME_FEATURE_RXHASH);
g_assert (d);
g_assert_cmpint (d->id, ==, NM_ETHTOOL_ID_FEATURE_RXHASH);
g_assert_cmpstr (d->optname, ==, NM_ETHTOOL_OPTNAME_FEATURE_RXHASH);
}
/*****************************************************************************/
shared: add nm_utils_g_main_context_create_integrate_source() for integrating a GMainContext in another We will rework NMClient entirely. Then, the synchronous initialization will also use the asynchronous code paths. The difference will be that with synchronous initialization, all D-Bus interaction will be done with an internal GMainContext as current thread default, and that internal context will run until initialization completes. Note that even after initialization completes, it cannot be swapped back to the user's (outer) GMainContext. That is because contexts are essentially the queue for our D-Bus events, and we cannot swap from one queue to the other in a race free manner (or a full resync). In other words, the two contexts are not in sync, so after using the internal context NMClient needs to stick to that (at least, until the name owner gets lost, which gives an opportunity to resync and switch back to the user's main context). We thus need to hook the internal (inner) GMainContext with the user's (outer) context, so when the user iterates the outer context, events on the inner context get dispatched. Add nm_utils_g_main_context_create_integrate_source() to create such a GSource for integrating two contexts. Note that the use-case here is limited: the integrated, inner main context must not be explicitly iterated except from being dispatched by the integrating source. Otherwise, you'd get recursive runs, possible deadlocks and general ugliness. NMClient must show restrain how to use the inner context while it is integrated.
2019-11-02 16:55:43 +01:00
typedef struct {
GMainLoop *loop1;
GMainContext *c2;
GSource *extra_sources[2];
bool got_signal[5];
int fd_2;
} IntegData;
static gboolean
_test_integrate_cb_handle (IntegData *d, int signal)
{
int i;
g_assert (d);
g_assert (signal >= 0);
g_assert (signal < G_N_ELEMENTS (d->got_signal));
g_assert (!d->got_signal[signal]);
d->got_signal[signal] = TRUE;
for (i = 0; i < G_N_ELEMENTS (d->got_signal); i++) {
if (!d->got_signal[i])
break;
}
if (i == G_N_ELEMENTS (d->got_signal))
g_main_loop_quit (d->loop1);
return G_SOURCE_REMOVE;
}
static gboolean
_test_integrate_cb_timeout_1 (gpointer user_data)
{
return _test_integrate_cb_handle (user_data, 0);
}
static gboolean
_test_integrate_cb_fd_2 (int fd,
GIOCondition condition,
gpointer user_data)
{
IntegData *d = user_data;
g_assert (d->got_signal[1]);
g_assert (d->got_signal[2]);
g_assert (d->got_signal[3]);
g_assert (d->extra_sources[0]);
g_assert (d->extra_sources[1]);
return _test_integrate_cb_handle (d, 4);
}
static gboolean
_test_integrate_cb_idle_2 (gpointer user_data)
{
IntegData *d = user_data;
GSource *extra_source;
g_assert (d->got_signal[1]);
g_assert (d->got_signal[2]);
g_assert (d->extra_sources[0]);
g_assert (!d->extra_sources[1]);
all: use nm_g_unix_fd_source_new() instead of g_unix_fd_source_new() Its source-func argument has the right signature. Otherwise, this is an easy to make mistake.
2020-01-01 10:30:27 +01:00
extra_source = nm_g_unix_fd_source_new (d->fd_2,
G_IO_IN,
G_PRIORITY_DEFAULT,
_test_integrate_cb_fd_2,
d,
NULL);
shared: add nm_utils_g_main_context_create_integrate_source() for integrating a GMainContext in another We will rework NMClient entirely. Then, the synchronous initialization will also use the asynchronous code paths. The difference will be that with synchronous initialization, all D-Bus interaction will be done with an internal GMainContext as current thread default, and that internal context will run until initialization completes. Note that even after initialization completes, it cannot be swapped back to the user's (outer) GMainContext. That is because contexts are essentially the queue for our D-Bus events, and we cannot swap from one queue to the other in a race free manner (or a full resync). In other words, the two contexts are not in sync, so after using the internal context NMClient needs to stick to that (at least, until the name owner gets lost, which gives an opportunity to resync and switch back to the user's main context). We thus need to hook the internal (inner) GMainContext with the user's (outer) context, so when the user iterates the outer context, events on the inner context get dispatched. Add nm_utils_g_main_context_create_integrate_source() to create such a GSource for integrating two contexts. Note that the use-case here is limited: the integrated, inner main context must not be explicitly iterated except from being dispatched by the integrating source. Otherwise, you'd get recursive runs, possible deadlocks and general ugliness. NMClient must show restrain how to use the inner context while it is integrated.
2019-11-02 16:55:43 +01:00
g_source_attach (extra_source, d->c2);
d->extra_sources[1] = extra_source;
return _test_integrate_cb_handle (d, 3);
}
static gboolean
_test_integrate_cb_idle_1 (gpointer user_data)
{
IntegData *d = user_data;
GSource *extra_source;
g_assert (d->got_signal[2]);
g_assert (!d->extra_sources[0]);
extra_source = g_idle_source_new ();
g_source_set_callback (extra_source, _test_integrate_cb_idle_2, d, NULL);
g_source_attach (extra_source, d->c2);
d->extra_sources[0] = extra_source;
return _test_integrate_cb_handle (d, 1);
}
static gboolean
_test_integrate_cb_fd_1 (int fd,
GIOCondition condition,
gpointer user_data)
{
IntegData *d = user_data;
g_assert (!d->got_signal[1]);
return _test_integrate_cb_handle (d, 2);
}
static gboolean
_test_integrate_maincontext_cb_idle1 (gpointer user_data)
{
guint32 *p_count = user_data;
g_assert (*p_count < 5);
(*p_count)++;
return G_SOURCE_CONTINUE;
}
static void
test_integrate_maincontext (gconstpointer test_data)
{
const guint TEST_IDX = GPOINTER_TO_UINT (test_data);
GMainContext *c1 = g_main_context_default ();
nm_auto_unref_gmaincontext GMainContext *c2 = g_main_context_new ();
nm_auto_destroy_and_unref_gsource GSource *integ_source = NULL;
integ_source = nm_utils_g_main_context_create_integrate_source (c2);
g_source_attach (integ_source, c1);
if (TEST_IDX == 1) {
nm_auto_destroy_and_unref_gsource GSource *idle_source_1 = NULL;
guint32 count = 0;
idle_source_1 = g_idle_source_new ();
g_source_set_callback (idle_source_1, _test_integrate_maincontext_cb_idle1, &count, NULL);
g_source_attach (idle_source_1, c2);
shared/tests/trivial: rename nmtst_main_context_iterate_until() to nmtst_main_context_iterate_until_assert() nmtst_main_context_iterate_until*() iterates until the condition is satisfied. If that doesn't happen within timeout, it fails an assertion. Rename the function to make that clearer.
2020-01-04 08:57:47 +01:00
nmtst_main_context_iterate_until_assert (c1, 2000, count == 5);
shared: add nm_utils_g_main_context_create_integrate_source() for integrating a GMainContext in another We will rework NMClient entirely. Then, the synchronous initialization will also use the asynchronous code paths. The difference will be that with synchronous initialization, all D-Bus interaction will be done with an internal GMainContext as current thread default, and that internal context will run until initialization completes. Note that even after initialization completes, it cannot be swapped back to the user's (outer) GMainContext. That is because contexts are essentially the queue for our D-Bus events, and we cannot swap from one queue to the other in a race free manner (or a full resync). In other words, the two contexts are not in sync, so after using the internal context NMClient needs to stick to that (at least, until the name owner gets lost, which gives an opportunity to resync and switch back to the user's main context). We thus need to hook the internal (inner) GMainContext with the user's (outer) context, so when the user iterates the outer context, events on the inner context get dispatched. Add nm_utils_g_main_context_create_integrate_source() to create such a GSource for integrating two contexts. Note that the use-case here is limited: the integrated, inner main context must not be explicitly iterated except from being dispatched by the integrating source. Otherwise, you'd get recursive runs, possible deadlocks and general ugliness. NMClient must show restrain how to use the inner context while it is integrated.
2019-11-02 16:55:43 +01:00
}
if (TEST_IDX == 2) {
nm_auto_destroy_and_unref_gsource GSource *main_timeout_source = NULL;
nm_auto_destroy_and_unref_gsource GSource *timeout_source_1 = NULL;
nm_auto_destroy_and_unref_gsource GSource *idle_source_1 = NULL;
nm_auto_destroy_and_unref_gsource GSource *fd_source_1 = NULL;
nm_auto_unref_gmainloop GMainLoop *loop1 = NULL;
nm_auto_close int fd_1 = -1;
nm_auto_close int fd_2 = -1;
IntegData d;
int i;
main_timeout_source = g_timeout_source_new (3000);
g_source_set_callback (main_timeout_source, nmtst_g_source_assert_not_called, NULL, NULL);
g_source_attach (main_timeout_source, c1);
loop1 = g_main_loop_new (c1, FALSE);
d = (IntegData) {
.loop1 = loop1,
.c2 = c2,
};
fd_1 = open ("/dev/null", O_RDONLY | O_CLOEXEC);
g_assert (fd_1 >= 0);
all: use nm_g_unix_fd_source_new() instead of g_unix_fd_source_new() Its source-func argument has the right signature. Otherwise, this is an easy to make mistake.
2020-01-01 10:30:27 +01:00
fd_source_1 = nm_g_unix_fd_source_new (fd_1,
G_IO_IN,
G_PRIORITY_DEFAULT,
_test_integrate_cb_fd_1,
&d,
NULL);
shared: add nm_utils_g_main_context_create_integrate_source() for integrating a GMainContext in another We will rework NMClient entirely. Then, the synchronous initialization will also use the asynchronous code paths. The difference will be that with synchronous initialization, all D-Bus interaction will be done with an internal GMainContext as current thread default, and that internal context will run until initialization completes. Note that even after initialization completes, it cannot be swapped back to the user's (outer) GMainContext. That is because contexts are essentially the queue for our D-Bus events, and we cannot swap from one queue to the other in a race free manner (or a full resync). In other words, the two contexts are not in sync, so after using the internal context NMClient needs to stick to that (at least, until the name owner gets lost, which gives an opportunity to resync and switch back to the user's main context). We thus need to hook the internal (inner) GMainContext with the user's (outer) context, so when the user iterates the outer context, events on the inner context get dispatched. Add nm_utils_g_main_context_create_integrate_source() to create such a GSource for integrating two contexts. Note that the use-case here is limited: the integrated, inner main context must not be explicitly iterated except from being dispatched by the integrating source. Otherwise, you'd get recursive runs, possible deadlocks and general ugliness. NMClient must show restrain how to use the inner context while it is integrated.
2019-11-02 16:55:43 +01:00
g_source_attach (fd_source_1, c2);
fd_2 = open ("/dev/null", O_RDONLY | O_CLOEXEC);
g_assert (fd_2 >= 0);
d.fd_2 = fd_2;
idle_source_1 = g_idle_source_new ();
g_source_set_callback (idle_source_1, _test_integrate_cb_idle_1, &d, NULL);
g_source_attach (idle_source_1, c2);
timeout_source_1 = g_timeout_source_new (5);
g_source_set_callback (timeout_source_1, _test_integrate_cb_timeout_1, &d, NULL);
g_source_attach (timeout_source_1, c2);
g_main_loop_run (loop1);
for (i = 0; i < G_N_ELEMENTS (d.extra_sources); i++) {
g_assert (d.extra_sources[i]);
nm_clear_pointer (&d.extra_sources[i], nm_g_source_destroy_and_unref);
}
}
}
/*****************************************************************************/
libnm/tests: test nm_ip_addr_zero is all-zero and compares to IP addresse as expected
2020-01-09 10:34:30 +01:00
static void
test_nm_ip_addr_zero (void)
{
in_addr_t a4 = nmtst_inet4_from_string ("0.0.0.0");
struct in6_addr a6 = *nmtst_inet6_from_string ("::");
char buf[NM_UTILS_INET_ADDRSTRLEN];
shared: add NM_IP_ADDR_ZERO macro for initializing NMIPAddr to zero
2020-01-09 12:32:28 +01:00
NMIPAddr a = NM_IP_ADDR_INIT;
g_assert (memcmp (&a, &nm_ip_addr_zero, sizeof (a)) == 0);
libnm/tests: test nm_ip_addr_zero is all-zero and compares to IP addresse as expected
2020-01-09 10:34:30 +01:00
g_assert (IN6_IS_ADDR_UNSPECIFIED (&nm_ip_addr_zero.addr6));
g_assert (memcmp (&nm_ip_addr_zero.addr6, &in6addr_any, sizeof (in6addr_any)) == 0);
g_assert (memcmp (&nm_ip_addr_zero, &a4, sizeof (a4)) == 0);
g_assert (memcmp (&nm_ip_addr_zero, &a6, sizeof (a6)) == 0);
all: use _nm_utils_inet4_ntop() instead of nm_utils_inet4_ntop() and _nm_utils_inet6_ntop() instead of nm_utils_inet6_ntop(). nm_utils_inet4_ntop()/nm_utils_inet6_ntop() are public API of libnm. For one, that means they are only available in code that links with libnm/libnm-core. But such basic helpers should be available everywhere. Also, they accept NULL as destination buffers. We keep that behavior for potential libnm users, but internally we never want to use the static buffers. This patch needs to take care that there are no callers of _nm_utils_inet[46]_ntop() that pass NULL buffers. Also, _nm_utils_inet[46]_ntop() are inline functions and the compiler can get rid of them. We should consistently use the same variant of the helper. The only downside is that the "good" name is already taken. The leading underscore is rather ugly and inconsistent. Also, with our internal variants we can use "static array indices in function parameter declarations" next. Thereby the compiler helps to ensure that the provided buffers are of the right size.
2020-01-09 11:29:27 +01:00
g_assert_cmpstr (_nm_utils_inet4_ntop (nm_ip_addr_zero.addr4, buf), ==, "0.0.0.0");
g_assert_cmpstr (_nm_utils_inet6_ntop (&nm_ip_addr_zero.addr6, buf), ==, "::");
libnm/tests: test nm_ip_addr_zero is all-zero and compares to IP addresse as expected
2020-01-09 10:34:30 +01:00
g_assert_cmpstr (nm_utils_inet_ntop (AF_INET, &nm_ip_addr_zero, buf), ==, "0.0.0.0");
g_assert_cmpstr (nm_utils_inet_ntop (AF_INET6, &nm_ip_addr_zero, buf), ==, "::");
shared: add NM_IP_ADDR_ZERO macro for initializing NMIPAddr to zero
2020-01-09 12:32:28 +01:00
G_STATIC_ASSERT_EXPR (sizeof (a) == sizeof (a.array));
libnm/tests: test nm_ip_addr_zero is all-zero and compares to IP addresse as expected
2020-01-09 10:34:30 +01:00
}
libnm-core: use different ifname validation function for OVS bridges, ports and interfaces OVS bridges and ports do not have the length limitation of 15 bytes, the only requirements are that all chars must be alphanumeric and not be forward or backward slashes. For OVS interfaces only 'patch' types do not have the length limit, all the other types do (according to whether they have a corresponding kernel link or not). Add related unit test. https://bugzilla.redhat.com/show_bug.cgi?id=1788432
2020-02-06 17:17:49 +01:00
static void
test_connection_ovs_ifname (gconstpointer test_data)
{
const guint TEST_CASE = GPOINTER_TO_UINT (test_data);
gs_unref_object NMConnection *con = NULL;
NMSettingConnection *s_con = NULL;
NMSettingOvsBridge *s_ovs_bridge = NULL;
NMSettingOvsPort *s_ovs_port = NULL;
NMSettingOvsInterface *s_ovs_iface = NULL;
NMSettingOvsPatch *s_ovs_patch = NULL;
const char *ovs_iface_type = NULL;
switch (TEST_CASE) {
case 1:
con = nmtst_create_minimal_connection ("test_connection_ovs_ifname_bridge",
NULL,
NM_SETTING_OVS_BRIDGE_SETTING_NAME, &s_con);
s_ovs_bridge = nm_connection_get_setting_ovs_bridge (con);
g_assert (s_ovs_bridge);
break;
case 2:
con = nmtst_create_minimal_connection ("test_connection_ovs_ifname_port",
NULL,
NM_SETTING_OVS_PORT_SETTING_NAME, &s_con);
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER,
"master0",
NM_SETTING_CONNECTION_SLAVE_TYPE,
NM_SETTING_OVS_BRIDGE_SETTING_NAME,
NULL);
s_ovs_port = nm_connection_get_setting_ovs_port (con);
g_assert (s_ovs_port);
break;
case 3:
con = nmtst_create_minimal_connection ("test_connection_ovs_ifname_interface_patch",
NULL,
NM_SETTING_OVS_INTERFACE_SETTING_NAME, &s_con);
s_ovs_iface = nm_connection_get_setting_ovs_interface (con);
g_assert (s_ovs_iface);
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER,
"master0",
NM_SETTING_CONNECTION_SLAVE_TYPE,
NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_iface,
NM_SETTING_OVS_INTERFACE_TYPE,
"patch",
NULL);
s_ovs_patch = NM_SETTING_OVS_PATCH (nm_setting_ovs_patch_new());
g_assert (s_ovs_patch);
g_object_set (s_ovs_patch,
NM_SETTING_OVS_PATCH_PEER, "1.2.3.4",
NULL);
nm_connection_add_setting (con, NM_SETTING (s_ovs_patch));
s_ovs_patch = nm_connection_get_setting_ovs_patch (con);
g_assert (s_ovs_patch);
ovs_iface_type = "patch";
break;
case 4:
con = nmtst_create_minimal_connection ("test_connection_ovs_ifname_interface_internal",
NULL,
NM_SETTING_OVS_INTERFACE_SETTING_NAME, &s_con);
s_ovs_iface = nm_connection_get_setting_ovs_interface (con);
g_assert (s_ovs_iface);
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER,
"master0",
NM_SETTING_CONNECTION_SLAVE_TYPE,
NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_iface,
NM_SETTING_OVS_INTERFACE_TYPE,
"internal",
NULL);
ovs_iface_type = "internal";
break;
case 5:
con = nmtst_create_minimal_connection ("test_connection_ovs_ifname_interface_system",
NULL,
NM_SETTING_WIRED_SETTING_NAME, &s_con);
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER,
"master0",
NM_SETTING_CONNECTION_SLAVE_TYPE,
NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
s_ovs_iface = NM_SETTING_OVS_INTERFACE (nm_setting_ovs_interface_new());
g_assert (s_ovs_iface);
g_object_set (s_ovs_iface,
NM_SETTING_OVS_INTERFACE_TYPE,
"system",
NULL);
nm_connection_add_setting (con, NM_SETTING (s_ovs_iface));
s_ovs_iface = nm_connection_get_setting_ovs_interface (con);
g_assert (s_ovs_iface);
ovs_iface_type = "system";
break;
case 6:
con = nmtst_create_minimal_connection ("test_connection_ovs_ifname_interface_dpdk",
NULL,
NM_SETTING_OVS_INTERFACE_SETTING_NAME, &s_con);
s_ovs_iface = nm_connection_get_setting_ovs_interface (con);
g_assert (s_ovs_iface);
g_object_set (s_con,
NM_SETTING_CONNECTION_MASTER,
"master0",
NM_SETTING_CONNECTION_SLAVE_TYPE,
NM_SETTING_OVS_PORT_SETTING_NAME,
NULL);
g_object_set (s_ovs_iface,
NM_SETTING_OVS_INTERFACE_TYPE,
"dpdk",
NULL);
ovs_iface_type = "dpdk";
break;
}
if (!nm_streq0 (ovs_iface_type, "system")) {
/* wrong: contains backward slash */
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "ovs\\0",
NULL);
nmtst_assert_connection_unnormalizable (con,
NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
/* wrong: contains forward slash */
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "ovs/0",
NULL);
nmtst_assert_connection_unnormalizable (con,
NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
}
/* wrong: contains space */
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "ovs 0",
NULL);
nmtst_assert_connection_unnormalizable (con,
NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
/* good */
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "ovs0",
NULL);
nmtst_assert_connection_verifies (con);
nm-shared-utils: relax ovs ifname check to accept any (non-space) ASCII printable char quoting 'man ovs-vswitchd.conf.db': "The name must be alphanumeric and must not contain forward or backward slashes." OVS actually accepts a wider range of chars (all printable UTF-8 chars), NetworkManager restricts this to ASCII char as it's a safer option for now since OVS is not well documented on this matter. https://bugzilla.redhat.com/show_bug.cgi?id=1788432 Fixes: e7d72a14f6 ('libnm-core: use different ifname validation function for OVS bridges, ports and interfaces')
2020-02-18 12:16:02 +01:00
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "ovs-br0",
NULL);
nmtst_assert_connection_verifies (con);
libnm-core: use different ifname validation function for OVS bridges, ports and interfaces OVS bridges and ports do not have the length limitation of 15 bytes, the only requirements are that all chars must be alphanumeric and not be forward or backward slashes. For OVS interfaces only 'patch' types do not have the length limit, all the other types do (according to whether they have a corresponding kernel link or not). Add related unit test. https://bugzilla.redhat.com/show_bug.cgi?id=1788432
2020-02-06 17:17:49 +01:00
/* good if bridge, port, or patch interface */
g_object_set (s_con,
NM_SETTING_CONNECTION_INTERFACE_NAME, "ovs123123123123130123123",
NULL);
if (!ovs_iface_type || nm_streq (ovs_iface_type, "patch"))
nmtst_assert_connection_verifies (con);
else {
nmtst_assert_connection_unnormalizable (con,
NM_CONNECTION_ERROR,
NM_CONNECTION_ERROR_INVALID_PROPERTY);
}
}
shared: add nm_utils_strsplit_quoted() We want to parse "/proc/cmdline". That is space separated with support for quoting and escaping. Our implementation becomes part of stable behavior, and we should interpret the kernel command line the same way as the system does. That means, our implementation should match systemd's.
2020-06-22 19:43:35 +02:00
/*****************************************************************************/
static gboolean
_strsplit_quoted_char_needs_escaping (char ch)
{
return NM_IN_SET (ch, '\'', '\"', '\\')
|| strchr (NM_ASCII_WHITESPACES, ch);
}
static char *
_strsplit_quoted_create_str_rand (gssize len)
{
NMStrBuf strbuf = NM_STR_BUF_INIT (nmtst_get_rand_uint32 () % 200, nmtst_get_rand_bool ());
g_assert (len >= -1);
if (len == -1)
len = nmtst_get_rand_word_length (NULL);
while (len-- > 0) {
char ch;
ch = nmtst_rand_select ('a', ' ', '\\', '"', '\'', nmtst_get_rand_uint32 () % 255 + 1);
g_assert (ch);
nm_str_buf_append_c (&strbuf, ch);
}
if (!strbuf.allocated)
nm_str_buf_maybe_expand (&strbuf, 1, nmtst_get_rand_bool ());
return nm_str_buf_finalize (&strbuf, NULL);
}
static char **
_strsplit_quoted_create_strv_rand (void)
{
guint len = nmtst_get_rand_word_length (NULL);
char **ptr;
guint i;
libnm-core: use different ifname validation function for OVS bridges, ports and interfaces OVS bridges and ports do not have the length limitation of 15 bytes, the only requirements are that all chars must be alphanumeric and not be forward or backward slashes. For OVS interfaces only 'patch' types do not have the length limit, all the other types do (according to whether they have a corresponding kernel link or not). Add related unit test. https://bugzilla.redhat.com/show_bug.cgi?id=1788432
2020-02-06 17:17:49 +01:00
shared: add nm_utils_strsplit_quoted() We want to parse "/proc/cmdline". That is space separated with support for quoting and escaping. Our implementation becomes part of stable behavior, and we should interpret the kernel command line the same way as the system does. That means, our implementation should match systemd's.
2020-06-22 19:43:35 +02:00
ptr = g_new (char *, len + 1);
for (i = 0; i < len; i++)
ptr[i] = _strsplit_quoted_create_str_rand (-1);
ptr[i] = NULL;
return ptr;
}
static char *
_strsplit_quoted_join_strv_rand (const char *const*strv)
{
NMStrBuf strbuf = NM_STR_BUF_INIT (nmtst_get_rand_uint32 () % 200, nmtst_get_rand_bool ());
char *result;
gsize l;
gsize l2;
gsize *p_l2 = nmtst_get_rand_bool () ? &l2 : NULL;
gsize i;
g_assert (strv);
nm_str_buf_append_c_repeated (&strbuf, ' ', nmtst_get_rand_word_length (NULL) / 4);
for (i = 0; strv[i]; i++) {
const char *s = strv[i];
gsize j;
char quote;
nm_str_buf_append_c_repeated (&strbuf, ' ', 1 + nmtst_get_rand_word_length (NULL) / 4);
j = 0;
quote = '\0';
while (TRUE) {
char ch = s[j++];
/* extract_first_word*/
if (quote != '\0') {
if (ch == '\0') {
nm_str_buf_append_c (&strbuf, quote);
break;
}
if ( ch == quote
|| ch == '\\'
|| nmtst_get_rand_uint32 () % 5 == 0)
nm_str_buf_append_c (&strbuf, '\\');
nm_str_buf_append_c (&strbuf, ch);
if (nmtst_get_rand_uint32 () % 3 == 0) {
nm_str_buf_append_c (&strbuf, quote);
quote = '\0';
goto next_maybe_quote;
}
continue;
}
if (ch == '\0') {
if (s == strv[i]) {
quote = nmtst_rand_select ('\'', '"');
nm_str_buf_append_c_repeated (&strbuf, quote, 2);
}
break;
}
if ( _strsplit_quoted_char_needs_escaping (ch)
|| nmtst_get_rand_uint32 () % 5 == 0)
nm_str_buf_append_c (&strbuf, '\\');
nm_str_buf_append_c (&strbuf, ch);
next_maybe_quote:
if (nmtst_get_rand_uint32 () % 5 == 0) {
quote = nmtst_rand_select ('\'', '\"');
nm_str_buf_append_c (&strbuf, quote);
if (nmtst_get_rand_uint32 () % 5 == 0) {
nm_str_buf_append_c (&strbuf, quote);
quote = '\0';
}
}
}
}
nm_str_buf_append_c_repeated (&strbuf, ' ', nmtst_get_rand_word_length (NULL) / 4);
nm_str_buf_maybe_expand (&strbuf, 1, nmtst_get_rand_bool ());
l = strbuf.len;
result = nm_str_buf_finalize (&strbuf, p_l2);
g_assert (!p_l2 || l == *p_l2);
g_assert (strlen (result) == l);
return result;
}
static void
_strsplit_quoted_assert_strv (const char *topic,
const char *str,
const char *const*strv1,
const char *const*strv2)
{
nm_auto_str_buf NMStrBuf s1 = { };
nm_auto_str_buf NMStrBuf s2 = { };
gs_free char *str_escaped = NULL;
int i;
g_assert (str);
g_assert (strv1);
g_assert (strv2);
if (_nm_utils_strv_equal ((char **) strv1, (char **) strv2))
return;
for (i = 0; strv1[i]; i++) {
gs_free char *s = g_strescape (strv1[i], NULL);
g_print (">>> [%s] strv1[%d] = \"%s\"\n", topic, i, s);
if (i > 0)
nm_str_buf_append_c (&s1, ' ');
nm_str_buf_append_printf (&s1, "\"%s\"", s);
}
for (i = 0; strv2[i]; i++) {
gs_free char *s = g_strescape (strv2[i], NULL);
g_print (">>> [%s] strv2[%d] = \"%s\"\n", topic, i, s);
if (i > 0)
nm_str_buf_append_c (&s2, ' ');
nm_str_buf_append_printf (&s2, "\"%s\"", s);
}
nm_str_buf_maybe_expand (&s1, 1, FALSE);
nm_str_buf_maybe_expand (&s2, 1, FALSE);
str_escaped = g_strescape (str, NULL);
g_error ("compared words differs: [%s] str=\"%s\"; strv1=%s; strv2=%s", topic, str_escaped, nm_str_buf_get_str (&s1), nm_str_buf_get_str (&s2));
}
static void
_strsplit_quoted_test (const char *str,
const char *const*strv_expected)
{
gs_strfreev char **strv_systemd = NULL;
gs_strfreev char **strv_nm = NULL;
int r;
g_assert (str);
r = nmtst_systemd_extract_first_word_all (str, &strv_systemd);
g_assert_cmpint (r, ==, 1);
g_assert (strv_systemd);
if (!strv_expected)
strv_expected = (const char *const*) strv_systemd;
_strsplit_quoted_assert_strv ("systemd", str, strv_expected, (const char *const*) strv_systemd);
strv_nm = nm_utils_strsplit_quoted (str);
g_assert (strv_nm);
_strsplit_quoted_assert_strv ("nm", str, strv_expected, (const char *const*) strv_nm);
}
static void
test_strsplit_quoted (void)
{
int i_run;
_strsplit_quoted_test ("", NM_MAKE_STRV ());
_strsplit_quoted_test (" ", NM_MAKE_STRV ());
_strsplit_quoted_test (" ", NM_MAKE_STRV ());
_strsplit_quoted_test (" \t", NM_MAKE_STRV ());
_strsplit_quoted_test ("a b", NM_MAKE_STRV ("a", "b"));
_strsplit_quoted_test ("a\\ b", NM_MAKE_STRV ("a b"));
_strsplit_quoted_test (" a\\ \"b\"", NM_MAKE_STRV ("a b"));
_strsplit_quoted_test (" a\\ \"b\" c \n", NM_MAKE_STRV ("a b", "c"));
for (i_run = 0; i_run < 1000; i_run++) {
gs_strfreev char **strv = NULL;
gs_free char *str = NULL;
/* create random strv array and join them carefully so that splitting
* them will yield the original value. */
strv = _strsplit_quoted_create_strv_rand ();
str = _strsplit_quoted_join_strv_rand ((const char *const*) strv);
_strsplit_quoted_test (str, (const char *const*) strv);
}
/* Create random words and assert that systemd and our implementation can
* both split them (and in the exact same way). */
for (i_run = 0; i_run < 1000; i_run++) {
gs_free char *s = _strsplit_quoted_create_str_rand (nmtst_get_rand_uint32 () % 150);
_strsplit_quoted_test (s, NULL);
}
}
libnm-core: use different ifname validation function for OVS bridges, ports and interfaces OVS bridges and ports do not have the length limitation of 15 bytes, the only requirements are that all chars must be alphanumeric and not be forward or backward slashes. For OVS interfaces only 'patch' types do not have the length limit, all the other types do (according to whether they have a corresponding kernel link or not). Add related unit test. https://bugzilla.redhat.com/show_bug.cgi?id=1788432
2020-02-06 17:17:49 +01:00
libnm/tests: test nm_ip_addr_zero is all-zero and compares to IP addresse as expected
2020-01-09 10:34:30 +01:00
/*****************************************************************************/
libnm: refactor wifi frequency handling in libnm - mark global variables as const. This allows the linker to mark the variable as read only. - for nm_utils_wifi_[25]ghz_freqs(), don't generate a list based on bg_table/a_table. Instead, keep static array of frequencies. Since we have unit tests that check the consistency, this has little maintenance effort. - add unit tests
2020-06-21 21:59:32 +02:00
static void
_do_wifi_ghz_freqs (const guint *freqs, const char *band)
{
int len;
int j;
int i;
g_assert (NM_IN_STRSET (band, "a", "bg"));
g_assert (freqs);
g_assert (freqs[0] != 0);
for (i = 0; freqs[i]; i++) {
for (j = 0; j < i; j++)
g_assert (freqs[i] != freqs[j]);
}
len = i;
g_assert (nm_utils_wifi_freq_to_channel (0) == 0);
g_assert (nm_utils_wifi_channel_to_freq (0, "bg") == -1);
g_assert (nm_utils_wifi_channel_to_freq (0, "foo") == 0);
g_assert (!nm_utils_wifi_is_channel_valid (0, "bg"));
g_assert (!nm_utils_wifi_is_channel_valid (0, "foo"));
for (i = 0; i < len; i++) {
guint freq = freqs[i];
guint32 chan;
guint32 freq2;
chan = nm_utils_wifi_freq_to_channel (freq);
g_assert (chan != 0);
freq2 = nm_utils_wifi_channel_to_freq (chan, band);
g_assert (freq2 == freq);
g_assert (nm_utils_wifi_is_channel_valid (chan, band));
}
g_assert (freqs[len] == 0);
}
static void
test_nm_utils_wifi_ghz_freqs (void)
{
_do_wifi_ghz_freqs (nm_utils_wifi_2ghz_freqs (), "bg");
_do_wifi_ghz_freqs (nm_utils_wifi_5ghz_freqs (), "a");
}
/*****************************************************************************/
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NMTST_DEFINE ();
int main (int argc, char **argv)
{
nmtst_init (&argc, &argv, TRUE);
shared: add NM_ASCII_SPACES macro
2019-03-20 10:43:54 +01:00
g_test_add_func ("/core/general/test_nm_ascii_spaces", test_nm_ascii_spaces);
all: refactor hashing by introducing NMHashState The privious NM_HASH_* macros directly operated on a guint value and were thus close to the actual implementation. Replace them by adding a NMHashState struct and accessors to update the hash state. This hides the implementation better and would allow us to carry more state. For example, we could switch to siphash24() transparently. For now, we still do a form basically djb2 hashing, albeit with differing start seed. Also add nm_hash_str() and nm_str_hash(): - nm_hash_str() is our own string hashing implementation - nm_str_hash() is our own string implementation, but with a GHashFunc signature, suitable to pass it to g_hash_table_new(). Also, it has this name in order to remind you of g_str_hash(), which it is replacing.
2017-10-13 14:00:22 +02:00
g_test_add_func ("/core/general/test_nm_hash", test_nm_hash);
shared: add nm_g_slice_free_fcn() util Useful, when you need a GDestroyNotify function for g_slice_free() of a certain type.
2017-08-15 12:32:37 +02:00
g_test_add_func ("/core/general/test_nm_g_slice_free_fcn", test_nm_g_slice_free_fcn);
c-list: add c_list_sort() Add a stable, recursive merge sort for CList. This could be improved by doing an iterative implementation. The recursive implementation's stack depth is not an issue, as it is bound by O(ln(n)). But an iterative implementation would safe the overhead of O(n*log(n)) function calls and be potentially faster.
2017-07-08 13:30:00 +02:00
g_test_add_func ("/core/general/test_c_list_sort", test_c_list_sort);
shared: add NMDedupMultiIndex "nm-dedup-multi.h" Add the NMDedupMultiIndex cache. It basically tracks objects as doubly linked list. With the addition that each object and the list head is indexed by a hash table. Also, it supports tracking multiple distinct lists, all indexed by the idx-type instance. It also deduplicates the tracked objects and shares them. - the objects that can be put into the cache must be immutable and ref-counted. That is, the cache will deduplicate them and share the reference. Also, as these objects are immutable and ref-counted, it is safe that users outside the cache own them too (as long as they keep them immutable and manage their reference properly). The deduplication uses obj_id_hash_func() and obj_id_equal_func(). These functions must cover *every* aspect of the objects when comparing equality. For example nm_platform_ip4_route_cmp() would be a function that qualifies as obj_id_equal_func(). The cache creates references to the objects as needed and gives them back. This happens via obj_get_ref() and obj_put_ref(). Note that obj_get_ref() is free to create a new object, for example to convert a stack-allocated object to a (ref-counted) heap allocated one. The deduplication process creates NMDedupIndexBox instances which are the ref-counted entity. In principle, the objects themself don't need to be ref-counted as that is handled by the boxing instance. - The cache doesn't only do deduplication. It is a multi-index, meaning, callers add objects using a index handle NMDedupMultiIdxType. The NMDedupMultiIdxType instance is the access handle to lookup the list and objects inside the cache. Note that the idx-type instance may partition the objects in distinct lists. For all operations there are cross-references and hash table lookups. Hence, every operation of this data structure is O(1) and the memory overhead for an index tracking an object is constant. The cache preserves ordering (due to linked list) and exposes the list as public API. This allows users to iterate the list without any additional copying of elements.
2017-06-04 22:43:21 +02:00
g_test_add_func ("/core/general/test_dedup_multi", test_dedup_multi);
shared: add nm_utils_str_utf8safe_*() API to sanitize UTF-8 strings Use C-style backslash escaping to sanitize non-UTF-8 strings. The functions are compatible with glib's g_strcompress() and g_strescape(). The difference is only that g_strescape() escapes all non-printable, non ASCII character as well, while nm_utils_str_utf8safe_escape() -- depending on the flags -- preserves valid UTF-8 sequence except backslash. The flags allow to optionally escape ASCII control characters and all non-ASCII (valid UTF-8) characters. But the option to preserve valid UTF-8 (non-ASCII) characters verbatim, is what distinguishes from g_strescape().
2017-05-16 18:50:21 +02:00
g_test_add_func ("/core/general/test_utils_str_utf8safe", test_utils_str_utf8safe);
shared: add nm_utils_strsplit_set() helper A replacement for g_strsplit_set(). While g_strsplit_set() does (n+1) malloc and n slice allocations, this needs roughtly (O(log(n))) mallocs. Another difference from g_strsplit_set() is that this function treats multiple delimiters as one (and thus never returns empty words). While I can see that sometimes you may want to keep empty words (like parsing a CSV file and preserve empty cells), we usually use this function for splitting user input. In such case, we want to treat multiple delimiters as one.
2017-09-14 19:14:01 +02:00
g_test_add_func ("/core/general/test_nm_utils_strsplit_set", test_nm_utils_strsplit_set);
shared: add nm_utils_escaped_tokens_options_*() API This will be used for splitting and escaping option parameters in nmcli (vpn.data).
2020-03-27 09:29:43 +01:00
g_test_add_func ("/core/general/test_nm_utils_escaped_tokens", test_nm_utils_escaped_tokens);
macros: rework NM_IN_SET() macro to expand to a plain logical expression Let the preprocessor do more work, but generate a simple expression that the compiler can optimize (presumably) better.
2015-08-12 13:04:12 +02:00
g_test_add_func ("/core/general/test_nm_in_set", test_nm_in_set);
macros: add NM_IN_STRSET() Add macro similar to NM_IN_SET() that checks for C strings. NULL values are allowed and handled as one would expect.
2016-02-11 16:53:06 +01:00
g_test_add_func ("/core/general/test_nm_in_strset", test_nm_in_strset);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_setting_vpn_items", test_setting_vpn_items);
g_test_add_func ("/core/general/test_setting_vpn_update_secrets", test_setting_vpn_update_secrets);
g_test_add_func ("/core/general/test_setting_vpn_modify_during_foreach", test_setting_vpn_modify_during_foreach);
g_test_add_func ("/core/general/test_setting_ip4_config_labels", test_setting_ip4_config_labels);
libnm-core: change how new and legacy properties are serialized Although libnm filters out properties received from the daemon that it doesn't understand, there may be other clients that do not. In particular, a client might call GetSettings() on a connection, update the ipv4.addresses property in the returned dictionary, and then pass the dictionary to Update(). In that case, the updated dictionary would contain ipv4.address-data, but it would not reflect the changes the client intended to make. Fix this by changing the daemon side to prefer the legacy properties to the new ones if both are set, and changing the client side to not send the legacy properties (since we don't support new clients talking to old servers anyway).
2014-11-14 11:46:19 -05:00
g_test_add_func ("/core/general/test_setting_ip4_config_address_data", test_setting_ip4_config_address_data);
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
g_test_add_func ("/core/general/test_setting_ip_route_attributes", test_setting_ip_route_attributes);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_setting_gsm_apn_spaces", test_setting_gsm_apn_spaces);
g_test_add_func ("/core/general/test_setting_gsm_apn_bad_chars", test_setting_gsm_apn_bad_chars);
g_test_add_func ("/core/general/test_setting_gsm_apn_underscore", test_setting_gsm_apn_underscore);
g_test_add_func ("/core/general/test_setting_gsm_without_number", test_setting_gsm_without_number);
libnm/wwan: add GSM setting device-id, sim-id, and sim-operator-id properties These properties limit whether the connection applies to a certain WWAN modem based on the modem's device ID or SIM ID (as reported by the WWAN management service), or through the MCC/MNC ID of the operator that issued the SIM card.
2015-10-01 12:44:28 -05:00
g_test_add_func ("/core/general/test_setting_gsm_sim_operator_id", test_setting_gsm_sim_operator_id);
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
g_test_add_func ("/core/general/test_setting_to_dbus_all", test_setting_to_dbus_all);
g_test_add_func ("/core/general/test_setting_to_dbus_no_secrets", test_setting_to_dbus_no_secrets);
g_test_add_func ("/core/general/test_setting_to_dbus_only_secrets", test_setting_to_dbus_only_secrets);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
g_test_add_func ("/core/general/test_setting_to_dbus_transform", test_setting_to_dbus_transform);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
g_test_add_func ("/core/general/test_setting_to_dbus_enum", test_setting_to_dbus_enum);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_setting_compare_id", test_setting_compare_id);
libnm: fix comparing NMSettingIPConfig for address and route properties When comparing settings, nm_setting_compare() performs a complicated logic, which basically serializes each GObject property to a GVariant for the D-Bus representation. That is wrong for example for ipv4.addresses, which don't contain address labels. That is, the GObject property is called "addresses", but the D-Bus field "addresses" cannot encode every information and thus comparison fails. Instead, it would have to look into "address-data". Traditionally, we have virtual functions like compare_property() per NMSetting to do the comparison. That comparison is based on the GObject properties. I think that is wrong, because we should have a generic concept of what a property is, independent from GObject properties. With libnm, we added NMSettingProperty, which indeed is such an GObject independent representation to define properties. However, it is not used thoroughly, instead compare_property() is a hack of special cases, overloads from NMSettingProperty, overloads of compare_property(), and default behavior based on GParamSpec. This should be cleaned up. For now, just hack it by handle the properties with the problems explicitly.
2016-06-11 00:31:26 +02:00
g_test_add_func ("/core/general/test_setting_compare_addresses", test_setting_compare_addresses);
g_test_add_func ("/core/general/test_setting_compare_routes", test_setting_compare_routes);
device: extend MAC address handling including randomization for ethernet and wifi Extend the "ethernet.cloned-mac-address" and "wifi.cloned-mac-address" settings. Instead of specifying an explicit MAC address, the additional special values "permanent", "preserve", "random", "random-bia", "stable" and "stable-bia" are supported. "permanent" means to use the permanent hardware address. Previously that was the default if no explict cloned-mac-address was set. The default is thus still "permanent", but it can be overwritten by global configuration. "preserve" means not to configure the MAC address when activating the device. That was actually the default behavior before introducing MAC address handling with commit 1b49f941a69af910b0e68530be7339e8053068e5. "random" and "random-bia" use a randomized MAC address for each connection. "stable" and "stable-bia" use a generated, stable address based on some token. The "bia" suffix says to generate a burned-in address. The stable method by default uses as token the connection UUID, but the token can be explicitly choosen via "stable:<TOKEN>" and "stable-bia:<TOKEN>". On a D-Bus level, the "cloned-mac-address" is a bytestring and thus cannot express the new forms. It is replaced by the new "assigned-mac-address" field. For the GObject property, libnm's API, nmcli, keyfile, etc. the old name "cloned-mac-address" is still used. Deprecating the old field seems more complicated then just extending the use of the existing "cloned-mac-address" field, although the name doesn't match well with the extended meaning. There is some overlap with the "wifi.mac-address-randomization" setting. https://bugzilla.gnome.org/show_bug.cgi?id=705545 https://bugzilla.gnome.org/show_bug.cgi?id=708820 https://bugzilla.gnome.org/show_bug.cgi?id=758301
2016-05-24 15:57:16 +02:00
g_test_add_func ("/core/general/test_setting_compare_wired_cloned_mac_address", test_setting_compare_wired_cloned_mac_address);
g_test_add_func ("/core/general/test_setting_compare_wirless_cloned_mac_address", test_setting_compare_wireless_cloned_mac_address);
libnm-core: add NM_SETTING_COMPARE_FLAG_IGNORE_TIMESTAMP flag for ignoring timestamp when comparing settings or connections.
2014-11-18 14:20:12 +01:00
g_test_add_func ("/core/general/test_setting_compare_timestamp", test_setting_compare_timestamp);
test: fix duplicate test names New glib complains.
2015-09-03 18:15:46 +02:00
#define ADD_FUNC(name, func, secret_flags, comp_flags, remove_secret) \
g_test_add_data_func_full ("/core/general/" G_STRINGIFY (func) "_" name, \
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
test_data_compare_secrets_new (secret_flags, comp_flags, remove_secret), \
func, g_free)
test: fix duplicate test names New glib complains.
2015-09-03 18:15:46 +02:00
ADD_FUNC ("agent_owned", test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_AGENT_OWNED, NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS, TRUE);
ADD_FUNC ("not_saved", test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_NOT_SAVED, NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS, TRUE);
ADD_FUNC ("secrets", test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_IGNORE_SECRETS, TRUE);
ADD_FUNC ("exact", test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_EXACT, FALSE);
ADD_FUNC ("agent_owned", test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_AGENT_OWNED, NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS, TRUE);
ADD_FUNC ("not_saved", test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_NOT_SAVED, NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS, TRUE);
ADD_FUNC ("secrets", test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_IGNORE_SECRETS, TRUE);
ADD_FUNC ("exact", test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_EXACT, FALSE);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_setting_old_uuid", test_setting_old_uuid);
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
g_test_add_func ("/core/general/test_connection_to_dbus_setting_name", test_connection_to_dbus_setting_name);
libnm-core: drop previously-deprecated NMSetting properties Drop the NMSetting properties that were marked deprecated in libnm-util in 0.9.10, but use nm_setting_class_add_dbus_property() to deal with them appropriately when serializing/deserializing.
2014-07-29 18:42:02 -04:00
g_test_add_func ("/core/general/test_connection_to_dbus_deprecated_props", test_connection_to_dbus_deprecated_props);
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
g_test_add_func ("/core/general/test_setting_new_from_dbus", test_setting_new_from_dbus);
libnm-core: change all mac-address properties to G_TYPE_STRING Make all mac-address properties (including NMSettingBluetooth:bdaddr, NMSettingOlpcMesh:dhcp-anycast-addr, and NMSettingWireless:bssid) be strings, using _nm_setting_class_transform_property() to handle translating to/from binary form when dealing with D-Bus. Update everything accordingly for the change, and also add a test for transformed setting properties to test-general.
2014-07-30 10:57:45 -04:00
g_test_add_func ("/core/general/test_setting_new_from_dbus_transform", test_setting_new_from_dbus_transform);
libnm: make use of GParamSpecFlags and GParamSpecEnum Make enum- and flags-valued properties use GParamSpecEnum and GParamSpecFlags, for better introspectability/bindability. This requires no changes outside libnm-core/libnm since the expected data size is still the same with g_object_get()/g_object_set(), and GLib will internally convert between int/uint and enum/flags GValues when using g_object_get_property()/g_object_set_property().
2014-06-26 16:47:46 -04:00
g_test_add_func ("/core/general/test_setting_new_from_dbus_enum", test_setting_new_from_dbus_enum);
libnm-core: validate property types from D-Bus (rh #1182567) In _nm_setting_new_from_dbus(), verify that the properties have the right types, and return an error if not. (In particular, don't crash if someone tries to assign a GBytes-valued property a non-'ay' value.)
2015-01-15 12:19:06 -05:00
g_test_add_func ("/core/general/test_setting_new_from_dbus_bad", test_setting_new_from_dbus_bad);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_connection_replace_settings", test_connection_replace_settings);
g_test_add_func ("/core/general/test_connection_replace_settings_from_connection", test_connection_replace_settings_from_connection);
libnm-core: clean up nm_connection_replace_settings()'s semantics On failure, nm_connection_replace_settings() would leave the connection in an undefined state. Fix it so that either (a) the settings are replaced and the resulting connection is valid and we return TRUE, or (b) the connection is untouched and we return FALSE and an error. (And add a test case for this.)
2014-08-22 10:14:38 -04:00
g_test_add_func ("/core/general/test_connection_replace_settings_bad", test_connection_replace_settings_bad);
libnm-core: rename NMConnection to/from_hash methods Rename nm_connection_to_hash() to nm_connection_to_dbus(), and nm_connection_new_from_hash() to nm_connection_new_from_dbus(). In addition to clarifying that this is specifically the D-Bus serialization format, these names will also work better in the GDBus-based future where the serialization format is GVariant, not GHashTable. Also, move NMSettingHashFlags to nm-connection.h, and rename it NMConnectionSerializationFlags.
2014-08-06 19:35:31 -04:00
g_test_add_func ("/core/general/test_connection_new_from_dbus", test_connection_new_from_dbus);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_connection_normalize_virtual_iface_name", test_connection_normalize_virtual_iface_name);
libnm: normalize missing connection UUID Extend nm_connection_normalize() to add a connection UUID in case it is unset.
2014-12-01 13:03:23 +01:00
g_test_add_func ("/core/general/test_connection_normalize_uuid", test_connection_normalize_uuid);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_connection_normalize_type", test_connection_normalize_type);
g_test_add_func ("/core/general/test_connection_normalize_slave_type_1", test_connection_normalize_slave_type_1);
g_test_add_func ("/core/general/test_connection_normalize_slave_type_2", test_connection_normalize_slave_type_2);
libnm-core: add normalize of MTU for NMSettingInfiniband Previously, NMSettingInfiniband:verify() silently modifies the setting for invalid MTU. verify() should not do that. For libnm-core we can change behavior and implement normalization of MTU. This changes behavior for NMSettingInfiniband:verify() so that MTU gets no longer fixed by verify() alone. Instead verify() fails with a verification error. Due the possibility to normalize the MTU, NM still can receive invalid settings and fix it. For libnm-core we don't change behavior, merely add a code comment. Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-13 02:35:26 +02:00
g_test_add_func ("/core/general/test_connection_normalize_infiniband_mtu", test_connection_normalize_infiniband_mtu);
libnm-core: remove gateway when never-default=yes in NMSettingIPConfig Having a gateway defined when never-default=yes causes troubles in connection matching and anyway makes no sense. If the combination is found, remove the gateway during the normalization phase. https://bugzilla.redhat.com/show_bug.cgi?id=1313091
2016-04-22 13:45:08 +02:00
g_test_add_func ("/core/general/test_connection_normalize_gateway_never_default", test_connection_normalize_gateway_never_default);
libnm-core: ip-config: normalize may-fail for disabled IP methods Since commit 7d1709d7f649 ("device: check may_fail when progressing to IP_CHECK") NM correctly checks the may-fail properties to decide whether a connection must fail after the completion of IP configuration. But for ipv4.method=disabled and ipv6.method=ignore the IP configuration is always considered failed and thus setting may-fail=no results in a connection that can never succeed. To prevent such wrong configuration, force may-fail to TRUE for those methods during connection normalization. https://bugzilla.redhat.com/show_bug.cgi?id=1334884
2016-07-01 17:05:42 +02:00
g_test_add_func ("/core/general/test_connection_normalize_may_fail", test_connection_normalize_may_fail);
libnm-core: drop extra IPs from shared connections during normalization The core only consider the first address for shared connections, don't pretend we accept multiple addresses. This change doesn't prevent supporting multiple addresses in the future. https://bugzilla.gnome.org/show_bug.cgi?id=763937
2016-09-12 18:51:00 +02:00
g_test_add_func ("/core/general/test_connection_normalize_shared_addresses", test_connection_normalize_shared_addresses);
libnm: fix normalizing and verifying OVS connections Normalizing can be complicated, as settings depend on each other and possibly conflict. That is, because verify() must exactly anticipate whether normalization will succeed and how the result will look like. That is because we only want to modify the connection, if we are sure that the result will verify. Hence, verify() and normalize() are strongly related. The implementation should not be spread out between NMSettingOvsInterface:verify(), NMSettingOvsPatch:verify() and _normalize_ovs_interface_type(). Also, add some unit-tests.
2017-10-27 14:30:18 +02:00
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_system/1", GUINT_TO_POINTER (1), test_connection_normalize_ovs_interface_type_system);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_system/2", GUINT_TO_POINTER (2), test_connection_normalize_ovs_interface_type_system);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_system/3", GUINT_TO_POINTER (3), test_connection_normalize_ovs_interface_type_system);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_system/4", GUINT_TO_POINTER (4), test_connection_normalize_ovs_interface_type_system);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_system/5", GUINT_TO_POINTER (5), test_connection_normalize_ovs_interface_type_system);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_system/6", GUINT_TO_POINTER (6), test_connection_normalize_ovs_interface_type_system);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_system/7", GUINT_TO_POINTER (7), test_connection_normalize_ovs_interface_type_system);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/1", GUINT_TO_POINTER (1), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/2", GUINT_TO_POINTER (2), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/3", GUINT_TO_POINTER (3), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/4", GUINT_TO_POINTER (4), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/5", GUINT_TO_POINTER (5), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/6", GUINT_TO_POINTER (6), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/7", GUINT_TO_POINTER (7), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/8", GUINT_TO_POINTER (8), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/9", GUINT_TO_POINTER (9), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/10", GUINT_TO_POINTER (10), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/11", GUINT_TO_POINTER (11), test_connection_normalize_ovs_interface_type_ovs_interface);
g_test_add_data_func ("/core/general/test_connection_normalize_ovs_interface_type_ovs_interface/12", GUINT_TO_POINTER (12), test_connection_normalize_ovs_interface_type_ovs_interface);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
libnm-core: use different ifname validation function for OVS bridges, ports and interfaces OVS bridges and ports do not have the length limitation of 15 bytes, the only requirements are that all chars must be alphanumeric and not be forward or backward slashes. For OVS interfaces only 'patch' types do not have the length limit, all the other types do (according to whether they have a corresponding kernel link or not). Add related unit test. https://bugzilla.redhat.com/show_bug.cgi?id=1788432
2020-02-06 17:17:49 +01:00
g_test_add_data_func ("/core/general/test_connection_ovs_ifname/1", GUINT_TO_POINTER (1), test_connection_ovs_ifname);
g_test_add_data_func ("/core/general/test_connection_ovs_ifname/2", GUINT_TO_POINTER (2), test_connection_ovs_ifname);
g_test_add_data_func ("/core/general/test_connection_ovs_ifname/3", GUINT_TO_POINTER (3), test_connection_ovs_ifname);
g_test_add_data_func ("/core/general/test_connection_ovs_ifname/4", GUINT_TO_POINTER (4), test_connection_ovs_ifname);
g_test_add_data_func ("/core/general/test_connection_ovs_ifname/5", GUINT_TO_POINTER (5), test_connection_ovs_ifname);
g_test_add_data_func ("/core/general/test_connection_ovs_ifname/6", GUINT_TO_POINTER (6), test_connection_ovs_ifname);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_setting_connection_permissions_helpers", test_setting_connection_permissions_helpers);
g_test_add_func ("/core/general/test_setting_connection_permissions_property", test_setting_connection_permissions_property);
g_test_add_func ("/core/general/test_connection_compare_same", test_connection_compare_same);
g_test_add_func ("/core/general/test_connection_compare_key_only_in_a", test_connection_compare_key_only_in_a);
g_test_add_func ("/core/general/test_connection_compare_setting_only_in_a", test_connection_compare_setting_only_in_a);
g_test_add_func ("/core/general/test_connection_compare_key_only_in_b", test_connection_compare_key_only_in_b);
g_test_add_func ("/core/general/test_connection_compare_setting_only_in_b", test_connection_compare_setting_only_in_b);
g_test_add_func ("/core/general/test_connection_diff_a_only", test_connection_diff_a_only);
g_test_add_func ("/core/general/test_connection_diff_same", test_connection_diff_same);
g_test_add_func ("/core/general/test_connection_diff_different", test_connection_diff_different);
g_test_add_func ("/core/general/test_connection_diff_no_secrets", test_connection_diff_no_secrets);
g_test_add_func ("/core/general/test_connection_diff_inferrable", test_connection_diff_inferrable);
g_test_add_func ("/core/general/test_connection_good_base_types", test_connection_good_base_types);
g_test_add_func ("/core/general/test_connection_bad_base_types", test_connection_bad_base_types);
g_test_add_func ("/core/general/test_hwaddr_aton_ether_normal", test_hwaddr_aton_ether_normal);
g_test_add_func ("/core/general/test_hwaddr_aton_ib_normal", test_hwaddr_aton_ib_normal);
g_test_add_func ("/core/general/test_hwaddr_aton_no_leading_zeros", test_hwaddr_aton_no_leading_zeros);
g_test_add_func ("/core/general/test_hwaddr_aton_malformed", test_hwaddr_aton_malformed);
g_test_add_func ("/core/general/test_hwaddr_equal", test_hwaddr_equal);
libnm-core: add nm_utils_hwaddr_canonical() Add a helper function for converting hardware addresses to canonical form.
2014-10-28 08:32:25 -04:00
g_test_add_func ("/core/general/test_hwaddr_canonical", test_hwaddr_canonical);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
g_test_add_func ("/core/general/test_ip4_prefix_to_netmask", test_ip4_prefix_to_netmask);
g_test_add_func ("/core/general/test_ip4_netmask_to_prefix", test_ip4_netmask_to_prefix);
g_test_add_func ("/core/general/test_connection_changed_signal", test_connection_changed_signal);
g_test_add_func ("/core/general/test_setting_connection_changed_signal", test_setting_connection_changed_signal);
g_test_add_func ("/core/general/test_setting_bond_changed_signal", test_setting_bond_changed_signal);
g_test_add_func ("/core/general/test_setting_ip4_changed_signal", test_setting_ip4_changed_signal);
g_test_add_func ("/core/general/test_setting_ip6_changed_signal", test_setting_ip6_changed_signal);
g_test_add_func ("/core/general/test_setting_vlan_changed_signal", test_setting_vlan_changed_signal);
g_test_add_func ("/core/general/test_setting_vpn_changed_signal", test_setting_vpn_changed_signal);
g_test_add_func ("/core/general/test_setting_wired_changed_signal", test_setting_wired_changed_signal);
g_test_add_func ("/core/general/test_setting_wireless_changed_signal", test_setting_wireless_changed_signal);
g_test_add_func ("/core/general/test_setting_wireless_security_changed_signal", test_setting_wireless_security_changed_signal);
g_test_add_func ("/core/general/test_setting_802_1x_changed_signal", test_setting_802_1x_changed_signal);
libnm-core: add NMSettingIPConfig:gateway, drop NMIPAddress:gateway The gateway is a global property of the IPv4/IPv6 configuration, not an attribute of any particular address. So represent it as such in the API; remove the gateway from NMIPAddress, and add it to NMSettingIPConfig. Behind the scenes, the gateway is still serialized along with the first address in NMSettingIPConfig:addresses, and is deserialized from that if the settings dictionary doesn't contain a 'gateway' key. Adjust nmcli's interactive mode to prompt for IP addresses and gateway separately. (Patch partly from Jirka Klimeš.)
2014-10-20 21:30:56 -04:00
g_test_add_func ("/core/general/test_setting_ip4_gateway", test_setting_ip4_gateway);
g_test_add_func ("/core/general/test_setting_ip6_gateway", test_setting_ip6_gateway);
libnm/test: test compare_property() to handle default values Add a test case to check whether compare/diff works correctly.
2015-05-20 17:53:25 +02:00
g_test_add_func ("/core/general/test_setting_compare_default_strv", test_setting_compare_default_strv);
libnm: add NMSettingUser This only adds new API for a NMSettingUser. The setting class is still entirely unused. The point is getting the new API into 1.8.0 release of libnm. It's easier to backport the use of the API to a stable branch then backporting public API. https://bugzilla.gnome.org/show_bug.cgi?id=776276 https://bugzilla.redhat.com/show_bug.cgi?id=1421429
2017-03-24 12:41:04 +01:00
g_test_add_func ("/core/general/test_setting_user_data", test_setting_user_data);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
libnm: add NMSockAddrEndpoint API NMSockAddrEndpoint is an immutable structure that contains the endpoint string of a service. It also includes the (naive) parsing of the host and port/service parts. This will be used for the endpoint of WireGuard's peers. But since endpoints are not something specific to WireGuard, give it a general name (and purpose) independent from WireGuard. Essentially, this structure takes a string in a manner that libnm understands, and uses it for node and service arguments for getaddrinfo(). NMSockAddrEndpoint allows to have endpoints that are not parsable into a host and port part. That is useful because our settings need to be able to hold invalid values. That is for forward compatibility (server sends a new endpoint format) and for better error handling (have invalid settings that can be constructed without loss, but fail later during the NMSetting:verify() step).
2018-12-29 13:01:28 +01:00
g_test_add_func ("/core/general/test_sock_addr_endpoint", test_sock_addr_endpoint);
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
g_test_add_func ("/core/general/hexstr2bin", test_hexstr2bin);
shared: add nm_strquote() util We already have nm_strquote_a(). That is useful, but uses alloca(), hence it is ill suited to be called from a macro, inside a loop, or from a function that should be inlined. Instead, add nm_strquote() that has the same purpose but writes to a provided string buffer.
2017-10-05 14:27:24 +02:00
g_test_add_func ("/core/general/nm_strquote", test_nm_strquote);
libnm/test: add test for nm_utils_uuid_generate_from_string()
2014-11-28 14:26:24 +01:00
g_test_add_func ("/core/general/test_nm_utils_uuid_generate_from_string", test_nm_utils_uuid_generate_from_string);
libnm: move _nm_utils_uuid_generate_from_strings() from src/ to libnm-core/
2015-02-22 20:15:52 +01:00
g_test_add_func ("/core/general/_nm_utils_uuid_generate_from_strings", test_nm_utils_uuid_generate_from_strings);
libnm,core: enhance nm_utils_hexstr2bin() Make the type return GBytes since most in-tree users want that. Allow the function to accept many more formats as valid hex, including bytes delimited by ':' and a leading '0x'.
2014-11-06 17:51:09 -06:00
libnm: move _nm_utils_ascii_str_to_int64() from src/ to libnm-core/
2015-02-22 11:54:03 +01:00
g_test_add_func ("/core/general/_nm_utils_ascii_str_to_int64", test_nm_utils_ascii_str_to_int64);
utils: add nm_utils_is_power_of_two() macro
2015-05-12 09:24:05 +02:00
g_test_add_func ("/core/general/nm_utils_is_power_of_two", test_nm_utils_is_power_of_two);
shared: move nm_utils_ptrarray_find_binary_search() to shared utils
2018-08-01 11:03:02 +02:00
g_test_add_func ("/core/general/nm_utils_ptrarray_find_binary_search", test_nm_utils_ptrarray_find_binary_search);
g_test_add_func ("/core/general/nm_utils_ptrarray_find_binary_search_with_duplicates", test_nm_utils_ptrarray_find_binary_search_with_duplicates);
libnm: add NMUtilsStrStrDictKey utility When having a hash-of-hashes where each hash is indexed by a name, (such as GKeyFile), you can either implement it as a hash-of-hashes or define your own version of indexes that pack both levels of names into one key. This is an implementation of such a key. Use it as: GHashTable *hash = g_hash_table_new_full (_nm_utils_strstrdictkey_hash, _nm_utils_strstrdictkey_equal, g_free, _destroy_value); and create keys via: NMUtilsStrStrDictKey *k = _nm_utils_strstrdictkey_create (s1, s2); For lookup you can use static strings (note that the static string might increase the size of the binary): g_hash_table_contains (hash, _nm_utils_strstrdictkey_static ("outer", "inner"))
2015-05-22 18:52:22 +02:00
g_test_add_func ("/core/general/_nm_utils_strstrdictkey", test_nm_utils_strstrdictkey);
shared: add NM_PTRARRAY_LEN() utility macro I used to use g_strv_length ((char **) p) instead, but that feels ugly because it g_strv_length() is not designed to operate on arbitrary pointer arrays.
2017-04-06 09:47:05 +02:00
g_test_add_func ("/core/general/nm_ptrarray_len", test_nm_ptrarray_len);
libnm: move _nm_utils_ascii_str_to_int64() from src/ to libnm-core/
2015-02-22 11:54:03 +01:00
libnm-core/test: add tests for DNS option utilities
2015-04-01 11:47:42 +02:00
g_test_add_func ("/core/general/_nm_utils_dns_option_validate", test_nm_utils_dns_option_validate);
g_test_add_func ("/core/general/_nm_utils_dns_option_find_idx", test_nm_utils_dns_option_find_idx);
libnm-core: use jansson to compare and check team configurations Optionally link libnm-core against jansson JSON library and use it to validate and compare team configurations.
2016-03-11 13:17:24 +01:00
g_test_add_func ("/core/general/_nm_utils_validate_json", test_nm_utils_check_valid_json);
g_test_add_func ("/core/general/_nm_utils_team_config_equal", test_nm_utils_team_config_equal);
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
g_test_add_func ("/core/general/test_nm_utils_enum", test_nm_utils_enum);
shared: add test for NM_SET_OUT()
2016-09-26 16:59:30 +02:00
g_test_add_func ("/core/general/nm-set-out", test_nm_set_out);
libnm-core: define known route attribute names and validation function This adds definition of a set of known route option attributes to libnm-core and helper functions. nm_ip_route_attribute_validate() performs the validation of the attribute type and, in case of a formatted string attribute, of its content. nm_ip_route_get_variant_attribute_spec() returns the attribute format specifier to be passed to nm_utils_parse_variant_attributes(). Since at the moment NMIPRoute is the only user of NMVariantAttributeSpec and the type is opaque to users of the library, the struct is extended to carry some other data useful for validation.
2017-02-15 15:06:24 +01:00
g_test_add_func ("/core/general/route_attributes/parse", test_route_attributes_parse);
g_test_add_func ("/core/general/route_attributes/format", test_route_attributes_format);
libnm: lookup route attributes attribute spec via binary search
2019-11-27 13:13:48 +01:00
g_test_add_func ("/core/general/test_variant_attribute_spec", test_variant_attribute_spec);
libnm-core: add enum conversion utilities Add functions nm_utils_enum_to_str() and nm_utils_enum_from_str() which can be used to perform conversions between enum values and strings, passing the GType automatically generated for every enum by glib-mkenums.
2015-06-04 15:05:33 +02:00
shared: minor cleanup of nm_utils_get_start_time_for_pid()
2018-05-25 15:38:21 +02:00
g_test_add_func ("/core/general/get_start_time_for_pid", test_get_start_time_for_pid);
shared: support zero arguments for NM_NARG() macro It relies on the GCC extension ##__VA_ARGS__, but we do that on various places already. Also add a test.
2018-07-10 08:45:39 +02:00
g_test_add_func ("/core/general/test_nm_va_args_macros", test_nm_va_args_macros);
libnm, cli, ifcfg-rh: add NMSettingEthtool setting Note that in NetworkManager API (D-Bus, libnm, and nmcli), the features are called "feature-xyz". The "feature-" prefix is used, because NMSettingEthtool possibly will gain support for options that are not only -K|--offload|--features, for example -C|--coalesce. The "xzy" suffix is either how ethtool utility calls the feature ("tso", "rx"). Or, if ethtool utility specifies no alias for that feature, it's the name from kernel's ETH_SS_FEATURES ("tx-tcp6-segmentation"). If possible, we prefer ethtool utility's naming. Also note, how the features "feature-sg", "feature-tso", and "feature-tx" actually refer to multiple underlying kernel features at once. This too follows what ethtool utility does. The functionality is not yet implemented server-side.
2018-07-16 23:37:55 +02:00
g_test_add_func ("/core/general/test_ethtool_offload", test_ethtool_offload);
shared: support zero arguments for NM_NARG() macro It relies on the GCC extension ##__VA_ARGS__, but we do that on various places already. Also add a test.
2018-07-10 08:45:39 +02:00
shared: add nm_utils_g_main_context_create_integrate_source() for integrating a GMainContext in another We will rework NMClient entirely. Then, the synchronous initialization will also use the asynchronous code paths. The difference will be that with synchronous initialization, all D-Bus interaction will be done with an internal GMainContext as current thread default, and that internal context will run until initialization completes. Note that even after initialization completes, it cannot be swapped back to the user's (outer) GMainContext. That is because contexts are essentially the queue for our D-Bus events, and we cannot swap from one queue to the other in a race free manner (or a full resync). In other words, the two contexts are not in sync, so after using the internal context NMClient needs to stick to that (at least, until the name owner gets lost, which gives an opportunity to resync and switch back to the user's main context). We thus need to hook the internal (inner) GMainContext with the user's (outer) context, so when the user iterates the outer context, events on the inner context get dispatched. Add nm_utils_g_main_context_create_integrate_source() to create such a GSource for integrating two contexts. Note that the use-case here is limited: the integrated, inner main context must not be explicitly iterated except from being dispatched by the integrating source. Otherwise, you'd get recursive runs, possible deadlocks and general ugliness. NMClient must show restrain how to use the inner context while it is integrated.
2019-11-02 16:55:43 +01:00
g_test_add_data_func ("/core/general/test_integrate_maincontext/1", GUINT_TO_POINTER (1), test_integrate_maincontext);
g_test_add_data_func ("/core/general/test_integrate_maincontext/2", GUINT_TO_POINTER (2), test_integrate_maincontext);
libnm/tests: test nm_ip_addr_zero is all-zero and compares to IP addresse as expected
2020-01-09 10:34:30 +01:00
g_test_add_func ("/core/general/test_nm_ip_addr_zero", test_nm_ip_addr_zero);
libnm: refactor wifi frequency handling in libnm - mark global variables as const. This allows the linker to mark the variable as read only. - for nm_utils_wifi_[25]ghz_freqs(), don't generate a list based on bg_table/a_table. Instead, keep static array of frequencies. Since we have unit tests that check the consistency, this has little maintenance effort. - add unit tests
2020-06-21 21:59:32 +02:00
g_test_add_func ("/core/general/test_nm_utils_wifi_ghz_freqs", test_nm_utils_wifi_ghz_freqs);
libnm/tests: test nm_ip_addr_zero is all-zero and compares to IP addresse as expected
2020-01-09 10:34:30 +01:00
shared: add nm_utils_strsplit_quoted() We want to parse "/proc/cmdline". That is space separated with support for quoting and escaping. Our implementation becomes part of stable behavior, and we should interpret the kernel command line the same way as the system does. That means, our implementation should match systemd's.
2020-06-22 19:43:35 +02:00
g_test_add_func ("/core/general/test_strsplit_quoted", test_strsplit_quoted);
tests: refactor tests to use g_test framework (g_test_add_func) Signed-off-by: Thomas Haller <thaller@redhat.com>
2014-08-19 00:49:43 +02:00
return g_test_run ();
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
}
Reference in a new issue Copy permalink