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

3431 lines
124 KiB
C
Raw Normal View History

libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
/* -*- Mode: C; tab-width: 4; indent-tabs-mode: t; c-basic-offset: 4 -*- */
/*
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2, or (at your option)
* any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*
* Copyright 2008 - 2011 Red Hat, Inc.
*
*/
#include <glib.h>
#include <string.h>
#include <nm-utils.h>
#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"
#include "nm-core-internal.h"
#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-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 libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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"
#include "nm-simple-connection.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-test-utils.h"
static void
vpn_check_func (const char *key, const char *value, gpointer user_data)
{
const char *test = user_data;
if (!strcmp (key, "foobar1")) {
ASSERT (strcmp (value, "blahblah1") == 0,
test, "unexpected vpn item '%s' / '%s'", key, value);
return;
}
if (!strcmp (key, "foobar2")) {
ASSERT (strcmp (value, "blahblah2") == 0,
test, "unexpected vpn item '%s' / '%s'", key, value);
return;
}
if (!strcmp (key, "foobar3")) {
ASSERT (strcmp (value, "blahblah3") == 0,
test, "unexpected vpn item '%s' / '%s'", key, value);
return;
}
if (!strcmp (key, "foobar4")) {
ASSERT (strcmp (value, "blahblah4") == 0,
test, "unexpected vpn item '%s' / '%s'", key, value);
return;
}
ASSERT (FALSE, test, "unexpected vpn item '%s'", key);
}
static void
vpn_check_empty_func (const char *key, const char *value, gpointer user_data)
{
const char *test = user_data;
/* We don't expect any values */
ASSERT (FALSE, test, "unexpected vpn item '%s'", key);
}
static void
test_setting_vpn_items (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
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
ASSERT (s_vpn != NULL,
"vpn-items",
"error creating vpn setting");
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 */
nm_setting_vpn_foreach_data_item (s_vpn, vpn_check_func, "vpn-data");
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");
nm_setting_vpn_add_secret (s_vpn, "foobar1", "blahblah1");
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 */
nm_setting_vpn_foreach_secret (s_vpn, vpn_check_func, "vpn-secrets");
nm_setting_vpn_remove_secret (s_vpn, "foobar1");
nm_setting_vpn_remove_secret (s_vpn, "foobar2");
nm_setting_vpn_remove_secret (s_vpn, "foobar3");
nm_setting_vpn_remove_secret (s_vpn, "foobar4");
/* Try to add some blank values and make sure they are rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*key != 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_data_item (s_vpn, NULL, NULL);
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*strlen (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 ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*item != 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_data_item (s_vpn, "foobar1", NULL);
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*strlen (item) > 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, "foobar1", "");
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*key != 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_data_item (s_vpn, NULL, "blahblah1");
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*strlen (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 ();
nm_setting_vpn_foreach_data_item (s_vpn, vpn_check_empty_func, "vpn-data-empty");
/* Try to add some blank secrets and make sure they are rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*key != 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, NULL, NULL);
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*strlen (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 ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*secret != 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", NULL);
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*strlen (secret) > 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, "foobar1", "");
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*key != 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, NULL, "blahblah1");
g_test_assert_expected_messages ();
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*strlen (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 ();
nm_setting_vpn_foreach_secret (s_vpn, vpn_check_empty_func, "vpn-secrets-empty");
g_object_unref (s_vpn);
}
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: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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 (connection != NULL,
"vpn-update-secrets",
"error creating 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
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
ASSERT (s_vpn != NULL,
"vpn-update-secrets",
"error creating vpn setting");
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);
ASSERT (success == TRUE,
"vpn-update-secrets", "failed to update VPN secrets: %s", error->message);
/* Read the secrets back out */
tmp = nm_setting_vpn_get_secret (s_vpn, key1);
ASSERT (tmp != NULL,
"vpn-update-secrets", "unexpected failure getting key #1");
ASSERT (strcmp (tmp, val1) == 0,
"vpn-update-secrets", "unexpected key #1 value");
tmp = nm_setting_vpn_get_secret (s_vpn, key2);
ASSERT (tmp != NULL,
"vpn-update-secrets", "unexpected failure getting key #2");
ASSERT (strcmp (tmp, val2) == 0,
"vpn-update-secrets", "unexpected key #2 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
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;
GVariant *dict, *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-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
addr = nm_ip_address_new (AF_INET, "1.1.1.1", 24, NULL, &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
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);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_verify (NM_SETTING (s_ip4), NULL, &error);
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
addr = nm_setting_ip_config_get_address (s_ip4, 0);
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
label = nm_ip_address_get_attribute (addr, "label");
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
/* addr 2 */
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
addr = nm_ip_address_new (AF_INET, "2.2.2.2", 24, NULL, &error);
g_assert_no_error (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
nm_ip_address_set_attribute (addr, "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);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_verify (NM_SETTING (s_ip4), NULL, &error);
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
addr = nm_setting_ip_config_get_address (s_ip4, 1);
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
label = nm_ip_address_get_attribute (addr, "label");
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-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
addr = nm_ip_address_new (AF_INET, "3.3.3.3", 24, NULL, &error);
g_assert_no_error (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
nm_ip_address_set_attribute (addr, "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);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_verify (NM_SETTING (s_ip4), NULL, &error);
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
addr = nm_setting_ip_config_get_address (s_ip4, 2);
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
label = nm_ip_address_get_attribute (addr, "label");
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);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_verify (NM_SETTING (s_ip4), NULL, &error);
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
addr = nm_setting_ip_config_get_address (s_ip4, 0);
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_cmpstr (nm_ip_address_get_address (addr), ==, "2.2.2.2");
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
label = nm_ip_address_get_attribute (addr, "label");
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-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_cmpstr (nm_ip_address_get_address (addr), ==, "3.3.3.3");
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
label = nm_ip_address_get_attribute (addr, "label");
g_assert (label == NULL);
/* The labels should appear in the D-Bus serialization */
conn = nmtst_create_minimal_connection ("label 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);
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", 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 (setting_dict);
g_variant_unref (value);
g_strfreev (labels);
/* And should be deserialized */
conn = nm_simple_connection_new_from_dbus (dict, &error);
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-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_cmpstr (nm_ip_address_get_address (addr), ==, "2.2.2.2");
label = nm_ip_address_get_attribute (addr, "label");
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-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_cmpstr (nm_ip_address_get_address (addr), ==, "3.3.3.3");
label = nm_ip_address_get_attribute (addr, "label");
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);
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_verify (NM_SETTING (s_ip4), NULL, &error);
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
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-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_cmpstr (nm_ip_address_get_address (addr), ==, "2.2.2.2");
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
label = nm_ip_address_get_attribute (addr, "label");
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-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_cmpstr (nm_ip_address_get_address (addr), ==, "3.3.3.3");
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
label = nm_ip_address_get_attribute (addr, "label");
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
}
static void
test_setting_gsm_apn_spaces (void)
{
NMSettingGsm *s_gsm;
const char *tmp;
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
ASSERT (s_gsm != NULL,
"gsm-apn-spaces",
"error creating GSM setting");
/* Trailing space */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar ", NULL);
tmp = nm_setting_gsm_get_apn (s_gsm);
ASSERT (tmp != NULL,
"gsm-apn-spaces", "empty APN");
ASSERT (strcmp (tmp, "foobar") == 0,
"gsm-apn-spaces", "unexpected APN");
/* Leading space */
g_object_set (s_gsm, NM_SETTING_GSM_APN, " foobar", NULL);
tmp = nm_setting_gsm_get_apn (s_gsm);
ASSERT (tmp != NULL,
"gsm-apn-spaces", "empty APN");
ASSERT (strcmp (tmp, "foobar") == 0,
"gsm-apn-spaces", "unexpected APN");
}
static void
test_setting_gsm_apn_bad_chars (void)
{
NMSettingGsm *s_gsm;
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
ASSERT (s_gsm != NULL,
"gsm-apn-bad-chars",
"error creating GSM setting");
g_object_set (s_gsm, NM_SETTING_GSM_NUMBER, "*99#", NULL);
/* Make sure a valid APN works */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar123.-baz", NULL);
ASSERT (nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL) == TRUE,
"gsm-apn-bad-chars", "unexpectedly invalid GSM setting");
/* Random invalid chars */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "@#%$@#%@#%", NULL);
ASSERT (nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL) == FALSE,
"gsm-apn-bad-chars", "unexpectedly valid GSM setting");
/* Spaces */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar baz", NULL);
ASSERT (nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL) == FALSE,
"gsm-apn-bad-chars", "unexpectedly valid GSM setting");
/* 0 characters long */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "", NULL);
ASSERT (nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL) == FALSE,
"gsm-apn-bad-chars", "unexpectedly valid GSM setting");
/* 65-character long */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijkl1", NULL);
ASSERT (nm_setting_verify (NM_SETTING (s_gsm), NULL, NULL) == FALSE,
"gsm-apn-bad-chars", "unexpectedly valid GSM setting");
}
static void
test_setting_gsm_apn_underscore (void)
{
NMSettingGsm *s_gsm;
GError *error = NULL;
gboolean success;
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
g_assert (s_gsm);
g_object_set (s_gsm, NM_SETTING_GSM_NUMBER, "*99#", NULL);
/* 65-character long */
g_object_set (s_gsm, NM_SETTING_GSM_APN, "foobar_baz", NULL);
success = nm_setting_verify (NM_SETTING (s_gsm), NULL, &error);
g_assert_no_error (error);
g_assert (success == TRUE);
}
static void
test_setting_gsm_without_number (void)
{
NMSettingGsm *s_gsm;
GError *error = NULL;
gboolean success;
s_gsm = (NMSettingGsm *) nm_setting_gsm_new ();
g_assert (s_gsm);
g_object_set (s_gsm, NM_SETTING_GSM_NUMBER, NULL, NULL);
success = nm_setting_verify (NM_SETTING (s_gsm), NULL, &error);
g_assert_no_error (error);
g_assert (success == TRUE);
g_object_set (s_gsm, NM_SETTING_GSM_NUMBER, "", NULL);
success = nm_setting_verify (NM_SETTING (s_gsm), NULL, &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: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_error_free (error);
}
static NMSettingWirelessSecurity *
make_test_wsec_setting (const char *detail)
{
NMSettingWirelessSecurity *s_wsec;
s_wsec = (NMSettingWirelessSecurity *) nm_setting_wireless_security_new ();
ASSERT (s_wsec != NULL, detail, "error creating setting");
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-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
#define ASSERT_CONTAINS(vardict, key, test_name, msg) \
{ \
GVariant *value; \
value = g_variant_lookup_value (vardict, key, NULL); \
ASSERT (value != NULL, test_name, msg); \
g_variant_unref (value); \
}
#define ASSERT_NOT_CONTAINS(vardict, key, test_name, msg) \
{ \
GVariant *value; \
value = g_variant_lookup_value (vardict, key, NULL); \
ASSERT (value == NULL, test_name, msg); \
}
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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-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_setting_to_dbus (NM_SETTING (s_wsec), NULL, 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
/* Make sure all keys are there */
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
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_KEY_MGMT,
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
"setting-to-dbus-all", "unexpectedly missing " NM_SETTING_WIRELESS_SECURITY_KEY_MGMT);
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
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME,
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
"setting-to-dbus-all", "unexpectedly missing " NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME);
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
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_PSK,
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
"setting-to-dbus-all", "unexpectedly missing " NM_SETTING_WIRELESS_SECURITY_PSK);
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
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY0,
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
"setting-to-dbus-all", "unexpectedly missing " 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-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_setting_to_dbus (NM_SETTING (s_wsec), NULL, NM_CONNECTION_SERIALIZE_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
/* Make sure non-secret keys are there */
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
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_KEY_MGMT,
"setting-to-dbus-no-secrets", "unexpectedly missing " NM_SETTING_WIRELESS_SECURITY_KEY_MGMT);
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME,
"setting-to-dbus-no-secrets", "unexpectedly missing " 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-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
ASSERT_NOT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_PSK,
"setting-to-dbus-no-secrets", "unexpectedly present " NM_SETTING_WIRELESS_SECURITY_PSK);
ASSERT_NOT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY0,
"setting-to-dbus-no-secrets", "unexpectedly present " 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-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_setting_to_dbus (NM_SETTING (s_wsec), NULL, NM_CONNECTION_SERIALIZE_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-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 */
ASSERT_NOT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_KEY_MGMT,
"setting-to-dbus-only-secrets", "unexpectedly present " NM_SETTING_WIRELESS_SECURITY_KEY_MGMT);
ASSERT_NOT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_LEAP_USERNAME,
"setting-to-dbus-only-secrets", "unexpectedly present " 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 */
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_PSK,
"setting-to-dbus-only-secrets", "unexpectedly missing " NM_SETTING_WIRELESS_SECURITY_PSK);
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_WEP_KEY0,
"setting-to-dbus-only-secrets", "unexpectedly missing " 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-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_setting_to_dbus (s_wired, NULL, NM_CONNECTION_SERIALIZE_ALL);
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);
dict = _nm_setting_to_dbus (s_ip6, NULL, NM_CONNECTION_SERIALIZE_ALL);
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);
dict = _nm_setting_to_dbus (s_wsec, NULL, NM_CONNECTION_SERIALIZE_ALL);
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);
dict = _nm_setting_to_dbus (s_serial, NULL, NM_CONNECTION_SERIALIZE_ALL);
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: 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-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
ASSERT_CONTAINS (dict, NM_SETTING_WIRELESS_SECURITY_SETTING_NAME,
"connection-to-dbus-setting-name", "unexpectedly missing " 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");
dict = _nm_setting_to_dbus (NM_SETTING (s_wsec), NULL, 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
g_object_unref (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
s_wsec = (NMSettingWirelessSecurity *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_WIRELESS_SECURITY, dict, NULL, NULL);
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: 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_wired = _nm_setting_new_from_dbus (NM_TYPE_SETTING_WIRED, dict, NULL, &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);
s_ip6 = (NMSettingIP6Config *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_IP6_CONFIG, dict, NULL, &error);
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);
s_wsec = (NMSettingWirelessSecurity *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_WIRELESS_SECURITY, dict, NULL, &error);
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);
s_serial = (NMSettingSerial *) _nm_setting_new_from_dbus (NM_TYPE_SETTING_SERIAL, dict, NULL, &error);
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: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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-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: 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
connection = nm_simple_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));
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
check_permission (NMSettingConnection *s_con,
guint32 idx,
const char *expected_uname,
const char *tag)
{
gboolean success;
const char *ptype = NULL, *pitem = NULL, *detail = NULL;
success = nm_setting_connection_get_permission (s_con, 0, &ptype, &pitem, &detail);
ASSERT (success == TRUE, tag, "unexpected failure getting added permission");
/* Permission type */
ASSERT (ptype != NULL, tag, "unexpected failure getting permission type");
ASSERT (strcmp (ptype, "user") == 0, tag, "retrieved unexpected permission type");
/* Permission item */
ASSERT (pitem != NULL, tag, "unexpected failure getting permission item");
ASSERT (strcmp (pitem, expected_uname) == 0, tag, "retrieved unexpected permission item");
ASSERT (detail == NULL, tag, "unexpected success getting permission detail");
}
#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 */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (success == FALSE,
"setting-connection-permissions-helpers", "unexpected success adding bad permission type #1");
/* Ensure a bad [type] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*ptype*");
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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 ();
ASSERT (success == FALSE,
"setting-connection-permissions-helpers", "unexpected success adding bad permission type #2");
/* Ensure a bad [item] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*uname*");
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (success == FALSE,
"setting-connection-permissions-helpers", "unexpected success adding bad permission item #1");
/* Ensure a bad [item] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*uname[0] != '\\0'*");
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (success == FALSE,
"setting-connection-permissions-helpers", "unexpected success adding bad permission item #2");
/* Ensure an [item] with ':' is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*strchr (uname, ':')*");
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (success == FALSE,
"setting-connection-permissions-helpers", "unexpected success adding bad permission item #3");
/* Ensure a non-UTF-8 [item] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*g_utf8_validate (uname, -1, NULL)*");
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (success == FALSE,
"setting-connection-permissions-helpers", "unexpected success adding bad permission item #4");
/* Ensure a non-NULL [detail] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (success == FALSE,
"setting-connection-permissions-helpers", "unexpected success adding bad detail");
/* Ensure a valid call results in success */
success = nm_setting_connection_add_permission (s_con, "user", TEST_UNAME, NULL);
ASSERT (success == TRUE,
"setting-connection-permissions-helpers", "unexpected failure adding valid user permisson");
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 1,
"setting-connection-permissions-helpers", "unexpected failure getting number of permissions");
check_permission (s_con, 0, TEST_UNAME, "setting-connection-permissions-helpers");
/* 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);
ASSERT (perms != NULL,
"setting-connection-permissions-helpers", "unexpected failure getting permissions");
ASSERT (g_strv_length (perms) == 1,
"setting-connection-permissions-helpers", "unexpected failure getting number of permissions");
ASSERT (strcmp (perms[0], expected_perm) == 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
"setting-connection-permissions-helpers", "unexpected permission property data");
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);
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-helpers", "unexpected failure removing permission");
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 */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad permission type #1");
/* Ensure a bad [type] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad permission type #2");
/* Ensure a bad [item] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad permission item #1");
/* Ensure a bad [item] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad permission item #2");
/* Ensure an [item] with ':' in the middle is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad permission item #3");
/* Ensure an [item] with ':' at the end is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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 ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad permission item #4");
/* Ensure a non-UTF-8 [item] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*g_utf8_validate (str, -1, 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
add_permission_property (s_con, "user", buf, (int) sizeof (buf), NULL);
g_test_assert_expected_messages ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad permission item #5");
/* Ensure a non-NULL [detail] is rejected */
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*(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 ();
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected success adding bad detail");
/* Ensure a valid call results in success */
success = nm_setting_connection_add_permission (s_con, "user", TEST_UNAME, NULL);
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 1,
"setting-connection-permissions-property", "unexpected failure adding valid user permisson");
check_permission (s_con, 0, TEST_UNAME, "setting-connection-permissions-property");
/* Now remove that permission and ensure we have 0 permissions */
nm_setting_connection_remove_permission (s_con, 0);
ASSERT (nm_setting_connection_get_num_permissions (s_con) == 0,
"setting-connection-permissions-property", "unexpected failure removing permission");
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 },
{ 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: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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 },
{ NM_SETTING_CONNECTION_SECONDARIES, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_CONNECTION_GATEWAY_PING_TIMEOUT, NM_SETTING_DIFF_RESULT_IN_A },
{ 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 },
{ 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 },
{ 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 },
{ NM_SETTING_IP_CONFIG_ADDRESSES, NM_SETTING_DIFF_RESULT_IN_A },
{ NM_SETTING_IP_CONFIG_ROUTES, NM_SETTING_DIFF_RESULT_IN_A },
{ 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 libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_IP4_CONFIG_DHCP_CLIENT_ID, 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 },
{ 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 libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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_NUMBER, "*99#",
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)
{
NMSetting *old, *new;
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);
}
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: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NMSetting *old, *new;
gboolean success;
/* 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
/* Clear the PSK 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
g_object_set (new, NM_SETTING_WIRELESS_SECURITY_PSK, NULL, NULL);
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
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;
libnm: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-glib-test.c libnm/nm-glib-enum-types.[ch] cp libnm-glib/tests/*.[ch] libnm/tests/ ]
2014-07-24 08:53:33 -04:00
NMSetting *old, *new;
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));
}
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;
}
static void
test_ip4_prefix_to_netmask (void)
{
int i;
for (i = 0; i<=32; i++) {
guint32 netmask = nm_utils_ip4_prefix_to_netmask (i);
int plen = nm_utils_ip4_netmask_to_prefix (netmask);
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++) {
guint32 netmask = nm_utils_ip4_prefix_to_netmask (i);
guint32 netmask_lowest_bit = netmask & ~nm_utils_ip4_prefix_to_netmask (i-1);
g_assert_cmpint (i, ==, nm_utils_ip4_netmask_to_prefix (netmask));
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. */
prefix_holey = nm_utils_ip4_netmask_to_prefix (netmask_holey);
g_assert_cmpint (i, ==, prefix_holey);
}
}
g_rand_free (rand);
}
#define ASSERT_CHANGED(statement) \
{ \
changed = FALSE; \
statement; \
g_assert (changed); \
}
#define ASSERT_UNCHANGED(statement) \
{ \
changed = FALSE; \
statement; \
g_assert (!changed); \
}
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;
char *uuid;
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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->dns->len*");
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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->dns_search->len*");
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, 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
addr = nm_ip_address_new (AF_INET, "22.33.0.0", 24, NULL, &error);
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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->addresses->len*");
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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->routes->len*");
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, 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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->dns->len*");
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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->dns_search->len*");
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, 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
addr = nm_ip_address_new (AF_INET6, "1:2:3::4:5:6", 64, NULL, &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_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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->addresses->len*");
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
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_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*i < priv->routes->len*");
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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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));
libnm: add libnm/libnm-core (part 2) This fixes up the code from the previous "clean" import, and adds build infrastructure. [There are two slightly orthogonal sets of changes in this patch. First, the files added in the previous commit were modified as followed: # Replace internal references to "libnm-util" and "libnm-glib" with "libnm" perl -pi -e 's/libnm-(util|glib)/libnm/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] # Fix includes of the enum-types files perl -pi -e 's/nm-utils-enum-types/nm-core-enum-types/;' libnm-core/*.[ch] libnm-core/tests/*.[ch] libnm/*.[ch] libnm/tests/*.[ch] perl -pi -e 's/nm-glib-enum-types/nm-enum-types/;' libnm/*.[ch] libnm/tests/*.[ch] # Fix some python example code perl -pi -e 's/import NMClient/import NM/;' -e 's/NMClient.Client\(\)/NM.Client()/;' libnm/nm-client.c Then, the build infrastructure was added (without further modifying any existing files in libnm-core or libnm.) Note: to regenerate libnm.ver after rebase: (head -2 libnm-util/libnm-util.ver; (grep -h '\s'nm_ libnm-util/libnm-util.ver libnm-glib/libnm-glib.ver | env LANG=C sort); tail -3 libnm-util/libnm-util.ver) > libnm/libnm.ver ]
2014-07-04 13:26:57 -04:00
g_test_expect_message ("libnm", G_LOG_LEVEL_CRITICAL, "*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)
{
GError *error = NULL;
NMSetting *setting;
gboolean success;
/* 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);
success = nm_setting_verify (NM_SETTING (setting), NULL, &error);
g_assert_no_error (error);
g_assert (success == TRUE);
}
/*
* 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
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
/* If vlan.interface-name is invalid, deserialization will fail. */
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
con = nm_simple_connection_new_from_dbus (connection_dict, &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: 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_clear_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: 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: 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
con = nm_simple_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: 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
con = nm_simple_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);
g_assert (_nm_setting_type_is_base_type (base_type));
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);
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);
g_assert (_nm_setting_type_is_base_type (base_type));
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);
g_assert (_nm_setting_type_is_base_type (base_type));
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);
}
g_assert (!nm_connection_get_connection_type (con));
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);
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);
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);
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;
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",
NM_SETTING_INFINIBAND_MTU, (guint) 2044,
NULL);
nmtst_assert_connection_verifies_without_normalization (con);
g_assert_cmpint (2044, ==, nm_setting_infiniband_get_mtu (s_infini));
g_object_set (s_infini,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "datagram",
NM_SETTING_INFINIBAND_MTU, (guint) 2045,
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);
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 (2044, ==, nm_setting_infiniband_get_mtu (s_infini));
g_object_set (s_infini,
NM_SETTING_INFINIBAND_TRANSPORT_MODE, "connected",
NM_SETTING_INFINIBAND_MTU, (guint) 65520,
NULL);
nmtst_assert_connection_verifies_without_normalization (con);
g_assert_cmpint (65520, ==, nm_setting_infiniband_get_mtu (s_infini));
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);
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: add libnm/libnm-core (part 1) This commit begins creating the new "libnm", which will replace libnm-util and libnm-glib. The main reason for the libnm-util/libnm-glib split is that the daemon needs to link to libnm-util (to get NMSettings, NMConnection, etc), but can't link to libnm-glib (because it uses many of the same type names as the NetworkManager daemon. eg, NMDevice). So the daemon links to only libnm-util, but basically all clients link to both. With libnm, there will be only a single client-visible library, and NetworkManager will internally link against a private "libnm-core" containing the parts that used to be in libnm-util. (The "libnm-core" parts still need to be in their own directory so that the daemon can see those header files without also seeing the ones in libnm/ that conflict with its own headers.) [This commit just copies the source code from libnm-util/ to libnm-core/, and libnm-glib/ to libnm/: mkdir -p libnm-core/tests/ mkdir -p libnm/tests/ cp libnm-util/*.[ch] libnm-util/nm-version.h.in libnm-core/ rm -f libnm-core/nm-version.h libnm-core/nm-setting-template.[ch] libnm-core/nm-utils-enum-types.[ch] cp libnm-util/tests/*.[ch] libnm-core/tests/ cp libnm-glib/*.[ch] libnm/ rm -f libnm/libnm_glib.[ch] libnm/libnm-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);
/* The tests */
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);
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-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);
#define ADD_FUNC(func, secret_flags, comp_flags, remove_secret) \
g_test_add_data_func_full ("/core/general/" G_STRINGIFY (func), \
test_data_compare_secrets_new (secret_flags, comp_flags, remove_secret), \
func, g_free)
ADD_FUNC (test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_AGENT_OWNED, NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS, TRUE);
ADD_FUNC (test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_NOT_SAVED, NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS, TRUE);
ADD_FUNC (test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_IGNORE_SECRETS, TRUE);
ADD_FUNC (test_setting_compare_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_EXACT, FALSE);
ADD_FUNC (test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_AGENT_OWNED, NM_SETTING_COMPARE_FLAG_IGNORE_AGENT_OWNED_SECRETS, TRUE);
ADD_FUNC (test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_NOT_SAVED, NM_SETTING_COMPARE_FLAG_IGNORE_NOT_SAVED_SECRETS, TRUE);
ADD_FUNC (test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_IGNORE_SECRETS, TRUE);
ADD_FUNC (test_setting_compare_vpn_secrets, NM_SETTING_SECRET_FLAG_NONE, NM_SETTING_COMPARE_FLAG_EXACT, FALSE);
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);
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);
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);
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);
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