fdo-mirrors/NetworkManager
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/NetworkManager/NetworkManager.git synced 2025-12-20 14:00:05 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
NetworkManager/clients/cli/devices.c

3949 lines
129 KiB
C
Raw Normal View History

cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
/* nmcli - command-line tool to control NetworkManager
*
* 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 of the License, 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.
*
cli: strip errors from D-Bus before presenting to a user
2014-10-01 10:31:03 +02:00
* Copyright 2010 - 2014 Red Hat, Inc.
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
*/
all: cleanup includes and let "nm-default.h" include "config.h" - All internal source files (except "examples", which are not internal) should include "config.h" first. As also all internal source files should include "nm-default.h", let "config.h" be included by "nm-default.h" and include "nm-default.h" as first in every source file. We already wanted to include "nm-default.h" before other headers because it might contains some fixes (like "nm-glib.h" compatibility) that is required first. - After including "nm-default.h", we optinally allow for including the corresponding header file for the source file at hand. The idea is to ensure that each header file is self contained. - Don't include "config.h" or "nm-default.h" in any header file (except "nm-sd-adapt.h"). Public headers anyway must not include these headers, and internal headers are never included after "nm-default.h", as of the first previous point. - Include all internal headers with quotes instead of angle brackets. In practice it doesn't matter, because in our public headers we must include other headers with angle brackets. As we use our public headers also to compile our interal source files, effectively the result must be the same. Still do it for consistency. - Except for <config.h> itself. Include it with angle brackets as suggested by https://www.gnu.org/software/autoconf/manual/autoconf.html#Configuration-Headers
2016-02-19 14:57:48 +01:00
#include "nm-default.h"
build: do the right thing with config.h (bgo #644664) 1) it shouldn't be included in headers 2) it should be the first thing included in source files 3) it's needed for getting translation right
2011-03-14 01:00:56 -05:00
cli: move NmcMetaGenericInfo to "utils.h"
2017-04-04 13:52:13 +02:00
#include "devices.h"
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <errno.h>
cli: use readline also in other commands (besides interactive editor) It is used for '--ask' option. The main use is for questionnaires (nmcli -a con add).
2014-05-14 09:22:50 +02:00
#include <readline/readline.h>
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: add support for secret agent to 'nmcli dev connect' too And move secrets getting code to common.c (without changes).
2014-11-20 12:36:14 +01:00
#include "nm-secret-agent-simple.h"
cli: move utils function from common.h to nm-meta-setting-desc.c These functions are only used by nm-meta-setting-desc.c. Make them internal. Unfortunately, they are part of "common.h" which cannot be used without the rest of nmcli. Still todo.
2017-03-28 12:16:31 +02:00
#include "nm-client-utils.h"
cli: add 'nmcli agent' command (bgo #739568) Synopsis: nmcli agent { secret | polkit | all } The command runs separate NetworkManager secret agent or session polkit agent, or both. It is useful when - no other secret agent is available (such as GUI nm-applet, gnome-shell, KDE applet) - no other polkit agent is available (such as polkit-gnome-authentication-agent-1, polkit-kde-authentication-agent-1 or lxpolkit) https://bugzilla.gnome.org/show_bug.cgi?id=739568
2014-10-30 15:45:43 +01:00
#include "polkit-agent.h"
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
#include "utils.h"
cli: move IP config printing code from devices.c to common.c That allows sharing the code between devices and active connections.
2012-01-06 13:51:33 +01:00
#include "common.h"
cli/device: add modify command It modifies the applied connection using the Reapply API.
2016-06-22 18:25:48 +02:00
#include "connections.h"
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
/* define some prompts */
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
#define PROMPT_INTERFACE _("Interface: ")
#define PROMPT_INTERFACES _("Interface(s): ")
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_status[] = {
NMC_META_GENERIC ("DEVICE"), /* 0 */
NMC_META_GENERIC ("TYPE"), /* 1 */
NMC_META_GENERIC ("STATE"), /* 2 */
NMC_META_GENERIC ("DBUS-PATH"), /* 3 */
NMC_META_GENERIC ("CONNECTION"), /* 4 */
NMC_META_GENERIC ("CON-UUID"), /* 5 */
NMC_META_GENERIC ("CON-PATH"), /* 6 */
NULL,
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
};
cli: add active connections to 'nmcli device status'
2013-12-09 12:21:44 +01:00
#define NMC_FIELDS_DEV_STATUS_COMMON "DEVICE,TYPE,STATE,CONNECTION"
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_general[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("DEVICE"), /* 1 */
NMC_META_GENERIC ("TYPE"), /* 2 */
NMC_META_GENERIC ("NM-TYPE"), /* 3 */
NMC_META_GENERIC ("VENDOR"), /* 4 */
NMC_META_GENERIC ("PRODUCT"), /* 5 */
NMC_META_GENERIC ("DRIVER"), /* 6 */
NMC_META_GENERIC ("DRIVER-VERSION"), /* 7 */
NMC_META_GENERIC ("FIRMWARE-VERSION"), /* 8 */
NMC_META_GENERIC ("HWADDR"), /* 9 */
NMC_META_GENERIC ("MTU"), /* 10 */
NMC_META_GENERIC ("STATE"), /* 11 */
NMC_META_GENERIC ("REASON"), /* 12 */
NMC_META_GENERIC ("UDI"), /* 13 */
NMC_META_GENERIC ("IP-IFACE"), /* 14 */
NMC_META_GENERIC ("IS-SOFTWARE"), /* 15 */
NMC_META_GENERIC ("NM-MANAGED"), /* 16 */
NMC_META_GENERIC ("AUTOCONNECT"), /* 17 */
NMC_META_GENERIC ("FIRMWARE-MISSING"), /* 18 */
NMC_META_GENERIC ("NM-PLUGIN-MISSING"), /* 19 */
NMC_META_GENERIC ("PHYS-PORT-ID"), /* 20 */
NMC_META_GENERIC ("CONNECTION"), /* 21 */
NMC_META_GENERIC ("CON-UUID"), /* 22 */
NMC_META_GENERIC ("CON-PATH"), /* 23 */
NMC_META_GENERIC ("METERED"), /* 24 */
NULL,
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
};
cli: rename 'device list' -> 'device show' and remove 'iface' keyword Examples: nmcli device show - shows details for all available devices nmcli device show eth0 - shows details for eth0 nmcli device show em1.25 - shows details for em1.25 VLAN device
2013-04-03 10:01:42 +02:00
#define NMC_FIELDS_DEV_SHOW_GENERAL_COMMON "NAME,DEVICE,TYPE,VENDOR,PRODUCT,DRIVER,HWADDR,STATE"
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_connections[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("AVAILABLE-CONNECTION-PATHS"), /* 1 */
NMC_META_GENERIC ("AVAILABLE-CONNECTIONS"), /* 2 */
NULL,
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
};
nmcli: fix missing CONNECTIONS section name from nmcli -f all -m tab dev show <dev> When a full section is specified as the field parameter in terse tabular mode, the section name should be printed followed by all the field values belonging to that section separated by ':'. The NAME of section CONNECTIONS was missing. sample command: $ nmcli -m tab -t -f CONNECTIONS device show ens3 previous output was: /org/freedesktop/NetworkManager/Settings/{1}:5059XXX-XXXX.. | ens3-dhcp now: CONNECTIONS:/org/freedesktop/NetworkManager/Settings/{1}:5059XXX-XXXX.. | ens3-dhcp
2017-03-24 12:38:01 +01:00
#define NMC_FIELDS_DEV_SHOW_CONNECTIONS_COMMON "NAME,AVAILABLE-CONNECTION-PATHS,AVAILABLE-CONNECTIONS"
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_cap[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("CARRIER-DETECT"), /* 1 */
NMC_META_GENERIC ("SPEED"), /* 2 */
NMC_META_GENERIC ("IS-SOFTWARE"), /* 3 */
all: detect SR-IOV device support
2017-04-14 23:16:11 +02:00
NMC_META_GENERIC ("SRIOV"), /* 4 */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
NULL,
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
};
cli: show whether the device is software-based in CAPABILITIES
2014-07-11 10:33:11 +02:00
#define NMC_FIELDS_DEV_SHOW_CAP_COMMON "NAME,CARRIER-DETECT,SPEED,IS-SOFTWARE"
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_wired_prop[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("CARRIER"), /* 1 */
NMC_META_GENERIC ("S390-SUBCHANNELS"), /* 2 */
NULL,
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
};
cli: print s390 subchannels for devices that have that
2015-09-23 16:07:08 +02:00
#define NMC_FIELDS_DEV_SHOW_WIRED_PROP_COMMON "NAME,CARRIER,S390-SUBCHANNELS"
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_wifi_prop[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("WEP"), /* 1 */
NMC_META_GENERIC ("WPA"), /* 2 */
NMC_META_GENERIC ("WPA2"), /* 3 */
NMC_META_GENERIC ("TKIP"), /* 4 */
NMC_META_GENERIC ("CCMP"), /* 5 */
NMC_META_GENERIC ("AP"), /* 6 */
NMC_META_GENERIC ("ADHOC"), /* 7 */
NMC_META_GENERIC ("2GHZ"), /* 8 */
NMC_META_GENERIC ("5GHZ"), /* 9 */
NULL,
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
};
cli: rename 'device list' -> 'device show' and remove 'iface' keyword Examples: nmcli device show - shows details for all available devices nmcli device show eth0 - shows details for eth0 nmcli device show em1.25 - shows details for em1.25 VLAN device
2013-04-03 10:01:42 +02:00
#define NMC_FIELDS_DEV_SHOW_WIFI_PROP_COMMON "NAME,WEP,WPA,WPA2,TKIP,CCMP,AP,ADHOC"
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_wimax_prop[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("CTR-FREQ"), /* 1 */
NMC_META_GENERIC ("RSSI"), /* 2 */
NMC_META_GENERIC ("CINR"), /* 3 */
NMC_META_GENERIC ("TX-POW"), /* 4 */
NMC_META_GENERIC ("BSID"), /* 5 */
NULL,
cli: add support for WiMAX devices and connections
2011-01-06 17:01:55 -06:00
};
cli: rename 'device list' -> 'device show' and remove 'iface' keyword Examples: nmcli device show - shows details for all available devices nmcli device show eth0 - shows details for eth0 nmcli device show em1.25 - shows details for em1.25 VLAN device
2013-04-03 10:01:42 +02:00
#define NMC_FIELDS_DEV_SHOW_WIMAX_PROP_COMMON "NAME,CTR-FREQ,RSSI,CINR,TX-POW,BSID"
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_wifi_list[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("SSID"), /* 1 */
NMC_META_GENERIC ("SSID-HEX"), /* 2 */
NMC_META_GENERIC ("BSSID"), /* 3 */
NMC_META_GENERIC ("MODE"), /* 4 */
NMC_META_GENERIC ("CHAN"), /* 5 */
NMC_META_GENERIC ("FREQ"), /* 6 */
NMC_META_GENERIC ("RATE"), /* 7 */
NMC_META_GENERIC ("SIGNAL"), /* 8 */
NMC_META_GENERIC ("BARS"), /* 9 */
NMC_META_GENERIC ("SECURITY"), /* 10 */
NMC_META_GENERIC ("WPA-FLAGS"), /* 11 */
NMC_META_GENERIC ("RSN-FLAGS"), /* 12 */
NMC_META_GENERIC ("DEVICE"), /* 13 */
NMC_META_GENERIC ("ACTIVE"), /* 14 */
NMC_META_GENERIC ("IN-USE"), /* 15 */
NMC_META_GENERIC ("DBUS-PATH"), /* 16 */
NULL,
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
};
cli: 'dev wifi list': change default displayed columns Now they are #define NMC_FIELDS_DEV_WIFI_LIST_COMMON "IN-USE,SSID,MODE,CHAN,RATE,SIGNAL,BARS,SECURITY"
2013-05-20 15:50:47 +02:00
#define NMC_FIELDS_DEV_WIFI_LIST_COMMON "IN-USE,SSID,MODE,CHAN,RATE,SIGNAL,BARS,SECURITY"
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
#define NMC_FIELDS_DEV_WIFI_LIST_FOR_DEV_LIST "NAME,"NMC_FIELDS_DEV_WIFI_LIST_COMMON
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_wimax_list[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("NSP"), /* 1 */
NMC_META_GENERIC ("SIGNAL"), /* 2 */
NMC_META_GENERIC ("TYPE"), /* 3 */
NMC_META_GENERIC ("DEVICE"), /* 4 */
NMC_META_GENERIC ("ACTIVE"), /* 5 */
NMC_META_GENERIC ("DBUS-PATH"), /* 6 */
NULL,
cli: add support for WiMAX devices and connections
2011-01-06 17:01:55 -06:00
};
#define NMC_FIELDS_DEV_WIMAX_LIST_COMMON "NSP,SIGNAL,TYPE,DEVICE,ACTIVE"
#define NMC_FIELDS_DEV_WIMAX_LIST_FOR_DEV_LIST "NAME,"NMC_FIELDS_DEV_WIMAX_LIST_COMMON
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_master_prop[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("SLAVES"), /* 1 */
NULL,
cli: add BOND section to 'nmcli dev list' for bond devices Now, it just contains one item - SLAVES listing all slave devices of the bond device.
2012-10-22 14:31:11 +02:00
};
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
#define NMC_FIELDS_DEV_SHOW_MASTER_PROP_COMMON "NAME,SLAVES"
cli: add BOND section to 'nmcli dev list' for bond devices Now, it just contains one item - SLAVES listing all slave devices of the bond device.
2012-10-22 14:31:11 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_team_prop[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("SLAVES"), /* 1 */
NMC_META_GENERIC ("CONFIG"), /* 2 */
NULL,
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
};
#define NMC_FIELDS_DEV_SHOW_TEAM_PROP_COMMON "NAME,SLAVES,CONFIG"
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_vlan_prop[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("PARENT"), /* 1 */
NMC_META_GENERIC ("ID"), /* 2 */
NULL,
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
};
cli: add parent device information for VLAN devices
2014-07-21 13:08:53 +02:00
#define NMC_FIELDS_DEV_SHOW_VLAN_PROP_COMMON "NAME,PARENT,ID"
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_bluetooth[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("CAPABILITIES"), /* 1 */
NULL,
cli: show capabilities for Bluetooth devices
2014-09-24 12:00:47 +02:00
};
#define NMC_FIELDS_DEV_SHOW_BLUETOOTH_COMMON "NAME,CAPABILITIES"
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
/* Available sections for 'device show' */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_show_sections[] = {
NMC_META_GENERIC_WITH_NESTED ("GENERAL", nmc_fields_dev_show_general + 1), /* 0 */
NMC_META_GENERIC_WITH_NESTED ("CAPABILITIES", nmc_fields_dev_show_cap + 1), /* 1 */
NMC_META_GENERIC_WITH_NESTED ("WIFI-PROPERTIES", nmc_fields_dev_show_wifi_prop + 1), /* 2 */
NMC_META_GENERIC_WITH_NESTED ("AP", nmc_fields_dev_wifi_list + 1), /* 3 */
NMC_META_GENERIC_WITH_NESTED ("WIRED-PROPERTIES", nmc_fields_dev_show_wired_prop + 1), /* 4 */
NMC_META_GENERIC_WITH_NESTED ("WIMAX-PROPERTIES", nmc_fields_dev_show_wimax_prop + 1), /* 5 */
NMC_META_GENERIC_WITH_NESTED ("NSP", nmc_fields_dev_wimax_list + 1), /* 6 */
cli: use nmc_print() to output device's IP4 info The IP4 info adds a new type: to expose strv arguments for addresses, etc.
2017-04-06 15:14:23 +02:00
NMC_META_GENERIC_WITH_NESTED ("IP4", metagen_ip4_config), /* 7 */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
NMC_META_GENERIC_WITH_NESTED ("DHCP4", nmc_fields_dhcp4_config + 1), /* 8 */
NMC_META_GENERIC_WITH_NESTED ("IP6", nmc_fields_ip6_config + 1), /* 9 */
NMC_META_GENERIC_WITH_NESTED ("DHCP6", nmc_fields_dhcp6_config + 1), /* 10 */
NMC_META_GENERIC_WITH_NESTED ("BOND", nmc_fields_dev_show_master_prop + 1), /* 11 */
NMC_META_GENERIC_WITH_NESTED ("TEAM", nmc_fields_dev_show_team_prop + 1), /* 12 */
NMC_META_GENERIC_WITH_NESTED ("BRIDGE", nmc_fields_dev_show_master_prop + 1), /* 13 */
NMC_META_GENERIC_WITH_NESTED ("VLAN", nmc_fields_dev_show_vlan_prop + 1), /* 14 */
NMC_META_GENERIC_WITH_NESTED ("BLUETOOTH", nmc_fields_dev_show_bluetooth + 1), /* 15 */
NMC_META_GENERIC_WITH_NESTED ("CONNECTIONS", nmc_fields_dev_show_connections + 1), /* 16 */
NULL,
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
};
cli: display MTU for devices
2013-12-16 17:08:28 +01:00
#define NMC_FIELDS_DEV_SHOW_SECTIONS_COMMON "GENERAL.DEVICE,GENERAL.TYPE,GENERAL.HWADDR,GENERAL.MTU,GENERAL.STATE,"\
"GENERAL.CONNECTION,GENERAL.CON-PATH,WIRED-PROPERTIES,IP4,IP6"
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NmcMetaGenericInfo *const nmc_fields_dev_lldp_list[] = {
NMC_META_GENERIC ("NAME"), /* 0 */
NMC_META_GENERIC ("DEVICE"), /* 1 */
NMC_META_GENERIC ("CHASSIS-ID"), /* 2 */
NMC_META_GENERIC ("PORT-ID"), /* 3 */
NMC_META_GENERIC ("PORT-DESCRIPTION"), /* 4 */
NMC_META_GENERIC ("SYSTEM-NAME"), /* 5 */
NMC_META_GENERIC ("SYSTEM-DESCRIPTION"), /* 6 */
NMC_META_GENERIC ("SYSTEM-CAPABILITIES"), /* 7 */
NMC_META_GENERIC ("IEEE-802-1-PVID"), /* 8 */
NMC_META_GENERIC ("IEEE-802-1-PPVID"), /* 9 */
NMC_META_GENERIC ("IEEE-802-1-PPVID-FLAGS"), /* 10 */
NMC_META_GENERIC ("IEEE-802-1-VID"), /* 11 */
NMC_META_GENERIC ("IEEE-802-1-VLAN-NAME"), /* 12 */
NMC_META_GENERIC ("DESTINATION"), /* 13 */
NMC_META_GENERIC ("CHASSIS-ID-TYPE"), /* 14 */
NMC_META_GENERIC ("PORT-ID-TYPE"), /* 15 */
NULL,
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
};
#define NMC_FIELDS_DEV_LLDP_LIST_COMMON "DEVICE,CHASSIS-ID,PORT-ID,PORT-DESCRIPTION,SYSTEM-NAME,SYSTEM-DESCRIPTION," \
"SYSTEM-CAPABILITIES"
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
static guint progress_id = 0; /* ID of event source for displaying progress */
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static void
usage (void)
{
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Usage: nmcli device { COMMAND | help }\n\n"
nmcli: fix 'nmcli device modify' description Fixes: 8b4494598d04b6754a69463a4dc7274fdff8763f
2016-07-14 22:06:10 +02:00
"COMMAND := { status | show | set | connect | reapply | modify | disconnect | delete | monitor | wifi | lldp }\n\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
" status\n\n"
" show [<ifname>]\n\n"
nmcli: extend `nmcli device set` command to accept explicit ifname specifier Without it, you cannot set the properties of a device named "help". Now you can with: $ nmcli device set ifname help autoconnect no
2015-09-18 10:49:19 +02:00
" set [ifname] <ifname> [autoconnect yes|no] [managed yes|no]\n\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
" connect <ifname>\n\n"
cli: don't allow multiple <ifname> arguments to device-reapply Just like `nmcli device connect` only allows one argument, don't allow multiple device arguments for reapply. Allowing multiple device names makes it more complicated to add additional options to the command. For example, it would be useful to have a nmcli device reapply eth0 connection id other-connection but when allowing multiple device names, it gets more complicated in documentation, command line parsing and bash completion. Note that the user can achieve a very similar outcome by using the shell: for DEV in eth0 eth1 eth2; do nmcli device reapply $DEV & done wait argubaly, this doesn't report the exit status properly. To properly handle that would require more effort. Also, it is somewhat less efficient, but well. This is an API change, however it is very new API that probably nobody is using much. Also, the documentation (man nmcli) didn't mention the possibility to pass multiple device names.
2016-04-30 14:14:31 +02:00
" reapply <ifname>\n\n"
nmcli: fix 'nmcli device modify' description Fixes: 8b4494598d04b6754a69463a4dc7274fdff8763f
2016-07-14 22:06:10 +02:00
" modify <ifname> ([+|-]<setting>.<property> <value>)+\n\n"
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
" disconnect <ifname> ...\n\n"
" delete <ifname> ...\n\n"
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
" monitor <ifname> ...\n\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
" wifi [list [ifname <ifname>] [bssid <BSSID>]]\n\n"
" wifi connect <(B)SSID> [password <password>] [wep-key-type key|phrase] [ifname <ifname>]\n"
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
" [bssid <BSSID>] [name <name>] [private yes|no] [hidden yes|no]\n\n"
cli: add global --show-secrets option - it controls echoing passwords input on terminal - it replaces --show-secrets in 'nmcli connection show', which is deprecated now - it replaces --show-password in 'nmcli device wifi hotspot', which is deprecated now
2015-12-04 13:29:42 +01:00
" wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>] [password <password>]\n\n"
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
" wifi rescan [ifname <ifname>] [[ssid <SSID to scan>] ...]\n\n"
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
" lldp [list [ifname <ifname>]]\n\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
));
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
static void
usage_device_status (void)
{
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Usage: nmcli device status { help }\n"
"\n"
"Show status for all devices.\n"
"By default, the following columns are shown:\n"
" DEVICE - interface name\n"
" TYPE - device type\n"
" STATE - device state\n"
" CONNECTION - connection activated on device (if any)\n"
"Displayed columns can be changed using '--fields' global option. 'status' is\n"
"the default command, which means 'nmcli device' calls 'nmcli device status'.\n\n"));
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
}
static void
usage_device_show (void)
{
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Usage: nmcli device show { ARGUMENTS | help }\n"
"\n"
"ARGUMENTS := [<ifname>]\n"
"\n"
"Show details of device(s).\n"
"The command lists details for all devices, or for a given device.\n\n"));
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
}
static void
usage_device_connect (void)
{
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Usage: nmcli device connect { ARGUMENTS | help }\n"
"\n"
"ARGUMENTS := <ifname>\n"
"\n"
"Connect the device.\n"
"NetworkManager will try to find a suitable connection that will be activated.\n"
"It will also consider connections that are not set to auto-connect.\n\n"));
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
}
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
static void
usage_device_reapply (void)
{
g_printerr (_("Usage: nmcli device reapply { ARGUMENTS | help }\n"
"\n"
cli: don't allow multiple <ifname> arguments to device-reapply Just like `nmcli device connect` only allows one argument, don't allow multiple device arguments for reapply. Allowing multiple device names makes it more complicated to add additional options to the command. For example, it would be useful to have a nmcli device reapply eth0 connection id other-connection but when allowing multiple device names, it gets more complicated in documentation, command line parsing and bash completion. Note that the user can achieve a very similar outcome by using the shell: for DEV in eth0 eth1 eth2; do nmcli device reapply $DEV & done wait argubaly, this doesn't report the exit status properly. To properly handle that would require more effort. Also, it is somewhat less efficient, but well. This is an API change, however it is very new API that probably nobody is using much. Also, the documentation (man nmcli) didn't mention the possibility to pass multiple device names.
2016-04-30 14:14:31 +02:00
"ARGUMENTS := <ifname>\n"
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
"\n"
"Attempts to update device with changes to the currently active connection\n"
"made since it was last applied.\n\n"));
}
cli/device: add modify command It modifies the applied connection using the Reapply API.
2016-06-22 18:25:48 +02:00
static void
usage_device_modify (void)
{
nmcli: fix 'nmcli device modify' description Fixes: 8b4494598d04b6754a69463a4dc7274fdff8763f
2016-07-14 22:06:10 +02:00
g_printerr (_("Usage: nmcli device modify { ARGUMENTS | --help }\n"
cli/device: add modify command It modifies the applied connection using the Reapply API.
2016-06-22 18:25:48 +02:00
"\n"
"ARGUMENTS := <ifname> ([+|-]<setting>.<property> <value>)+\n"
"\n"
"Modify one or more properties currently active on the device without modifying\n"
"the connection profile. The changes have immediate effect. For multi-valued\n"
"properties you can use optional '+' or '-' prefix to the property name.\n"
"The '+' sign allows appending items instead of overwriting the whole value.\n"
"The '-' sign allows removing selected items instead of the whole value.\n"
"\n"
"Examples:\n"
"nmcli dev mod em1 ipv4.method manual ipv4.addr \"192.168.1.2/24, 10.10.1.5/8\"\n"
"nmcli dev mod em1 +ipv4.dns 8.8.4.4\n"
"nmcli dev mod em1 -ipv4.dns 1\n"
"nmcli dev mod em1 -ipv6.addr \"abbe::cafe/56\"\n"));
}
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
static void
usage_device_disconnect (void)
{
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Usage: nmcli device disconnect { ARGUMENTS | help }\n"
"\n"
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
"ARGUMENTS := <ifname> ...\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
"\n"
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
"Disconnect devices.\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
"The command disconnects the device and prevents it from auto-activating\n"
"further connections without user/manual intervention.\n\n"));
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
}
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
static void
usage_device_delete (void)
{
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Usage: nmcli device delete { ARGUMENTS | help }\n"
"\n"
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
"ARGUMENTS := <ifname> ...\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
"\n"
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
"Delete the software devices.\n"
"The command removes the interfaces. It only works for software devices\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
"(like bonds, bridges, etc.). Hardware devices cannot be deleted by the\n"
"command.\n\n"));
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
}
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
static void
usage_device_set (void)
{
g_printerr (_("Usage: nmcli device set { ARGUMENTS | help }\n"
"\n"
nmcli: extend `nmcli device set` command to accept explicit ifname specifier Without it, you cannot set the properties of a device named "help". Now you can with: $ nmcli device set ifname help autoconnect no
2015-09-18 10:49:19 +02:00
"ARGUMENTS := DEVICE { PROPERTY [ PROPERTY ... ] }\n"
"DEVICE := [ifname] <ifname> \n"
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
"PROPERTY := { autoconnect { yes | no } |\n"
" { managed { yes | no }\n"
"\n"
"Modify device properties.\n\n"));
}
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
static void
usage_device_monitor (void)
{
g_printerr (_("Usage: nmcli device monitor { ARGUMENTS | help }\n"
"\n"
"ARGUMENTS := [<ifname>] ...\n"
"\n"
"Monitor device activity.\n"
"This command prints a line whenever the specified devices change state.\n"
"Monitors all devices in case no interface is specified.\n\n"));
}
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
static void
usage_device_wifi (void)
{
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Usage: nmcli device wifi { ARGUMENTS | help }\n"
"\n"
"Perform operation on Wi-Fi devices.\n"
"\n"
"ARGUMENTS := [list [ifname <ifname>] [bssid <BSSID>]]\n"
"\n"
"List available Wi-Fi access points. The 'ifname' and 'bssid' options can be\n"
"used to list APs for a particular interface, or with a specific BSSID.\n"
"\n"
"ARGUMENTS := connect <(B)SSID> [password <password>] [wep-key-type key|phrase] [ifname <ifname>]\n"
cli: fix synopsis for hotspot arguments for `nmcli device wifi --help`
2016-10-12 15:06:04 +02:00
" [bssid <BSSID>] [name <name>] [private yes|no] [hidden yes|no]\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
"\n"
"Connect to a Wi-Fi network specified by SSID or BSSID. The command creates\n"
"a new connection and then activates it on a device. This is a command-line\n"
"counterpart of clicking an SSID in a GUI client. The command always creates\n"
"a new connection and thus it is mainly useful for connecting to new Wi-Fi\n"
"networks. If a connection for the network already exists, it is better to\n"
"bring up the existing profile as follows: nmcli con up id <name>. Note that\n"
"only open, WEP and WPA-PSK networks are supported at the moment. It is also\n"
"assumed that IP configuration is obtained via DHCP.\n"
"\n"
cli: fix synopsis for hotspot arguments for `nmcli device wifi --help`
2016-10-12 15:06:04 +02:00
"ARGUMENTS := hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>]\n"
" [band a|bg] [channel <channel>] [password <password>]\n"
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
"\n"
"Create a Wi-Fi hotspot. Use 'connection down' or 'device disconnect'\n"
"to stop the hotspot.\n"
"Parameters of the hotspot can be influenced by the optional parameters:\n"
"ifname - Wi-Fi device to use\n"
"con-name - name of the created hotspot connection profile\n"
"ssid - SSID of the hotspot\n"
"band - Wi-Fi band to use\n"
"channel - Wi-Fi channel to use\n"
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
"password - password to use for the hotspot\n"
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
"\n"
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
"ARGUMENTS := rescan [ifname <ifname>] [[ssid <SSID to scan>] ...]\n"
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
"\n"
"Request that NetworkManager immediately re-scan for available access points.\n"
"NetworkManager scans Wi-Fi networks periodically, but in some cases it might\n"
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
"be useful to start scanning manually. 'ssid' allows scanning for a specific\n"
"SSID, which is useful for APs with hidden SSIDs. More 'ssid' parameters can be\n"
"given. Note that this command does not show the APs,\n"
"use 'nmcli device wifi list' for that.\n\n"));
cli: add 'help' for second level commands (bgo #698237) (rh #1034119) In addition to 'nmcli connection help', you can now use: nmcli con up help nmcli con down help nmcli dev wifi help etc. The help option can also be in the form of --help/-h, or even shortened. https://bugzilla.gnome.org/show_bug.cgi?id=698237 https://bugzilla.redhat.com/show_bug.cgi?id=1034119
2013-12-18 14:49:06 +01:00
}
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
static void
usage_device_lldp (void)
{
g_printerr (_("Usage: nmcli device lldp { ARGUMENTS | help }\n"
"\n"
"ARGUMENTS := [list [ifname <ifname>]]\n"
"\n"
"List neighboring devices discovered through LLDP. The 'ifname' option can be\n"
"used to list neighbors for a particular interface.\n\n"));
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static void
quit (void)
{
cli: use nm_clear_g_source() util in quit() functions
2017-11-29 10:19:12 +01:00
if (nm_clear_g_source (&progress_id))
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
nmc_terminal_erase_line ();
cli: move global "loop" variable to nmcli.h header
2017-04-06 19:39:36 +02:00
g_main_loop_quit (loop);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
static int
compare_devices (const void *a, const void *b)
{
NMDevice *da = *(NMDevice **)a;
NMDevice *db = *(NMDevice **)b;
cli: prioritize devices with default routes and addresses This makes "nmcli" and "nmcli d" output slightly more useful.
2017-06-26 13:22:59 +02:00
NMActiveConnection *da_ac;
NMActiveConnection *db_ac;
NMIPConfig *da_ip;
NMIPConfig *db_ip;
int da_num_addrs;
int db_num_addrs;
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
int cmp;
/* Sort by later device states first */
cmp = nm_device_get_state (db) - nm_device_get_state (da);
if (cmp != 0)
return cmp;
cli: prioritize devices with default routes and addresses This makes "nmcli" and "nmcli d" output slightly more useful.
2017-06-26 13:22:59 +02:00
da_ac = nm_device_get_active_connection (da);
db_ac = nm_device_get_active_connection (db);
/* Prioritize devices with active connections */
if (da_ac)
cmp++;
if (db_ac)
cmp--;
if (cmp != 0)
return cmp;
/* VPNs go on the top if possible */
if (da_ac && !nm_active_connection_get_vpn (da_ac))
cmp++;
if (db_ac && !nm_active_connection_get_vpn (db_ac))
cmp--;
if (cmp != 0)
return cmp;
/* Default devices are prioritized */
if (da_ac && !nm_active_connection_get_default (da_ac))
cmp++;
if (db_ac && !nm_active_connection_get_default (db_ac))
cmp--;
if (cmp != 0)
return cmp;
/* Default IPv6 devices are prioritized */
if (da_ac && !nm_active_connection_get_default6 (da_ac))
cmp++;
if (db_ac && !nm_active_connection_get_default6 (db_ac))
cmp--;
if (cmp != 0)
return cmp;
/* Sort by number of addresses. */
da_ip = da_ac ? nm_active_connection_get_ip4_config (da_ac) : NULL;
da_num_addrs = da_ip ? nm_ip_config_get_addresses (da_ip)->len : 0;
db_ip = db_ac ? nm_active_connection_get_ip4_config (db_ac) : NULL;
db_num_addrs = db_ip ? nm_ip_config_get_addresses (db_ip)->len : 0;
da_ip = da_ac ? nm_active_connection_get_ip6_config (da_ac) : NULL;
da_num_addrs += da_ip ? nm_ip_config_get_addresses (da_ip)->len : 0;
db_ip = db_ac ? nm_active_connection_get_ip6_config (db_ac) : NULL;
db_num_addrs += db_ip ? nm_ip_config_get_addresses (db_ip)->len : 0;
cmp = db_num_addrs - da_num_addrs;
if (cmp != 0)
return cmp;
/* Fall back to alphanumeric sort by description and interface. */
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
cmp = g_strcmp0 (nm_device_get_type_description (da),
nm_device_get_type_description (db));
if (cmp != 0)
return cmp;
return g_strcmp0 (nm_device_get_iface (da),
nm_device_get_iface (db));
}
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
NMDevice **
nmc_get_devices_sorted (NMClient *client)
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
{
const GPtrArray *devs;
NMDevice **sorted;
devs = nm_client_get_devices (client);
sorted = g_new (NMDevice *, devs->len + 1);
memcpy (sorted, devs->pdata, devs->len * sizeof (NMDevice *));
sorted[devs->len] = NULL;
qsort (sorted, devs->len, sizeof (NMDevice *), compare_devices);
return sorted;
}
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
static void
cli/device: make it possible to limit completion to wifis
2016-07-12 10:15:01 +02:00
complete_device (NMDevice **devices, const char *prefix, gboolean wifi_only)
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
{
int i;
for (i = 0; devices[i]; i++) {
const char *iface = nm_device_get_iface (devices[i]);
cli/device: make it possible to limit completion to wifis
2016-07-12 10:15:01 +02:00
if (wifi_only && !NM_IS_DEVICE_WIFI (devices[i]))
continue;
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (g_str_has_prefix (iface, prefix))
g_print ("%s\n", iface);
}
}
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
void
nmc_complete_device (NMClient *client, const char *prefix, gboolean wifi_only)
{
gs_free NMDevice **devices = NULL;
devices = nmc_get_devices_sorted (client);
complete_device (devices, prefix, wifi_only);
}
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
static GSList *
cli/trivial: rename device_list() to get_device_list() It will look nicer when we have get_device().
2016-06-22 18:33:54 +02:00
get_device_list (NmCli *nmc, int argc, char **argv)
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
{
int arg_num = argc;
char **arg_arr = NULL;
char **arg_ptr = argv;
NMDevice **devices;
GSList *queue = NULL;
NMDevice *device;
int i;
if (argc == 0) {
if (nmc->ask) {
char *line = nmc_readline (PROMPT_INTERFACES);
nmc_string_to_arg_array (line, NULL, FALSE, &arg_arr, &arg_num);
g_free (line);
arg_ptr = arg_arr;
}
if (arg_num == 0) {
g_string_printf (nmc->return_text, _("Error: No interface specified."));
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
goto error;
}
}
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
devices = nmc_get_devices_sorted (nmc->client);
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
while (arg_num > 0) {
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (arg_num == 1 && nmc->complete)
cli/device: make it possible to limit completion to wifis
2016-07-12 10:15:01 +02:00
complete_device (devices, *arg_ptr, FALSE);
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
device = NULL;
for (i = 0; devices[i]; i++) {
if (!g_strcmp0 (nm_device_get_iface (devices[i]), *arg_ptr)) {
device = devices[i];
break;
}
}
if (device) {
if (!g_slist_find (queue, device))
queue = g_slist_prepend (queue, device);
else
g_printerr (_("Warning: argument '%s' is duplicated.\n"), *arg_ptr);
} else {
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (!nmc->complete)
g_printerr (_("Error: Device '%s' not found.\n"), *arg_ptr);
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
g_string_printf (nmc->return_text, _("Error: not all devices found."));
nmc->return_value = NMC_RESULT_ERROR_NOT_FOUND;
}
/* Take next argument */
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc->ask ? NULL : nmc, &arg_num, &arg_ptr, NULL);
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
}
g_free (devices);
error:
g_strfreev (arg_arr);
return queue;
}
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
static NMDevice *
get_device (NmCli *nmc, int *argc, char ***argv, GError **error)
{
gs_free NMDevice **devices = NULL;
gs_free char *ifname_ask = NULL;
const char *ifname = NULL;
int i;
if (*argc == 0) {
if (nmc->ask)
ifname = ifname_ask = nmc_readline (PROMPT_INTERFACE);
if (!ifname_ask) {
g_set_error_literal (error, NMCLI_ERROR, NMC_RESULT_ERROR_USER_INPUT,
nmcli: remove the extra full stop from error messages $ nmcli dev modify Error: No interface specified.. $ nmcli dev modify bla Error: Device 'bla' not found..
2016-07-15 11:38:01 +02:00
_("No interface specified"));
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
return NULL;
}
} else {
ifname = **argv;
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, argc, argv, NULL);
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
}
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
devices = nmc_get_devices_sorted (nmc->client);
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
for (i = 0; devices[i]; i++) {
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (!g_strcmp0 (nm_device_get_iface (devices[i]), ifname))
break;
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
}
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (nmc->complete && !*argc)
cli/device: make it possible to limit completion to wifis
2016-07-12 10:15:01 +02:00
complete_device (devices, ifname, FALSE);
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (devices[i] == NULL) {
g_set_error (error, NMCLI_ERROR, NMC_RESULT_ERROR_NOT_FOUND,
nmcli: remove the extra full stop from error messages $ nmcli dev modify Error: No interface specified.. $ nmcli dev modify bla Error: Device 'bla' not found..
2016-07-15 11:38:01 +02:00
_("Device '%s' not found"), ifname);
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
}
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
return devices[i];
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
}
cli: sort access point list According to signal strength, frequency and rate.
2015-02-26 11:51:07 +01:00
static int
compare_aps (gconstpointer a, gconstpointer b, gpointer user_data)
{
NMAccessPoint *apa = *(NMAccessPoint **)a;
NMAccessPoint *apb = *(NMAccessPoint **)b;
int cmp;
cmp = nm_access_point_get_strength (apb) - nm_access_point_get_strength (apa);
if (cmp != 0)
return cmp;
cmp = nm_access_point_get_frequency (apa) - nm_access_point_get_frequency (apb);
if (cmp != 0)
return cmp;
return nm_access_point_get_max_bitrate (apb) - nm_access_point_get_max_bitrate (apa);
}
static GPtrArray *
sort_access_points (const GPtrArray *aps)
{
GPtrArray *sorted;
int i;
cli: sort_access_points() doesn't take a NULL argument The device never returns NULL access point list, it only deallocates it on dispose(). Make this clear for Coverity: CID 59387 (#1 of 1): Dereference before null check (REVERSE_INULL) check_after_deref: Null-checking aps suggests that it may be null, but it has already been dereferenced on all paths leading to the check.
2016-02-04 16:42:00 +01:00
g_return_val_if_fail (aps, NULL);
cli: sort access point list According to signal strength, frequency and rate.
2015-02-26 11:51:07 +01:00
sorted = g_ptr_array_sized_new (aps->len);
cli: sort_access_points() doesn't take a NULL argument The device never returns NULL access point list, it only deallocates it on dispose(). Make this clear for Coverity: CID 59387 (#1 of 1): Dereference before null check (REVERSE_INULL) check_after_deref: Null-checking aps suggests that it may be null, but it has already been dereferenced on all paths leading to the check.
2016-02-04 16:42:00 +01:00
for (i = 0; i < aps->len; i++)
cli: sort access point list According to signal strength, frequency and rate.
2015-02-26 11:51:07 +01:00
g_ptr_array_add (sorted, aps->pdata[i]);
g_ptr_array_sort_with_data (sorted, compare_aps, NULL);
return sorted;
}
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
static void
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
wifi_signal_to_color (guint8 strength, NMMetaTermColor *color, NMMetaTermFormat *color_fmt)
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
{
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_NORMAL;
*color_fmt = NM_META_TERM_FORMAT_NORMAL;
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
if (strength > 80)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_GREEN;
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
else if (strength > 55)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_YELLOW;
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
else if (strength > 30)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_MAGENTA;
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
else if (strength > 5)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_CYAN;
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
else
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color_fmt = NM_META_TERM_FORMAT_DIM;
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static char *
cli: libnm-glib now uses enums for NMAccessPoint flags accessors
2011-02-21 15:41:12 +01:00
ap_wpa_rsn_flags_to_string (NM80211ApSecurityFlags flags)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
char *flags_str[16]; /* Enough space for flags and terminating NULL */
char *ret_str;
int i = 0;
if (flags & NM_802_11_AP_SEC_PAIR_WEP40)
flags_str[i++] = g_strdup ("pair_wpe40");
if (flags & NM_802_11_AP_SEC_PAIR_WEP104)
flags_str[i++] = g_strdup ("pair_wpe104");
if (flags & NM_802_11_AP_SEC_PAIR_TKIP)
flags_str[i++] = g_strdup ("pair_tkip");
if (flags & NM_802_11_AP_SEC_PAIR_CCMP)
flags_str[i++] = g_strdup ("pair_ccmp");
if (flags & NM_802_11_AP_SEC_GROUP_WEP40)
flags_str[i++] = g_strdup ("group_wpe40");
if (flags & NM_802_11_AP_SEC_GROUP_WEP104)
flags_str[i++] = g_strdup ("group_wpe104");
if (flags & NM_802_11_AP_SEC_GROUP_TKIP)
flags_str[i++] = g_strdup ("group_tkip");
if (flags & NM_802_11_AP_SEC_GROUP_CCMP)
flags_str[i++] = g_strdup ("group_ccmp");
if (flags & NM_802_11_AP_SEC_KEY_MGMT_PSK)
flags_str[i++] = g_strdup ("psk");
if (flags & NM_802_11_AP_SEC_KEY_MGMT_802_1X)
flags_str[i++] = g_strdup ("802.1X");
if (i == 0)
flags_str[i++] = g_strdup (_("(none)"));
flags_str[i] = NULL;
ret_str = g_strjoinv (" ", flags_str);
i = 0;
while (flags_str[i])
cli: trivial, fix up spacing
2012-04-28 22:54:02 +02:00
g_free (flags_str[i++]);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
return ret_str;
}
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
typedef struct {
NmCli *nmc;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
int index;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
guint32 output_flags;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
const char* active_bssid;
const char* device;
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
GPtrArray *output_data;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
} APInfo;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static void
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
fill_output_access_point (gpointer data, gpointer user_data)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
NMAccessPoint *ap = NM_ACCESS_POINT (data);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
APInfo *info = (APInfo *) user_data;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
NmcOutputField *arr;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
gboolean active = FALSE;
cli: libnm-glib now uses enums for NMAccessPoint flags accessors
2011-02-21 15:41:12 +01:00
NM80211ApFlags flags;
NM80211ApSecurityFlags wpa_flags, rsn_flags;
guint32 freq, bitrate;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
guint8 strength;
libnm: drop NM_TYPE_SSID, use G_TYPE_BYTES Make NMAccessPoint:ssid be G_TYPE_BYTES and update the corresponding APIs accordingly.
2014-07-02 14:25:43 -04:00
GBytes *ssid;
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
const char *bssid;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
NM80211Mode mode;
cli: handle hidden-SSID access points correctly
2014-01-08 11:39:22 -06:00
char *channel_str, *freq_str, *ssid_str = NULL, *ssid_hex_str = NULL,
*bitrate_str, *strength_str, *wpa_flags_str, *rsn_flags_str;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
GString *security_str;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
char *ap_name;
cli: 'dev wifi list': add BARS column displaying signal semi-graphically using unicode characters: '_' '▂' '▄' '▆' '█' 0x5f 0x2582 0x2584 0x2586 0x2588
2013-05-17 11:23:21 +02:00
const char *sig_bars;
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
NMMetaTermColor color;
NMMetaTermFormat color_fmt;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (info->active_bssid) {
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
const char *current_bssid = nm_access_point_get_bssid (ap);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (current_bssid && !strcmp (current_bssid, info->active_bssid))
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
active = TRUE;
}
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
/* Get AP properties */
flags = nm_access_point_get_flags (ap);
wpa_flags = nm_access_point_get_wpa_flags (ap);
rsn_flags = nm_access_point_get_rsn_flags (ap);
ssid = nm_access_point_get_ssid (ap);
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
bssid = nm_access_point_get_bssid (ap);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
freq = nm_access_point_get_frequency (ap);
mode = nm_access_point_get_mode (ap);
bitrate = nm_access_point_get_max_bitrate (ap);
coverity: fix various warnings detected with Coverity These are (most likely) only warnings and not severe bugs. Some of these changes are mostly made to get a clean run of Coverity without any warnings. Error found by running Coverity scan https://bugzilla.redhat.com/show_bug.cgi?id=1025894 Co-Authored-By: Jiří Klimeš <jklimes@redhat.com> Signed-off-by: Thomas Haller <thaller@redhat.com>
2013-10-31 14:13:33 +01:00
strength = MIN (nm_access_point_get_strength (ap), 100);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
/* Convert to strings */
cli: handle hidden-SSID access points correctly
2014-01-08 11:39:22 -06:00
if (ssid) {
libnm: drop NM_TYPE_SSID, use G_TYPE_BYTES Make NMAccessPoint:ssid be G_TYPE_BYTES and update the corresponding APIs accordingly.
2014-07-02 14:25:43 -04:00
const guint8 *ssid_data;
gsize ssid_len;
ssid_data = g_bytes_get_data (ssid, &ssid_len);
ssid_str = nm_utils_ssid_to_utf8 (ssid_data, ssid_len);
ssid_hex_str = ssid_to_hex ((const char *) ssid_data, ssid_len);
cli: handle hidden-SSID access points correctly
2014-01-08 11:39:22 -06:00
}
cli: 'dev wifi list': add CHAN column
2013-05-17 10:38:09 +02:00
channel_str = g_strdup_printf ("%u", nm_utils_wifi_freq_to_channel (freq));
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
freq_str = g_strdup_printf (_("%u MHz"), freq);
nmcli: use correct Mbit/s units for maximal bitrate (rh #1080474) https://bugzilla.redhat.com/show_bug.cgi?id=1080474
2014-04-01 14:56:32 +02:00
bitrate_str = g_strdup_printf (_("%u Mbit/s"), bitrate/1000);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
strength_str = g_strdup_printf ("%u", strength);
wpa_flags_str = ap_wpa_rsn_flags_to_string (wpa_flags);
rsn_flags_str = ap_wpa_rsn_flags_to_string (rsn_flags);
libnm-core: add nm_utils_wifi_strength_bars(), use it in nmcli and nmtui Add nm_utils_wifi_strength_bars(), which figures out whether the terminal can display graphical wifi strength bars, and converts a numerical value to the appropriate Unicode or ASCII characters. This also now takes into consideration the fact that the console font doesn't contain all of the necessary characters, so we can't display the graphical bars there. (rh #1131491)
2014-09-19 16:51:07 -04:00
sig_bars = nm_utils_wifi_strength_bars (strength);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
security_str = g_string_new (NULL);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
if ( (flags & NM_802_11_AP_FLAGS_PRIVACY)
&& (wpa_flags == NM_802_11_AP_SEC_NONE)
cli: 'dev wifi list': change SECURITY values * remove "Encrypted" tag The "Encrypted: " stuff was initially copied from nm-tool, but it doesn't help here much. See also http://cgit.freedesktop.org/NetworkManager/NetworkManager/commit/?id=a734c836a56f3170202f0555f1a03c9b2835775c for APs with WPA & RSN IEs, but no privacy flag. * remove a space from "WPA ", etc. strings. Translators often leave it out and thus break output. * change "WPA" to "WPA1" to make it clearer * use "802.1X" instead of "Enterprise" to save some characters
2013-05-24 22:36:24 +02:00
&& (rsn_flags == NM_802_11_AP_SEC_NONE)) {
g_string_append (security_str, _("WEP"));
g_string_append_c (security_str, ' ');
}
if (wpa_flags != NM_802_11_AP_SEC_NONE) {
g_string_append (security_str, _("WPA1"));
g_string_append_c (security_str, ' ');
}
if (rsn_flags != NM_802_11_AP_SEC_NONE) {
g_string_append (security_str, _("WPA2"));
g_string_append_c (security_str, ' ');
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
if ( (wpa_flags & NM_802_11_AP_SEC_KEY_MGMT_802_1X)
cli: 'dev wifi list': change SECURITY values * remove "Encrypted" tag The "Encrypted: " stuff was initially copied from nm-tool, but it doesn't help here much. See also http://cgit.freedesktop.org/NetworkManager/NetworkManager/commit/?id=a734c836a56f3170202f0555f1a03c9b2835775c for APs with WPA & RSN IEs, but no privacy flag. * remove a space from "WPA ", etc. strings. Translators often leave it out and thus break output. * change "WPA" to "WPA1" to make it clearer * use "802.1X" instead of "Enterprise" to save some characters
2013-05-24 22:36:24 +02:00
|| (rsn_flags & NM_802_11_AP_SEC_KEY_MGMT_802_1X)) {
g_string_append (security_str, _("802.1X"));
g_string_append_c (security_str, ' ');
}
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (security_str->len > 0)
g_string_truncate (security_str, security_str->len-1); /* Chop off last space */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array ((const NMMetaAbstractInfo *const*) nmc_fields_dev_wifi_list,
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
info->output_flags);
cli: put indexes into [] for AP and NSP in 'nmcli dev list' output AP1 -> AP[1] NSP1 -> NSP[1] It's more readable (section names are AP an NSP) and it makes indexing format conform to other data, like IP4.DNS[1].
2012-08-29 10:39:09 +02:00
ap_name = g_strdup_printf ("AP[%d]", info->index++); /* AP */
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_str (arr, 0, ap_name);
set_val_str (arr, 1, ssid_str);
cli: 'dev wifi list' use nm_utils_ssid_to_utf8() to display SSID; add SSID-HEX This commit removes enclosing '' from SSID. SSID-HEX is printed as a string of hex characters (each 2 chars represent one byte).
2013-05-16 16:26:15 +02:00
set_val_str (arr, 2, ssid_hex_str);
set_val_strc (arr, 3, bssid);
set_val_strc (arr, 4, mode == NM_802_11_MODE_ADHOC ? _("Ad-Hoc")
cli: 'dev wifi list': 'Infrastructure' -> 'Infra', 'Unknown' -> 'N/A' in MODE
2013-05-20 15:40:58 +02:00
: mode == NM_802_11_MODE_INFRA ? _("Infra")
: _("N/A"));
cli: 'dev wifi list': add CHAN column
2013-05-17 10:38:09 +02:00
set_val_str (arr, 5, channel_str);
set_val_str (arr, 6, freq_str);
set_val_str (arr, 7, bitrate_str);
set_val_str (arr, 8, strength_str);
cli: 'dev wifi list': add BARS column displaying signal semi-graphically using unicode characters: '_' '▂' '▄' '▆' '█' 0x5f 0x2582 0x2584 0x2586 0x2588
2013-05-17 11:23:21 +02:00
set_val_strc (arr, 9, sig_bars);
set_val_str (arr, 10, security_str->str);
set_val_str (arr, 11, wpa_flags_str);
set_val_str (arr, 12, rsn_flags_str);
set_val_strc (arr, 13, info->device);
set_val_strc (arr, 14, active ? _("yes") : _("no"));
cli: 'dev wifi list': add IN-USE (*) field
2013-05-20 16:36:20 +02:00
set_val_strc (arr, 15, active ? "*" : " ");
set_val_strc (arr, 16, nm_object_get_path (NM_OBJECT (ap)));
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
/* Set colors */
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
wifi_signal_to_color (strength, &color, &color_fmt);
set_val_color_all (arr, color);
set_val_color_fmt_all (arr, color_fmt);
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
if (active)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
arr[15].color = NM_META_TERM_COLOR_GREEN;
cli: color output of 'nmcli device wifi list'
2015-02-24 15:10:57 +01:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (info->output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
g_string_free (security_str, FALSE);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: show capabilities for Bluetooth devices
2014-09-24 12:00:47 +02:00
static char *
bluetooth_caps_to_string (NMBluetoothCapabilities caps)
{
char *caps_str[8]; /* Enough space for caps and terminating NULL */
char *ret_str;
int i = 0;
if (caps & NM_BT_CAPABILITY_DUN)
caps_str[i++] = g_strdup ("DUN");
if (caps & NM_BT_CAPABILITY_NAP)
caps_str[i++] = g_strdup ("NAP");
if (i == 0)
caps_str[i++] = g_strdup (_("(none)"));
caps_str[i] = NULL;
ret_str = g_strjoinv (" ", caps_str);
i = 0;
while (caps_str[i])
g_free (caps_str[i++]);
return ret_str;
}
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
static char *
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
construct_header_name (const char *base, const char *spec)
{
if (spec == NULL)
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
return g_strdup (base);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
return g_strdup_printf ("%s (%s)", base, spec);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: more compact format for device info (nmcli device show) (rh #998003) https://bugzilla.redhat.com/show_bug.cgi?id=998003
2013-12-13 14:19:43 +01:00
static const char *
get_active_connection_id (NMDevice *device)
{
NMActiveConnection *ac;
ac = nm_device_get_active_connection (device);
if (!ac)
return NULL;
cli: get connection ID straight from the active connection Static connection profile may not be available and using active conection is easier anyway.
2014-10-16 18:09:17 +02:00
return nm_active_connection_get_id (ac);
cli: more compact format for device info (nmcli device show) (rh #998003) https://bugzilla.redhat.com/show_bug.cgi?id=998003
2013-12-13 14:19:43 +01:00
}
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
static gboolean
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
print_bond_bridge_info (NMDevice *device,
NmCli *nmc,
const char *group_prefix,
const char *one_field)
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
{
const GPtrArray *slaves = NULL;
GString *slaves_str;
int idx;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NMMetaAbstractInfo *const*tmpl;
NmcOutputField *arr;
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
if (NM_IS_DEVICE_BOND (device))
slaves = nm_device_bond_get_slaves (NM_DEVICE_BOND (device));
else if (NM_IS_DEVICE_BRIDGE (device))
slaves = nm_device_bridge_get_slaves (NM_DEVICE_BRIDGE (device));
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
else
g_return_val_if_reached (FALSE);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
slaves_str = g_string_new (NULL);
for (idx = 0; slaves && idx < slaves->len; idx++) {
NMDevice *slave = g_ptr_array_index (slaves, idx);
const char *iface = nm_device_get_iface (slave);
if (iface) {
g_string_append (slaves_str, iface);
g_string_append_c (slaves_str, ' ');
}
}
if (slaves_str->len > 0)
g_string_truncate (slaves_str, slaves_str->len-1); /* Chop off last space */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_master_prop;
out_indices = parse_output_fields (one_field,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
set_val_strc (arr, 0, group_prefix); /* i.e. BOND, TEAM, BRIDGE */
set_val_str (arr, 1, slaves_str->str);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
g_string_free (slaves_str, FALSE);
return TRUE;
}
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
static char *
sanitize_team_config (const char *config)
{
char *ret;
int i;
if (!config)
return NULL;
ret = g_strdup (config);
for (i = 0; i < strlen (ret); i++) {
if (ret[i] == '\n')
ret[i] = ' ';
}
return ret;
}
static gboolean
print_team_info (NMDevice *device,
NmCli *nmc,
const char *group_prefix,
const char *one_field)
{
const GPtrArray *slaves = NULL;
GString *slaves_str;
int idx;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NMMetaAbstractInfo *const* tmpl;
NmcOutputField *arr;
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
if (NM_IS_DEVICE_TEAM (device))
slaves = nm_device_team_get_slaves (NM_DEVICE_TEAM (device));
else
g_return_val_if_reached (FALSE);
slaves_str = g_string_new (NULL);
for (idx = 0; slaves && idx < slaves->len; idx++) {
NMDevice *slave = g_ptr_array_index (slaves, idx);
const char *iface = nm_device_get_iface (slave);
if (iface) {
g_string_append (slaves_str, iface);
g_string_append_c (slaves_str, ' ');
}
}
if (slaves_str->len > 0)
g_string_truncate (slaves_str, slaves_str->len-1); /* Chop off last space */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_team_prop;
out_indices = parse_output_fields (one_field,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
set_val_strc (arr, 0, group_prefix); /* TEAM */
set_val_str (arr, 1, slaves_str->str);
set_val_str (arr, 2, sanitize_team_config (nm_device_team_get_config (NM_DEVICE_TEAM (device))));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
team: expose current device configuration through D-Bus and nmcli Add a new "Config" property to the D-Bus interface for team devices and show its value through "nmcli device show". The property contains the full JSON configuration from teamd for the device. https://bugzilla.redhat.com/show_bug.cgi?id=1310435
2016-05-16 14:42:41 +02:00
g_string_free (slaves_str, FALSE);
return TRUE;
}
cli: fix printing error on incorrect 'nmcli -f blah dev show' Don't print extra empty lines before the error.
2014-01-10 13:19:14 +01:00
static gboolean
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
show_device_info (NMDevice *device, NmCli *nmc)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
GError *error = NULL;
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
const char *hwaddr = NULL;
cli: fix uninitialized variable usage and clean up a bit
2010-03-25 13:39:49 -07:00
NMDeviceState state = NM_DEVICE_STATE_UNKNOWN;
cli: add REASON field to 'nmcli dev list' output
2012-02-09 15:39:57 +01:00
NMDeviceStateReason reason = NM_DEVICE_STATE_REASON_NONE;
cli: libnm-glib now uses enums for wired and WiFi capabilities
2011-02-21 16:38:02 +01:00
NMDeviceCapabilities caps;
cli: add fields to 'general' group of 'nmcli dev list's output
2011-12-20 15:54:58 +01:00
NMActiveConnection *acon;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
guint32 speed;
cli: display MTU for devices
2013-12-16 17:08:28 +01:00
char *speed_str, *state_str, *reason_str, *mtu_str;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
GArray *sections_array;
int k;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const char *fields_str = NULL;
const NMMetaAbstractInfo *const*tmpl;
NmcOutputField *arr;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
gboolean was_output = FALSE;
libnm: create NMIPConfig as parent of NMIP4Config and NMIP6Config Create NMIPConfig as the parent of NMIP4Config and NMIP6Config, and remove the two subclasses from the public API; while it's convenient to still have both internally, they are now identical to the outside world.
2014-10-22 13:48:18 -04:00
NMIPConfig *cfg4, *cfg6;
libnm: create NMDhcpConfig as parent of NMDhcp4Config and NMDhcp6Config As with NMIP4Config and NMIP6Config, merge the two DHCP config classes into one in the public API.
2014-10-22 13:48:18 -04:00
NMDhcpConfig *dhcp4, *dhcp6;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
const char *base_hdr = _("Device details");
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
GPtrArray *fields_in_section = NULL;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
if (!nmc->required_fields || strcasecmp (nmc->required_fields, "common") == 0)
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
fields_str = NMC_FIELDS_DEV_SHOW_SECTIONS_COMMON;
else if (strcasecmp (nmc->required_fields, "all") == 0) {
} else
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
fields_str = nmc->required_fields;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
sections_array = parse_output_fields (fields_str, (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_sections, TRUE, &fields_in_section, &error);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
if (error) {
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
g_string_printf (nmc->return_text, _("Error: 'device show': %s"), error->message);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
g_error_free (error);
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli: fix printing error on incorrect 'nmcli -f blah dev show' Don't print extra empty lines before the error.
2014-01-10 13:19:14 +01:00
return FALSE;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
{
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
gs_unref_array GArray *out_indices = NULL;
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
gs_free char *header_name = NULL;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
gs_free NmcOutputField *row = NULL;
int i;
nmcli: (trivial) add comment
2017-02-03 18:11:24 +01:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
/* Main header (pretty only) */
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
header_name = construct_header_name (base_hdr, nm_device_get_iface (device));
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
/* Lazy way to retrieve sorted array from 0 to the number of dev fields */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
out_indices = parse_output_fields (NULL,
(const NMMetaAbstractInfo *const*) nmc_fields_dev_show_general,
FALSE, NULL, NULL);
row = g_new0 (NmcOutputField, _NM_META_SETTING_TYPE_NUM + 1);
for (i = 0; i < _NM_META_SETTING_TYPE_NUM; i++)
row[i].info = (const NMMetaAbstractInfo *) &nmc_fields_dev_show_general[i];
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
print_required_fields (&nmc->nmc_config, NMC_OF_FLAG_MAIN_HEADER_ONLY,
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
out_indices, header_name,
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
0, row);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
/* Loop through the required sections and print them. */
for (k = 0; k < sections_array->len; k++) {
int section_idx = g_array_index (sections_array, int, k);
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
char *section_fld = (char *) g_ptr_array_index (fields_in_section, k);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (nmc->nmc_config.print_output != NMC_PRINT_TERSE && !nmc->nmc_config.multiline_output && was_output)
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_print ("\n"); /* Print empty line between groups in tabular mode */
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
was_output = FALSE;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
libnm: simplify NMDevice:state-reason The NMDevice:state-reason property was awkward to work with since it was a tuple of two values (state and reason), and likewise, although we had nm_device_get_state() to return just the state, there was no corresponding function to get just the reason. Fix this; make the state-reason property contain just the NMDeviceStateReason, and make nm_device_get_state_reason() return just that.
2014-08-25 10:51:26 -04:00
state = nm_device_get_state (device);
reason = nm_device_get_state_reason (device);
cli: fix uninitialized variable usage and clean up a bit
2010-03-25 13:39:49 -07:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
/* section GENERAL */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[0]->name)) {
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_general;
out_indices = parse_output_fields (section_fld,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: show progress in --pretty mode for 'nmcli con up' We need to use a workaround for VPN connection state. There is a bug somewhere causing that vpn-state-changed signal is not issued on active VPN connection. Debugging showed that D-Bus PropertiesChanged and VpnStateChanged signals are not processed properly. That's why libnm-glib's VPN connection object is not updated. The issue may lie in libnm-glib or even in dbus-glib. I wasn't able to get VpnStateChanged signal via dbus-glib (dbus_g_proxy_add_signal() and dbus_g_proxy_connect_signal()), however received the signal successfully using GDBus. Until the issue is analyzed and fixed, we need to ask for VPN state of active VPN connection every second in order to show progress and exit correctly.
2012-05-09 12:35:27 +02:00
state_str = g_strdup_printf ("%d (%s)", state, nmc_device_state_to_string (state));
reason_str = g_strdup_printf ("%d (%s)", reason, nmc_device_reason_to_string (reason));
cli: simplify getting device hardware addresses
2013-12-16 16:36:49 +01:00
hwaddr = nm_device_get_hw_address (device);
cli: display MTU for devices
2013-12-16 17:08:28 +01:00
mtu_str = g_strdup_printf ("%u", nm_device_get_mtu (device));
cli: more compact format for device info (nmcli device show) (rh #998003) https://bugzilla.redhat.com/show_bug.cgi?id=998003
2013-12-13 14:19:43 +01:00
acon = nm_device_get_active_connection (device);
cli: add REASON field to 'nmcli dev list' output
2012-02-09 15:39:57 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
set_val_strc (arr, 0, nmc_fields_dev_show_sections[0]->name); /* "GENERAL"*/
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_strc (arr, 1, nm_device_get_iface (device));
set_val_strc (arr, 2, nm_device_get_type_description (device));
cli: better indicate when a Wi-Fi plugin might not be available (rh #1168573) * print an error message indicating NM Wi-Fi plugin may be missing, for nmcli device wifi ifname <dev-name> nmcli device wifi connect ifname <dev-name> * add NM-TYPE to 'nmcli device show' command displaying internal NM device type (like NMDeviceWifi, NMDeviceGeneric, ...) https://bugzilla.redhat.com/show_bug.cgi?id=1168573
2015-04-14 14:35:14 +02:00
set_val_strc (arr, 3, G_OBJECT_TYPE_NAME (device));
set_val_strc (arr, 4, nm_device_get_vendor (device));
set_val_strc (arr, 5, nm_device_get_product (device));
set_val_strc (arr, 6, nm_device_get_driver (device) ? nm_device_get_driver (device) : _("(unknown)"));
set_val_strc (arr, 7, nm_device_get_driver_version (device));
set_val_strc (arr, 8, nm_device_get_firmware_version (device));
set_val_strc (arr, 9, hwaddr ? hwaddr : _("(unknown)"));
set_val_str (arr, 10, mtu_str);
set_val_str (arr, 11, state_str);
set_val_str (arr, 12, reason_str);
set_val_strc (arr, 13, nm_device_get_udi (device));
set_val_strc (arr, 14, nm_device_get_ip_iface (device));
set_val_strc (arr, 15, nm_device_is_software (device) ? _("yes") : _("no"));
set_val_strc (arr, 16, nm_device_get_managed (device) ? _("yes") : _("no"));
set_val_strc (arr, 17, nm_device_get_autoconnect (device) ? _("yes") : _("no"));
set_val_strc (arr, 18, nm_device_get_firmware_missing (device) ? _("yes") : _("no"));
cli: add NM-PLUGIN-MISSING to devices
2015-04-14 14:56:08 +02:00
set_val_strc (arr, 19, nm_device_get_nm_plugin_missing (device) ? _("yes") : _("no"));
set_val_strc (arr, 20, nm_device_get_physical_port_id (device));
set_val_strc (arr, 21, get_active_connection_id (device));
set_val_strc (arr, 22, acon ? nm_active_connection_get_uuid (acon) : NULL);
set_val_strc (arr, 23, acon ? nm_object_get_path (NM_OBJECT (acon)) : NULL);
cli: add 'metered' property to device
2015-04-30 11:52:04 +02:00
set_val_strc (arr, 24, nmc_device_metered_to_string (nm_device_get_metered (device)));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
was_output = TRUE;
}
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
/* section CAPABILITIES */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[1]->name)) {
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_cap;
out_indices = parse_output_fields (section_fld,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
caps = nm_device_get_capabilities (device);
speed = 0;
if (NM_IS_DEVICE_ETHERNET (device)) {
/* Speed in Mb/s */
speed = nm_device_ethernet_get_speed (NM_DEVICE_ETHERNET (device));
} else if (NM_IS_DEVICE_WIFI (device)) {
/* Speed in b/s */
speed = nm_device_wifi_get_bitrate (NM_DEVICE_WIFI (device));
speed /= 1000;
}
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
speed_str = speed ? g_strdup_printf (_("%u Mb/s"), speed) : g_strdup (_("unknown"));
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
set_val_strc (arr, 0, nmc_fields_dev_show_sections[1]->name); /* "CAPABILITIES" */
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_strc (arr, 1, (caps & NM_DEVICE_CAP_CARRIER_DETECT) ? _("yes") : _("no"));
set_val_str (arr, 2, speed_str);
cli: show whether the device is software-based in CAPABILITIES
2014-07-11 10:33:11 +02:00
set_val_strc (arr, 3, (caps & NM_DEVICE_CAP_IS_SOFTWARE) ? _("yes") : _("no"));
all: detect SR-IOV device support
2017-04-14 23:16:11 +02:00
set_val_strc (arr, 4, (caps & NM_DEVICE_CAP_SRIOV) ? _("yes") : _("no"));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
was_output = TRUE;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
/* Wireless specific information */
if ((NM_IS_DEVICE_WIFI (device))) {
cli: libnm-glib now uses enums for wired and WiFi capabilities
2011-02-21 16:38:02 +01:00
NMDeviceWifiCapabilities wcaps;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
NMAccessPoint *active_ap = NULL;
const char *active_bssid = NULL;
cli: sort access point list According to signal strength, frequency and rate.
2015-02-26 11:51:07 +01:00
GPtrArray *aps;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
/* section WIFI-PROPERTIES */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[2]->name)) {
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
wcaps = nm_device_wifi_get_capabilities (NM_DEVICE_WIFI (device));
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_wifi_prop;
out_indices = parse_output_fields (section_fld,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
set_val_strc (arr, 0, nmc_fields_dev_show_sections[2]->name); /* "WIFI-PROPERTIES" */
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_strc (arr, 1, (wcaps & (NM_WIFI_DEVICE_CAP_CIPHER_WEP40 | NM_WIFI_DEVICE_CAP_CIPHER_WEP104)) ?
_("yes") : _("no"));
set_val_strc (arr, 2, (wcaps & NM_WIFI_DEVICE_CAP_WPA) ? _("yes") : _("no"));
set_val_strc (arr, 3, (wcaps & NM_WIFI_DEVICE_CAP_RSN) ? _("yes") : _("no"));
set_val_strc (arr, 4, (wcaps & NM_WIFI_DEVICE_CAP_CIPHER_TKIP) ? _("yes") : _("no"));
set_val_strc (arr, 5, (wcaps & NM_WIFI_DEVICE_CAP_CIPHER_CCMP) ? _("yes") : _("no"));
set_val_strc (arr, 6, (wcaps & NM_WIFI_DEVICE_CAP_AP) ? _("yes") : _("no"));
set_val_strc (arr, 7, (wcaps & NM_WIFI_DEVICE_CAP_ADHOC) ? _("yes") : _("no"));
wifi: indicate 2ghz and 5ghz wifi device capabilities Add new capabilities CAP_FREQ_2GHZ and CAP_FREQ_5GHZ to indicate the frequency bands supported by a Wifi device. Add also CAP_FREQ_VALID, which is set when the values of the other 2 capabilities are available. Original patch by Dan Williams <dcbw@redhat.com> https://bugzilla.gnome.org/show_bug.cgi?id=723295
2015-03-18 21:56:51 +01:00
set_val_strc (arr, 8, !(wcaps & NM_WIFI_DEVICE_CAP_FREQ_VALID) ? _("unknown") :
((wcaps & NM_WIFI_DEVICE_CAP_FREQ_2GHZ) ? _("yes") : _("no")));
set_val_strc (arr, 9, !(wcaps & NM_WIFI_DEVICE_CAP_FREQ_VALID) ? _("unknown") :
((wcaps & NM_WIFI_DEVICE_CAP_FREQ_5GHZ) ? _("yes") : _("no")));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
was_output = TRUE;
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
/* section AP */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[3]->name)) {
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: fix uninitialized variable usage and clean up a bit
2010-03-25 13:39:49 -07:00
if (state == NM_DEVICE_STATE_ACTIVATED) {
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
active_ap = nm_device_wifi_get_active_access_point (NM_DEVICE_WIFI (device));
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
active_bssid = active_ap ? nm_access_point_get_bssid (active_ap) : NULL;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
}
cli: convert 'dev list' to new format; checking options validity for terse
2010-03-22 18:43:28 +01:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_wifi_list;
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
out_indices = parse_output_fields (section_fld ? section_fld : NMC_FIELDS_DEV_WIFI_LIST_FOR_DEV_LIST,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
{
APInfo info = {
.nmc = nmc,
.index = 1,
.output_flags = NMC_OF_FLAG_SECTION_PREFIX,
.active_bssid = active_bssid,
.device = nm_device_get_iface (device),
.output_data = out.output_data,
};
aps = sort_access_points (nm_device_wifi_get_access_points (NM_DEVICE_WIFI (device)));
g_ptr_array_foreach (aps, fill_output_access_point, &info);
g_ptr_array_free (aps, FALSE);
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
was_output = TRUE;
}
} else if (NM_IS_DEVICE_ETHERNET (device)) {
/* WIRED-PROPERTIES */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[4]->name)) {
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_wired_prop;
out_indices = parse_output_fields (section_fld,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
set_val_strc (arr, 0, nmc_fields_dev_show_sections[4]->name); /* "WIRED-PROPERTIES" */
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_strc (arr, 1, (nm_device_ethernet_get_carrier (NM_DEVICE_ETHERNET (device))) ?
_("on") : _("off"));
cli: print s390 subchannels for devices that have that
2015-09-23 16:07:08 +02:00
set_val_arrc (arr, 2, ((const char **) nm_device_ethernet_get_s390_subchannels (NM_DEVICE_ETHERNET (device))));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
was_output = TRUE;
}
cli/build: make WiMAX support optional
2011-02-14 16:06:37 +01:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: update displaying device details (IP and DHCP configuration)
2012-01-03 16:37:53 +01:00
/* IP configuration info */
core: add IP configuration for unmanaged devices Use NMPlatform to read the IP address/route configuration of unmanaged devices, and export that via D-Bus like we do with NM-generated IP configs.
2013-04-04 10:20:24 -04:00
cfg4 = nm_device_get_ip4_config (device);
cfg6 = nm_device_get_ip6_config (device);
dhcp4 = nm_device_get_dhcp4_config (device);
dhcp6 = nm_device_get_dhcp6_config (device);
/* IP4 */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (cfg4 && !strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[7]->name))
cli: use nmc_print() to output device's IP4 info The IP4 info adds a new type: to expose strv arguments for addresses, etc.
2017-04-06 15:14:23 +02:00
was_output = print_ip4_config (cfg4, &nmc->nmc_config, section_fld);
core: add IP configuration for unmanaged devices Use NMPlatform to read the IP address/route configuration of unmanaged devices, and export that via D-Bus like we do with NM-generated IP configs.
2013-04-04 10:20:24 -04:00
/* DHCP4 */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (dhcp4 && !strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[8]->name))
was_output = print_dhcp4_config (dhcp4, &nmc->nmc_config, nmc_fields_dev_show_sections[8]->name, section_fld);
core: add IP configuration for unmanaged devices Use NMPlatform to read the IP address/route configuration of unmanaged devices, and export that via D-Bus like we do with NM-generated IP configs.
2013-04-04 10:20:24 -04:00
/* IP6 */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (cfg6 && !strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[9]->name))
was_output = print_ip6_config (cfg6, &nmc->nmc_config, nmc_fields_dev_show_sections[9]->name, section_fld);
core: add IP configuration for unmanaged devices Use NMPlatform to read the IP address/route configuration of unmanaged devices, and export that via D-Bus like we do with NM-generated IP configs.
2013-04-04 10:20:24 -04:00
/* DHCP6 */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (dhcp6 && !strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[10]->name))
was_output = print_dhcp6_config (dhcp6, &nmc->nmc_config, nmc_fields_dev_show_sections[10]->name, section_fld);
cli: add BOND section to 'nmcli dev list' for bond devices Now, it just contains one item - SLAVES listing all slave devices of the bond device.
2012-10-22 14:31:11 +02:00
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
/* Bond specific information */
if (NM_IS_DEVICE_BOND (device)) {
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[11]->name))
was_output = print_bond_bridge_info (device, nmc, nmc_fields_dev_show_sections[11]->name, section_fld);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
}
cli: add BOND section to 'nmcli dev list' for bond devices Now, it just contains one item - SLAVES listing all slave devices of the bond device.
2012-10-22 14:31:11 +02:00
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
/* Team specific information */
if (NM_IS_DEVICE_TEAM (device)) {
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[12]->name))
was_output = print_team_info (device, nmc, nmc_fields_dev_show_sections[12]->name, section_fld);
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
}
cli: add BOND section to 'nmcli dev list' for bond devices Now, it just contains one item - SLAVES listing all slave devices of the bond device.
2012-10-22 14:31:11 +02:00
cli: also show team and bridge specific details for devices Now it is only SLAVES field, common for bond, team and bridge devices.
2014-09-23 14:49:43 +02:00
/* Bridge specific information */
if (NM_IS_DEVICE_BRIDGE (device)) {
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[13]->name))
was_output = print_bond_bridge_info (device, nmc, nmc_fields_dev_show_sections[13]->name, section_fld);
cli: add BOND section to 'nmcli dev list' for bond devices Now, it just contains one item - SLAVES listing all slave devices of the bond device.
2012-10-22 14:31:11 +02:00
}
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
/* VLAN-specific information */
if ((NM_IS_DEVICE_VLAN (device))) {
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[14]->name)) {
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
char * vlan_id_str = g_strdup_printf ("%u", nm_device_vlan_get_vlan_id (NM_DEVICE_VLAN (device)));
cli: add parent device information for VLAN devices
2014-07-21 13:08:53 +02:00
NMDevice *parent = nm_device_vlan_get_parent (NM_DEVICE_VLAN (device));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_vlan_prop;
out_indices = parse_output_fields (section_fld,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
set_val_strc (arr, 0, nmc_fields_dev_show_sections[14]->name); /* "VLAN" */
cli: add parent device information for VLAN devices
2014-07-21 13:08:53 +02:00
set_val_strc (arr, 1, parent ? nm_device_get_iface (parent) : NULL);
set_val_str (arr, 2, vlan_id_str);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: add VLAN section to 'nmcli dev list' for VLAN devices Now, it just contains one item - ID showing VLAN ID of the device.
2012-10-22 15:18:34 +02:00
was_output = TRUE;
}
}
cli: show capabilities for Bluetooth devices
2014-09-24 12:00:47 +02:00
if (NM_IS_DEVICE_BT (device)) {
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[15]->name)) {
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_bluetooth;
out_indices = parse_output_fields (section_fld,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: show capabilities for Bluetooth devices
2014-09-24 12:00:47 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
set_val_strc (arr, 0, nmc_fields_dev_show_sections[15]->name); /* "BLUETOOTH" */
cli: show capabilities for Bluetooth devices
2014-09-24 12:00:47 +02:00
set_val_str (arr, 1, bluetooth_caps_to_string (nm_device_bt_get_capabilities (NM_DEVICE_BT (device))));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: show capabilities for Bluetooth devices
2014-09-24 12:00:47 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: show capabilities for Bluetooth devices
2014-09-24 12:00:47 +02:00
was_output = TRUE;
}
}
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
/* section CONNECTIONS */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
if (!strcasecmp (nmc_fields_dev_show_sections[section_idx]->name, nmc_fields_dev_show_sections[16]->name)) {
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
const GPtrArray *avail_cons;
GString *ac_paths_str;
char **ac_arr = NULL;
int i;
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_show_connections;
out_indices = parse_output_fields (section_fld,
cli: merge NmcPrintFields into NmcOutputData and pass it directly to print_required_fields()
2017-03-31 12:55:43 +02:00
tmpl, FALSE, NULL, NULL);
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
/* available-connections */
avail_cons = nm_device_get_available_connections (device);
ac_paths_str = g_string_new (NULL);
libnm: change empty-GPtrArray-return semantics libnm functions that return GPtrArrays of objects had a rule that if the array was empty, they would return NULL rather than a 0-length array. As it turns out, this is just a nuisance to clients, since in most places the code for the non-empty case would end up doing the right thing for the empty case as well (and where it doesn't, we can check "array->len == 0" just as easily as "array == NULL"). So just return the 0-length array instead.
2014-08-25 10:33:53 -04:00
if (avail_cons->len) {
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
ac_arr = g_new (char *, avail_cons->len + 1);
ac_arr[avail_cons->len] = NULL;
}
libnm: change empty-GPtrArray-return semantics libnm functions that return GPtrArrays of objects had a rule that if the array was empty, they would return NULL rather than a 0-length array. As it turns out, this is just a nuisance to clients, since in most places the code for the non-empty case would end up doing the right thing for the empty case as well (and where it doesn't, we can check "array->len == 0" just as easily as "array == NULL"). So just return the 0-length array instead.
2014-08-25 10:33:53 -04:00
for (i = 0; i < avail_cons->len; i++) {
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
NMRemoteConnection *avail_con = g_ptr_array_index (avail_cons, i);
const char *ac_path = nm_connection_get_path (NM_CONNECTION (avail_con));
const char *ac_id = nm_connection_get_id (NM_CONNECTION (avail_con));
const char *ac_uuid = nm_connection_get_uuid (NM_CONNECTION (avail_con));
ac_arr[i] = g_strdup_printf ("%s | %s", ac_uuid, ac_id);
if (i == 0)
g_string_printf (ac_paths_str, "%s/{", NM_DBUS_PATH_SETTINGS);
else
g_string_append_c (ac_paths_str, ',');
g_string_append (ac_paths_str, strrchr (ac_path, '/') + 1);
}
if (ac_paths_str->len > 0)
g_string_append_c (ac_paths_str, '}');
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
set_val_strc (arr, 0, nmc_fields_dev_show_sections[16]->name); /* "CONNECTIONS" */
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_str (arr, 1, ac_paths_str->str);
set_val_arr (arr, 2, (ac_arr));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
g_string_free (ac_paths_str, FALSE);
cli: print NMDevice's 'available-connections' property in CONNECTIONS section of 'nmcli device list' Example: nmcli -f connections d l iface wlan5
2012-08-31 12:48:27 +02:00
was_output = TRUE;
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
if (sections_array)
g_array_free (sections_array, TRUE);
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
if (fields_in_section)
g_ptr_array_free (fields_in_section, TRUE);
cli: fix printing error on incorrect 'nmcli -f blah dev show' Don't print extra empty lines before the error.
2014-01-10 13:19:14 +01:00
return TRUE;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
void
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
nmc_device_state_to_color (NMDeviceState state, NMMetaTermColor *color, NMMetaTermFormat *color_fmt)
cli,devices: factor out device_state_to_color() We'll use this for device status monitor too.
2015-03-26 17:52:35 +01:00
{
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_NORMAL;
*color_fmt = NM_META_TERM_FORMAT_NORMAL;
cli,devices: factor out device_state_to_color() We'll use this for device status monitor too.
2015-03-26 17:52:35 +01:00
if (state <= NM_DEVICE_STATE_UNAVAILABLE)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color_fmt= NM_META_TERM_FORMAT_DIM;
cli,devices: factor out device_state_to_color() We'll use this for device status monitor too.
2015-03-26 17:52:35 +01:00
else if (state == NM_DEVICE_STATE_DISCONNECTED)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_RED;
cli,devices: factor out device_state_to_color() We'll use this for device status monitor too.
2015-03-26 17:52:35 +01:00
else if (state >= NM_DEVICE_STATE_PREPARE && state <= NM_DEVICE_STATE_SECONDARIES)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_YELLOW;
cli,devices: factor out device_state_to_color() We'll use this for device status monitor too.
2015-03-26 17:52:35 +01:00
else if (state == NM_DEVICE_STATE_ACTIVATED)
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
*color = NM_META_TERM_COLOR_GREEN;
cli,devices: factor out device_state_to_color() We'll use this for device status monitor too.
2015-03-26 17:52:35 +01:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static void
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
fill_output_device_status (NMDevice *device, GPtrArray *output_data)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
cli: add active connections to 'nmcli device status'
2013-12-09 12:21:44 +01:00
NMActiveConnection *ac;
nmcli: show devices list in colors
2015-02-16 10:56:45 +01:00
NMDeviceState state;
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
NMMetaTermColor color;
NMMetaTermFormat color_fmt;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
NmcOutputField *arr = nmc_dup_fields_array ((const NMMetaAbstractInfo *const*) nmc_fields_dev_status,
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
0);
nmcli: show devices list in colors
2015-02-16 10:56:45 +01:00
state = nm_device_get_state (device);
cli: add active connections to 'nmcli device status'
2013-12-09 12:21:44 +01:00
ac = nm_device_get_active_connection (device);
nmcli: show devices list in colors
2015-02-16 10:56:45 +01:00
/* Show devices in color */
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
nmc_device_state_to_color (state, &color, &color_fmt);
set_val_color_all (arr, color);
set_val_color_fmt_all (arr, color_fmt);
nmcli: show devices list in colors
2015-02-16 10:56:45 +01:00
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_strc (arr, 0, nm_device_get_iface (device));
set_val_strc (arr, 1, nm_device_get_type_description (device));
nmcli: show devices list in colors
2015-02-16 10:56:45 +01:00
set_val_strc (arr, 2, nmc_device_state_to_string (state));
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
set_val_strc (arr, 3, nm_object_get_path (NM_OBJECT (device)));
cli: add active connections to 'nmcli device status'
2013-12-09 12:21:44 +01:00
set_val_strc (arr, 4, get_active_connection_id (device));
set_val_strc (arr, 5, ac ? nm_active_connection_get_uuid (ac) : NULL);
set_val_strc (arr, 6, ac ? nm_object_get_path (NM_OBJECT (ac)) : NULL);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (output_data, arr);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
static NMCResultCode
do_devices_status (NmCli *nmc, int argc, char **argv)
{
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
GError *error = NULL;
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
NMDevice **devices;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
int i;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const char *fields_str = NULL;
const NMMetaAbstractInfo *const*tmpl;
NmcOutputField *arr;
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
/* Nothing to complete */
if (nmc->complete)
return nmc->return_value;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
while (argc > 0) {
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Unknown parameter: %s\n"), *argv);
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (!nmc->required_fields || strcasecmp (nmc->required_fields, "common") == 0)
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
fields_str = NMC_FIELDS_DEV_STATUS_COMMON;
else if (!nmc->required_fields || strcasecmp (nmc->required_fields, "all") == 0) {
} else
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
fields_str = nmc->required_fields;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_status;
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
out_indices = parse_output_fields (fields_str, tmpl, FALSE, NULL, &error);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (error) {
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
g_string_printf (nmc->return_text, _("Error: 'device status': %s"), error->message);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
g_error_free (error);
cli/device: use nmc_do_cmd()
2016-06-23 12:19:10 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
/* Add headers */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_MAIN_HEADER_ADD | NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
devices = nmc_get_devices_sorted (nmc->client);
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
for (i = 0; devices[i]; i++)
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
fill_output_device_status (devices[i], out.output_data);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, _("Status of devices"), 0, &out);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
g_free (devices);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
return NMC_RESULT_SUCCESS;
}
static NMCResultCode
cli/trivial: rename some functions for consistency Some functions that take device lists use plural form in name.
2016-06-23 18:49:39 +02:00
do_device_show (NmCli *nmc, int argc, char **argv)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
gs_free_error GError *error = NULL;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli/device: use nmc_do_cmd()
2016-06-23 12:19:10 +02:00
if (!nmc->mode_specified)
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
nmc->nmc_config_mutable.multiline_output = TRUE; /* multiline mode is default for 'device show' */
cli/device: use nmc_do_cmd()
2016-06-23 12:19:10 +02:00
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
if (argc) {
NMDevice *device;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
device = get_device (nmc, &argc, &argv, &error);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
if (!device) {
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
g_string_printf (nmc->return_text, _("Error: %s."), error->message);
return error->code;
}
if (argc) {
g_string_printf (nmc->return_text, _("Error: invalid extra argument '%s'."), *argv);
return NMC_RESULT_ERROR_USER_INPUT;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (nmc->complete)
return nmc->return_value;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
show_device_info (device, nmc);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
} else {
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
NMDevice **devices = nmc_get_devices_sorted (nmc->client);
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
int i;
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
/* nmc_do_cmd() should not call this with argc=0. */
g_assert (!nmc->complete);
cli: rename 'device list' -> 'device show' and remove 'iface' keyword Examples: nmcli device show - shows details for all available devices nmcli device show eth0 - shows details for eth0 nmcli device show em1.25 - shows details for em1.25 VLAN device
2013-04-03 10:01:42 +02:00
/* Show details for all devices */
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
for (i = 0; devices[i]; i++) {
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
if (!show_device_info (devices[i], nmc))
cli: fix printing error on incorrect 'nmcli -f blah dev show' Don't print extra empty lines before the error.
2014-01-10 13:19:14 +01:00
break;
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
if (devices[i + 1])
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_print ("\n"); /* Empty line */
cli: rename 'device list' -> 'device show' and remove 'iface' keyword Examples: nmcli device show - shows details for all available devices nmcli device show eth0 - shows details for eth0 nmcli device show em1.25 - shows details for em1.25 VLAN device
2013-04-03 10:01:42 +02:00
}
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
g_free (devices);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
return nmc->return_value;
}
static gboolean
timeout_cb (gpointer user_data)
{
/* Time expired -> exit nmcli */
NmCli *nmc = (NmCli *) user_data;
g_string_printf (nmc->return_text, _("Error: Timeout %d sec expired."), nmc->timeout);
nmc->return_value = NMC_RESULT_ERROR_TIMEOUT_EXPIRED;
quit ();
return FALSE;
}
cli: show progress in --pretty mode for 'nmcli dev disconnect'
2012-05-03 11:54:52 +02:00
static gboolean
progress_cb (gpointer user_data)
{
NMDevice *device = (NMDevice *) user_data;
cli: show progress in --pretty mode for 'nmcli con up' We need to use a workaround for VPN connection state. There is a bug somewhere causing that vpn-state-changed signal is not issued on active VPN connection. Debugging showed that D-Bus PropertiesChanged and VpnStateChanged signals are not processed properly. That's why libnm-glib's VPN connection object is not updated. The issue may lie in libnm-glib or even in dbus-glib. I wasn't able to get VpnStateChanged signal via dbus-glib (dbus_g_proxy_add_signal() and dbus_g_proxy_connect_signal()), however received the signal successfully using GDBus. Until the issue is analyzed and fixed, we need to ask for VPN state of active VPN connection every second in order to show progress and exit correctly.
2012-05-09 12:35:27 +02:00
nmc_terminal_show_progress (device ? nmc_device_state_to_string (nm_device_get_state (device)) : "");
cli: show progress in --pretty mode for 'nmcli dev disconnect'
2012-05-03 11:54:52 +02:00
return TRUE;
}
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
static void connected_state_cb (NMDevice *device, NMActiveConnection *active);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
static void
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
device_state_cb (NMDevice *device, GParamSpec *pspec, gpointer user_data)
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
{
cli: fix crash if active connection doesn't have a device (rh #1064333) $ nmcli dev connect wlan0 This can happen for example when some other process disconnects the connection being connected. Also fix printing a message on success. https://bugzilla.redhat.com/show_bug.cgi?id=1064333
2014-02-12 16:24:10 +01:00
NMActiveConnection *active = (NMActiveConnection *) user_data;
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
connected_state_cb (device, active);
}
static void
active_state_cb (NMActiveConnection *active, GParamSpec *pspec, gpointer user_data)
{
NMDevice *device = (NMDevice *) user_data;
connected_state_cb (device, active);
}
static void
connected_state_cb (NMDevice *device, NMActiveConnection *active)
{
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
NMDeviceState state;
cli: Finish waiting for the device activation when it disconnects The device status alone is uninteresting as its changes can be due to deactivation of previously active connection. We should monitor the active connection changes instead of device state changes. However the device state changes is still interesting, as it contains the reason for the change, let's just ignore them while the connection is activating. Lastly, we need to handle failures as well. It should be noted that it's not sufficient to deal with NM_DEVICE_STATE_FAILED as the device will quickly draverse to NM_DEVICE_STATE_DISCONNECTED. This happens in case of a failure due to NM_DEVICE_STATE_REASON_NO_SECRETS as soon as the server makes sure it won't reconnect automatically.
2014-11-14 18:29:36 +01:00
NMDeviceStateReason reason;
NMActiveConnectionState ac_state;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
state = nm_device_get_state (device);
cli: Finish waiting for the device activation when it disconnects The device status alone is uninteresting as its changes can be due to deactivation of previously active connection. We should monitor the active connection changes instead of device state changes. However the device state changes is still interesting, as it contains the reason for the change, let's just ignore them while the connection is activating. Lastly, we need to handle failures as well. It should be noted that it's not sufficient to deal with NM_DEVICE_STATE_FAILED as the device will quickly draverse to NM_DEVICE_STATE_DISCONNECTED. This happens in case of a failure due to NM_DEVICE_STATE_REASON_NO_SECRETS as soon as the server makes sure it won't reconnect automatically.
2014-11-14 18:29:36 +01:00
ac_state = nm_active_connection_get_state (active);
if (ac_state == NM_ACTIVE_CONNECTION_STATE_ACTIVATING)
return;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
if (state == NM_DEVICE_STATE_ACTIVATED) {
cli: fix crash if active connection doesn't have a device (rh #1064333) $ nmcli dev connect wlan0 This can happen for example when some other process disconnects the connection being connected. Also fix printing a message on success. https://bugzilla.redhat.com/show_bug.cgi?id=1064333
2014-02-12 16:24:10 +01:00
nmc_terminal_erase_line ();
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_print (_("Device '%s' successfully activated with '%s'.\n"),
nm_device_get_iface (device),
nm_active_connection_get_uuid (active));
cli: Finish waiting for the device activation when it disconnects The device status alone is uninteresting as its changes can be due to deactivation of previously active connection. We should monitor the active connection changes instead of device state changes. However the device state changes is still interesting, as it contains the reason for the change, let's just ignore them while the connection is activating. Lastly, we need to handle failures as well. It should be noted that it's not sufficient to deal with NM_DEVICE_STATE_FAILED as the device will quickly draverse to NM_DEVICE_STATE_DISCONNECTED. This happens in case of a failure due to NM_DEVICE_STATE_REASON_NO_SECRETS as soon as the server makes sure it won't reconnect automatically.
2014-11-14 18:29:36 +01:00
} else if ( state <= NM_DEVICE_STATE_DISCONNECTED
|| state >= NM_DEVICE_STATE_DEACTIVATING) {
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
reason = nm_device_get_state_reason (device);
cli: Finish waiting for the device activation when it disconnects The device status alone is uninteresting as its changes can be due to deactivation of previously active connection. We should monitor the active connection changes instead of device state changes. However the device state changes is still interesting, as it contains the reason for the change, let's just ignore them while the connection is activating. Lastly, we need to handle failures as well. It should be noted that it's not sufficient to deal with NM_DEVICE_STATE_FAILED as the device will quickly draverse to NM_DEVICE_STATE_DISCONNECTED. This happens in case of a failure due to NM_DEVICE_STATE_REASON_NO_SECRETS as soon as the server makes sure it won't reconnect automatically.
2014-11-14 18:29:36 +01:00
g_print (_("Error: Connection activation failed: (%d) %s.\n"),
reason, nmc_device_reason_to_string (reason));
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
} else
return;
g_signal_handlers_disconnect_by_func (active, G_CALLBACK (active_state_cb), device);
g_signal_handlers_disconnect_by_func (device, G_CALLBACK (device_state_cb), active);
g_object_unref (active);
g_object_unref (device);
quit ();
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
}
typedef struct {
NmCli *nmc;
NMDevice *device;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
gboolean hotspot;
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
} AddAndActivateInfo;
static void
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
add_and_activate_cb (GObject *client,
GAsyncResult *result,
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
gpointer user_data)
{
AddAndActivateInfo *info = (AddAndActivateInfo *) user_data;
NmCli *nmc = info->nmc;
NMDevice *device = info->device;
NMActiveConnectionState state;
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
NMActiveConnection *active;
GError *error = NULL;
active = nm_client_add_and_activate_connection_finish (NM_CLIENT (client), result, &error);
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
if (error) {
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
if (info->hotspot)
g_string_printf (nmc->return_text, _("Error: Failed to setup a Wi-Fi hotspot: %s"),
error->message);
else
g_string_printf (nmc->return_text, _("Error: Failed to add/activate new connection: %s"),
error->message);
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
g_error_free (error);
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
nmc->return_value = NMC_RESULT_ERROR_CON_ACTIVATION;
quit ();
} else {
state = nm_active_connection_get_state (active);
if (state == NM_ACTIVE_CONNECTION_STATE_UNKNOWN) {
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
if (info->hotspot)
g_string_printf (nmc->return_text, _("Error: Failed to setup a Wi-Fi hotspot"));
else
g_string_printf (nmc->return_text, _("Error: Failed to add/activate new connection: Unknown error"));
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
nmc->return_value = NMC_RESULT_ERROR_CON_ACTIVATION;
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
g_object_unref (active);
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
quit ();
}
if (nmc->nowait_flag || state == NM_ACTIVE_CONNECTION_STATE_ACTIVATED) {
/* User doesn't want to wait or already activated */
if (state == NM_ACTIVE_CONNECTION_STATE_ACTIVATED) {
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (nmc->nmc_config.print_output == NMC_PRINT_PRETTY)
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
nmc_terminal_erase_line ();
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
if (info->hotspot)
g_print (_("Connection with UUID '%s' created and activated on device '%s'\n"),
nm_active_connection_get_uuid (active), nm_device_get_iface (device));
else
g_print (_("Hotspot '%s' activated on device '%s'\n"),
nm_active_connection_get_id (active), nm_device_get_iface (device));
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
}
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
g_object_unref (active);
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
quit ();
} else {
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
g_object_ref (device);
g_signal_connect (device, "notify::state", G_CALLBACK (device_state_cb), active);
g_signal_connect (active, "notify::state", G_CALLBACK (active_state_cb), device);
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
g_timeout_add_seconds (nmc->timeout, timeout_cb, nmc); /* Exit if timeout expires */
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (nmc->nmc_config.print_output == NMC_PRINT_PRETTY)
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
progress_id = g_timeout_add (120, progress_cb, device);
}
}
g_free (info);
}
static void
create_connect_connection_for_device (AddAndActivateInfo *info)
{
NMConnection *connection;
NMSettingConnection *s_con;
/* Create new connection and tie it to the device */
connection = nm_simple_connection_new ();
s_con = (NMSettingConnection *) nm_setting_connection_new ();
nm_connection_add_setting (connection, NM_SETTING (s_con));
g_object_set (s_con,
NM_SETTING_CONNECTION_ID, nm_device_get_iface (info->device),
NM_SETTING_CONNECTION_INTERFACE_NAME, nm_device_get_iface (info->device),
NULL);
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
nm_client_add_and_activate_connection_async (info->nmc->client,
connection,
info->device,
NULL,
NULL,
add_and_activate_cb,
info);
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
}
static void
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
connect_device_cb (GObject *client, GAsyncResult *result, gpointer user_data)
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
{
AddAndActivateInfo *info = (AddAndActivateInfo *) user_data;
NmCli *nmc = info->nmc;
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
NMActiveConnection *active;
GError *error = NULL;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
const GPtrArray *devices;
NMDevice *device;
NMDeviceState state;
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
active = nm_client_activate_connection_finish (NM_CLIENT (client), result, &error);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
if (error) {
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
/* If no connection existed for the device, create one and activate it */
libnm-core, core: move NMManagerError to nm-errors Move the definition of NMManagerError to nm-errors, register it with D-Bus, and verify in the tests that it maps correctly. NM_MANAGER_ERROR_INTERNAL gets renamed to NM_MANAGER_ERROR_FAILED for consistency. NM_MANAGER_ERROR_UNMANAGED_DEVICE is dropped since that name doesn't really describe the one place it was previously used in. NM_MANAGER_ERROR_SYSTEM_CONNECTION is dropped because it was't being used. NM_MANAGER_ERROR_UNSUPPORTED_CONNECTION_TYPE is dropped because it can be replaced with an NM_CONNECTION_ERROR. NM_MANAGER_ERROR_AUTOCONNECT_NOT_ALLOWED is turned into the more generic NM_MANAGER_ERROR_CONNECTION_NOT_AVAILABLE. Also, remove the <tp:possible-errors> sections from nm-manager.xml, since they were completely out of date.
2014-10-15 15:27:25 -04:00
if (g_error_matches (error, NM_MANAGER_ERROR, NM_MANAGER_ERROR_UNKNOWN_CONNECTION)) {
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
create_connect_connection_for_device (info);
return;
}
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
g_string_printf (nmc->return_text, _("Error: Device activation failed: %s"),
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
error->message);
g_error_free (error);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
nmc->return_value = NMC_RESULT_ERROR_CON_ACTIVATION;
quit ();
} else {
g_assert (active);
devices = nm_active_connection_get_devices (active);
libnm: change empty-GPtrArray-return semantics libnm functions that return GPtrArrays of objects had a rule that if the array was empty, they would return NULL rather than a 0-length array. As it turns out, this is just a nuisance to clients, since in most places the code for the non-empty case would end up doing the right thing for the empty case as well (and where it doesn't, we can check "array->len == 0" just as easily as "array == NULL"). So just return the 0-length array instead.
2014-08-25 10:33:53 -04:00
if (devices->len == 0) {
cli: fix crash if active connection doesn't have a device (rh #1064333) $ nmcli dev connect wlan0 This can happen for example when some other process disconnects the connection being connected. Also fix printing a message on success. https://bugzilla.redhat.com/show_bug.cgi?id=1064333
2014-02-12 16:24:10 +01:00
g_string_printf (nmc->return_text, _("Error: Device activation failed: device was disconnected"));
nmc->return_value = NMC_RESULT_ERROR_CON_ACTIVATION;
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
g_object_unref (active);
cli: fix crash if active connection doesn't have a device (rh #1064333) $ nmcli dev connect wlan0 This can happen for example when some other process disconnects the connection being connected. Also fix printing a message on success. https://bugzilla.redhat.com/show_bug.cgi?id=1064333
2014-02-12 16:24:10 +01:00
quit ();
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
g_free (info);
cli: fix crash if active connection doesn't have a device (rh #1064333) $ nmcli dev connect wlan0 This can happen for example when some other process disconnects the connection being connected. Also fix printing a message on success. https://bugzilla.redhat.com/show_bug.cgi?id=1064333
2014-02-12 16:24:10 +01:00
return;
}
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
cli: fix crash if active connection doesn't have a device (rh #1064333) $ nmcli dev connect wlan0 This can happen for example when some other process disconnects the connection being connected. Also fix printing a message on success. https://bugzilla.redhat.com/show_bug.cgi?id=1064333
2014-02-12 16:24:10 +01:00
device = g_ptr_array_index (devices, 0);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
state = nm_device_get_state (device);
if (nmc->nowait_flag || state == NM_DEVICE_STATE_ACTIVATED) {
/* Don't want to wait or device already activated */
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (state == NM_DEVICE_STATE_ACTIVATED && nmc->nmc_config.print_output == NMC_PRINT_PRETTY) {
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
nmc_terminal_erase_line ();
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_print (_("Device '%s' has been connected.\n"), nm_device_get_iface (device));
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
}
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
g_object_unref (active);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
quit ();
} else {
cli: add support for secret agent to 'nmcli dev connect' too And move secrets getting code to common.c (without changes).
2014-11-20 12:36:14 +01:00
if (nmc->secret_agent) {
NMRemoteConnection *connection = nm_active_connection_get_connection (active);
clients: merge nm_secret_agent_simple_set_connection_path() into nm_secret_agent_simple_enable() set_connection_path() is almost always called right before enable(), and it's unclear why it would be called anywhere else. So just merge the two methods.
2014-11-20 16:00:34 -06:00
nm_secret_agent_simple_enable (NM_SECRET_AGENT_SIMPLE (nmc->secret_agent),
nm_connection_get_path (NM_CONNECTION (connection)));
cli: add support for secret agent to 'nmcli dev connect' too And move secrets getting code to common.c (without changes).
2014-11-20 12:36:14 +01:00
}
cli: Watch for device as well as AC state changes when activating The signals might be delivered in no particular order and we need to wait for the device to reach stable state (whether it's successfully conntected or not) as well as the active connection to leave ACTIVATING state. https://bugzilla.gnome.org/show_bug.cgi?id=740320
2014-11-18 15:55:55 +01:00
g_object_ref (device);
g_signal_connect (device, "notify::state", G_CALLBACK (device_state_cb), active);
g_signal_connect (active, "notify::state", G_CALLBACK (active_state_cb), device);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
/* Start timer not to loop forever if "notify::state" signal is not issued */
g_timeout_add_seconds (nmc->timeout, timeout_cb, nmc);
}
}
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
g_free (info);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
}
static NMCResultCode
do_device_connect (NmCli *nmc, int argc, char **argv)
{
NMDevice *device = NULL;
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
AddAndActivateInfo *info;
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
gs_free_error GError *error = NULL;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
/* Set default timeout for connect operation. */
if (nmc->timeout == -1)
nmc->timeout = 90;
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
device = get_device (nmc, &argc, &argv, &error);
if (!device) {
g_string_printf (nmc->return_text, _("Error: %s."), error->message);
return error->code;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
}
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
if (*argv) {
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
g_string_printf (nmc->return_text, _("Error: extra argument not allowed: '%s'."), *argv);
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
}
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (nmc->complete)
return nmc->return_value;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
/*
* Use nowait_flag instead of should_wait, because exiting has to be postponed
* till connect_device_cb() is called, giving NM time to check our permissions.
*/
nmc->nowait_flag = (nmc->timeout == 0);
cli: turn should_wait into a semaphore Count the exit blockers. We'll want to terminate the device monitor in case all monitored devices vanish.
2015-03-27 12:39:17 +01:00
nmc->should_wait++;
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
cli: add support for secret agent to 'nmcli dev connect' too And move secrets getting code to common.c (without changes).
2014-11-20 12:36:14 +01:00
/* Create secret agent */
nmc->secret_agent = nm_secret_agent_simple_new ("nmcli-connect");
clients: add define for NMSecretAgentSimple signal name
2016-09-09 17:33:09 +02:00
if (nmc->secret_agent) {
g_signal_connect (nmc->secret_agent,
NM_SECRET_AGENT_SIMPLE_REQUEST_SECRETS,
G_CALLBACK (nmc_secrets_requested),
nmc);
}
cli: add support for secret agent to 'nmcli dev connect' too And move secrets getting code to common.c (without changes).
2014-11-20 12:36:14 +01:00
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
info = g_malloc0 (sizeof (AddAndActivateInfo));
info->nmc = nmc;
info->device = device;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
info->hotspot = FALSE;
cli: create a connection if none exist in 'nmcli dev connect' (rh #1113941) https://bugzilla.redhat.com/show_bug.cgi?id=1113941
2014-09-13 00:35:05 +02:00
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
nm_client_activate_connection_async (nmc->client,
NULL, /* let NM find a connection automatically */
device,
NULL,
NULL,
connect_device_cb,
info);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
/* Start progress indication */
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (nmc->nmc_config.print_output == NMC_PRINT_PRETTY)
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
progress_id = g_timeout_add (120, progress_cb, device);
return nmc->return_value;
}
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
typedef struct {
NmCli *nmc;
GSList *queue;
guint timeout_id;
gboolean cmd_disconnect;
} DeviceCbInfo;
static void device_cb_info_finish (DeviceCbInfo *info, NMDevice *device);
static gboolean
device_op_timeout_cb (gpointer user_data)
{
DeviceCbInfo *info = user_data;
timeout_cb (info->nmc);
device_cb_info_finish (info, NULL);
return G_SOURCE_REMOVE;
}
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
static void
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
device_removed_cb (NMClient *client, NMDevice *device, DeviceCbInfo *info)
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
{
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
/* Success: device has been removed.
* It can also happen when disconnecting a software device.
*/
if (!g_slist_find (info->queue, device))
return;
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
if (info->cmd_disconnect)
g_print (_("Device '%s' successfully disconnected.\n"),
nm_device_get_iface (device));
else
g_print (_("Device '%s' successfully removed.\n"),
nm_device_get_iface (device));
device_cb_info_finish (info, device);
}
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
static void
disconnect_state_cb (NMDevice *device, GParamSpec *pspec, DeviceCbInfo *info)
{
if (!g_slist_find (info->queue, device))
return;
cli: stop waiting for deactivation if device reaches unmanaged state default-unmanaged devices enter unmanaged state right after they're disconnected.
2015-04-23 10:00:47 -04:00
if (nm_device_get_state (device) <= NM_DEVICE_STATE_DISCONNECTED) {
cli: fix nmcli timeout when disconnecting a device $ nmcli dev disconnect nm-bond Error: Timeout 10 sec expired. When a software device is disconnected, it will be removed. And it may not go to NM_DEVICE_STATE_DISCONNECTED state before that. So we need to listen to "device-removed" signal. Fixes:Beaker:test_log-NetworkManager_Test189_bond_activate
2014-11-10 16:56:34 +01:00
g_print (_("Device '%s' successfully disconnected.\n"),
nm_device_get_iface (device));
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
device_cb_info_finish (info, device);
cli: add support for 'nmcli dev connect ifname XXX'
2013-09-16 21:03:34 -05:00
}
}
cli: fix nmcli timeout when disconnecting a device $ nmcli dev disconnect nm-bond Error: Timeout 10 sec expired. When a software device is disconnected, it will be removed. And it may not go to NM_DEVICE_STATE_DISCONNECTED state before that. So we need to listen to "device-removed" signal. Fixes:Beaker:test_log-NetworkManager_Test189_bond_activate
2014-11-10 16:56:34 +01:00
static void
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
destroy_queue_element (gpointer data)
cli: fix nmcli timeout when disconnecting a device $ nmcli dev disconnect nm-bond Error: Timeout 10 sec expired. When a software device is disconnected, it will be removed. And it may not go to NM_DEVICE_STATE_DISCONNECTED state before that. So we need to listen to "device-removed" signal. Fixes:Beaker:test_log-NetworkManager_Test189_bond_activate
2014-11-10 16:56:34 +01:00
{
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
g_signal_handlers_disconnect_matched (data, G_SIGNAL_MATCH_FUNC, 0, 0, 0,
disconnect_state_cb, NULL);
g_object_unref (data);
}
static void
device_cb_info_finish (DeviceCbInfo *info, NMDevice *device)
{
if (device) {
GSList *elem = g_slist_find (info->queue, device);
if (!elem)
return;
info->queue = g_slist_delete_link (info->queue, elem);
destroy_queue_element (device);
} else {
g_slist_free_full (info->queue, destroy_queue_element);
info->queue = NULL;
}
if (info->queue)
return;
if (info->timeout_id)
g_source_remove (info->timeout_id);
g_signal_handlers_disconnect_by_func (info->nmc->client, device_removed_cb, info);
g_slice_free (DeviceCbInfo, info);
cli: fix nmcli timeout when disconnecting a device $ nmcli dev disconnect nm-bond Error: Timeout 10 sec expired. When a software device is disconnected, it will be removed. And it may not go to NM_DEVICE_STATE_DISCONNECTED state before that. So we need to listen to "device-removed" signal. Fixes:Beaker:test_log-NetworkManager_Test189_bond_activate
2014-11-10 16:56:34 +01:00
quit ();
}
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
static void
reapply_device_cb (GObject *object, GAsyncResult *result, gpointer user_data)
{
NMDevice *device = NM_DEVICE (object);
DeviceCbInfo *info = (DeviceCbInfo *) user_data;
NmCli *nmc = info->nmc;
GError *error = NULL;
if (!nm_device_reapply_finish (device, result, &error)) {
cli: don't allow multiple <ifname> arguments to device-reapply Just like `nmcli device connect` only allows one argument, don't allow multiple device arguments for reapply. Allowing multiple device names makes it more complicated to add additional options to the command. For example, it would be useful to have a nmcli device reapply eth0 connection id other-connection but when allowing multiple device names, it gets more complicated in documentation, command line parsing and bash completion. Note that the user can achieve a very similar outcome by using the shell: for DEV in eth0 eth1 eth2; do nmcli device reapply $DEV & done wait argubaly, this doesn't report the exit status properly. To properly handle that would require more effort. Also, it is somewhat less efficient, but well. This is an API change, however it is very new API that probably nobody is using much. Also, the documentation (man nmcli) didn't mention the possibility to pass multiple device names.
2016-04-30 14:14:31 +02:00
g_string_printf (nmc->return_text, _("Error: Reapplying connection to device '%s' (%s) failed: %s"),
nm_device_get_iface (device),
nm_object_get_path (NM_OBJECT (device)),
error->message);
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
g_error_free (error);
nmc->return_value = NMC_RESULT_ERROR_DEV_DISCONNECT;
device_cb_info_finish (info, device);
} else {
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (nmc->nmc_config.print_output == NMC_PRINT_PRETTY)
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
nmc_terminal_erase_line ();
g_print (_("Connection successfully reapplied to device '%s'.\n"),
cli: refactor cleanup in do_device_reapply()
2016-04-30 13:46:18 +02:00
nm_device_get_iface (device));
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
device_cb_info_finish (info, device);
}
}
static NMCResultCode
do_device_reapply (NmCli *nmc, int argc, char **argv)
{
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
NMDevice *device;
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
DeviceCbInfo *info = NULL;
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
gs_free_error GError *error = NULL;
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
/* Set default timeout for reapply operation. */
if (nmc->timeout == -1)
nmc->timeout = 10;
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
device = get_device (nmc, &argc, &argv, &error);
if (!device) {
g_string_printf (nmc->return_text, _("Error: %s."), error->message);
return error->code;
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
}
cli/device: split get_device() Parsing a single device name from the command line is generally useful. Remove the open coded versions in reapply, connect & status.
2016-06-27 11:54:25 +02:00
if (argc) {
g_string_printf (nmc->return_text, _("Error: invalid extra argument '%s'."), *argv);
return NMC_RESULT_ERROR_USER_INPUT;
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
}
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (nmc->complete)
return nmc->return_value;
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
nmc->nowait_flag = (nmc->timeout == 0);
cli: use should_wait consistently It's a semaphore, not a boolean.
2016-06-22 17:44:39 +02:00
nmc->should_wait++;
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
cli: don't allow multiple <ifname> arguments to device-reapply Just like `nmcli device connect` only allows one argument, don't allow multiple device arguments for reapply. Allowing multiple device names makes it more complicated to add additional options to the command. For example, it would be useful to have a nmcli device reapply eth0 connection id other-connection but when allowing multiple device names, it gets more complicated in documentation, command line parsing and bash completion. Note that the user can achieve a very similar outcome by using the shell: for DEV in eth0 eth1 eth2; do nmcli device reapply $DEV & done wait argubaly, this doesn't report the exit status properly. To properly handle that would require more effort. Also, it is somewhat less efficient, but well. This is an API change, however it is very new API that probably nobody is using much. Also, the documentation (man nmcli) didn't mention the possibility to pass multiple device names.
2016-04-30 14:14:31 +02:00
info = g_slice_new0 (DeviceCbInfo);
info->nmc = nmc;
info->queue = g_slist_prepend (info->queue, g_object_ref (device));
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
cli: don't allow multiple <ifname> arguments to device-reapply Just like `nmcli device connect` only allows one argument, don't allow multiple device arguments for reapply. Allowing multiple device names makes it more complicated to add additional options to the command. For example, it would be useful to have a nmcli device reapply eth0 connection id other-connection but when allowing multiple device names, it gets more complicated in documentation, command line parsing and bash completion. Note that the user can achieve a very similar outcome by using the shell: for DEV in eth0 eth1 eth2; do nmcli device reapply $DEV & done wait argubaly, this doesn't report the exit status properly. To properly handle that would require more effort. Also, it is somewhat less efficient, but well. This is an API change, however it is very new API that probably nobody is using much. Also, the documentation (man nmcli) didn't mention the possibility to pass multiple device names.
2016-04-30 14:14:31 +02:00
/* Now reapply the connection to the device */
nm_device_reapply_async (device, NULL, 0, 0, NULL, reapply_device_cb, info);
cli: add nmcli d reapply Client support for O.FD.NM.Device.Reapply().
2015-05-14 18:33:53 +02:00
return nmc->return_value;
}
cli/device: add modify command It modifies the applied connection using the Reapply API.
2016-06-22 18:25:48 +02:00
typedef struct {
NmCli *nmc;
int argc;
char **argv;
} ModifyInfo;
static void
modify_reapply_cb (GObject *object, GAsyncResult *result, gpointer user_data)
{
NMDevice *device = NM_DEVICE (object);
ModifyInfo *info = user_data;
NmCli *nmc = info->nmc;
GError *error = NULL;
if (!nm_device_reapply_finish (device, result, &error)) {
g_string_printf (nmc->return_text, _("Error: Reapplying connection to device '%s' (%s) failed: %s"),
nm_device_get_iface (device),
nm_object_get_path (NM_OBJECT (device)),
error->message);
g_error_free (error);
nmc->return_value = NMC_RESULT_ERROR_DEV_DISCONNECT;
} else {
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (nmc->nmc_config.print_output == NMC_PRINT_PRETTY)
cli/device: add modify command It modifies the applied connection using the Reapply API.
2016-06-22 18:25:48 +02:00
nmc_terminal_erase_line ();
g_print (_("Connection successfully reapplied to device '%s'.\n"),
nm_device_get_iface (device));
}
g_slice_free (ModifyInfo, info);
quit ();
}
static void
modify_get_applied_cb (GObject *object,
GAsyncResult *result,
gpointer user_data)
{
NMDevice *device = NM_DEVICE (object);
ModifyInfo *info = user_data;
NmCli *nmc = info->nmc;
gs_free_error GError *error = NULL;
NMConnection *connection;
guint64 version_id;
connection = nm_device_get_applied_connection_finish (device,
result,
&version_id,
&error);
if (!connection) {
g_string_printf (nmc->return_text, _("Error: Reading applied connection from device '%s' (%s) failed: %s"),
nm_device_get_iface (device),
nm_object_get_path (NM_OBJECT (device)),
error->message);
nmc->return_value = NMC_RESULT_ERROR_UNKNOWN;
g_slice_free (ModifyInfo, info);
quit ();
return;
}
if (!nmc_read_connection_properties (info->nmc, connection, &info->argc, &info->argv, &error)) {
g_string_assign (nmc->return_text, error->message);
nmc->return_value = error->code;
g_slice_free (ModifyInfo, info);
quit ();
return;
}
if (nmc->complete)
quit ();
else
nm_device_reapply_async (device, connection, version_id, 0, NULL, modify_reapply_cb, info);
}
static NMCResultCode
do_device_modify (NmCli *nmc, int argc, char **argv)
{
NMDevice *device = NULL;
ModifyInfo *info = NULL;
gs_free_error GError *error = NULL;
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli/device: add modify command It modifies the applied connection using the Reapply API.
2016-06-22 18:25:48 +02:00
device = get_device (nmc, &argc, &argv, &error);
if (!device) {
g_string_printf (nmc->return_text, _("Error: %s."), error->message);
return error->code;
}
if (nmc->timeout == -1)
nmc->timeout = 10;
nmc->nowait_flag = (nmc->timeout == 0);
nmc->should_wait++;
info = g_slice_new0 (ModifyInfo);
info->nmc = nmc;
info->argc = argc;
info->argv = argv;
nm_device_get_applied_connection_async (device, 0, NULL, modify_get_applied_cb, info);
return nmc->return_value;
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static void
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
disconnect_device_cb (GObject *object, GAsyncResult *result, gpointer user_data)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
NMDevice *device = NM_DEVICE (object);
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
DeviceCbInfo *info = (DeviceCbInfo *) user_data;
NmCli *nmc = info->nmc;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
NMDeviceState state;
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
GError *error = NULL;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
if (!nm_device_disconnect_finish (device, result, &error)) {
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
g_string_printf (nmc->return_text, _("Error: not all devices disconnected."));
g_printerr (_("Error: Device '%s' (%s) disconnecting failed: %s\n"),
nm_device_get_iface (device),
nm_object_get_path (NM_OBJECT (device)),
error->message);
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
g_error_free (error);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
nmc->return_value = NMC_RESULT_ERROR_DEV_DISCONNECT;
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
device_cb_info_finish (info, device);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
} else {
state = nm_device_get_state (device);
cli: stop waiting for deactivation if device reaches unmanaged state default-unmanaged devices enter unmanaged state right after they're disconnected.
2015-04-23 10:00:47 -04:00
if (nmc->nowait_flag || state <= NM_DEVICE_STATE_DISCONNECTED) {
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
/* Don't want to wait or device already disconnected */
cli: stop waiting for deactivation if device reaches unmanaged state default-unmanaged devices enter unmanaged state right after they're disconnected.
2015-04-23 10:00:47 -04:00
if (state <= NM_DEVICE_STATE_DISCONNECTED) {
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
if (nmc->nmc_config.print_output == NMC_PRINT_PRETTY)
cli: always print success message (not only in --pretty mode) (rh #1006444) The commands performing actions without an output only printed a successfull message for --pretty mode. Some users don't like this silent operation. This commit makes nmcli print operation success unconditionally. I think that doesn't hurt; scripts/users can ignore the message if not interested. https://bugzilla.redhat.com/show_bug.cgi?id=1006444
2013-09-19 15:15:10 +02:00
nmc_terminal_erase_line ();
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
g_print (_("Device '%s' successfully disconnected.\n"),
nm_device_get_iface (device));
cli: show progress in --pretty mode for 'nmcli dev disconnect'
2012-05-03 11:54:52 +02:00
}
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
device_cb_info_finish (info, device);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
}
}
static NMCResultCode
cli/trivial: rename some functions for consistency Some functions that take device lists use plural form in name.
2016-06-23 18:49:39 +02:00
do_devices_disconnect (NmCli *nmc, int argc, char **argv)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
NMDevice *device;
DeviceCbInfo *info = NULL;
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
GSList *queue, *iter;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: remove '--nowait' and '--timeout' options and use global '--wait' instead '--wait 0' corresponds to '--nowait' '--wait n' for n > 0 corresponds to '--timeout n'
2013-05-14 12:43:43 +02:00
/* Set default timeout for disconnect operation. */
if (nmc->timeout == -1)
nmc->timeout = 10;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli/trivial: rename device_list() to get_device_list() It will look nicer when we have get_device().
2016-06-22 18:33:54 +02:00
queue = get_device_list (nmc, argc, argv);
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
if (!queue)
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
return nmc->return_value;
if (nmc->complete)
goto out;
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
queue = g_slist_reverse (queue);
info = g_slice_new0 (DeviceCbInfo);
info->nmc = nmc;
info->cmd_disconnect = TRUE;
if (nmc->timeout > 0)
info->timeout_id = g_timeout_add_seconds (nmc->timeout, device_op_timeout_cb, info);
g_signal_connect (nmc->client, NM_CLIENT_DEVICE_REMOVED,
G_CALLBACK (device_removed_cb), info);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: remove '--nowait' and '--timeout' options and use global '--wait' instead '--wait 0' corresponds to '--nowait' '--wait n' for n > 0 corresponds to '--timeout n'
2013-05-14 12:43:43 +02:00
nmc->nowait_flag = (nmc->timeout == 0);
cli: turn should_wait into a semaphore Count the exit blockers. We'll want to terminate the device monitor in case all monitored devices vanish.
2015-03-27 12:39:17 +01:00
nmc->should_wait++;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
for (iter = queue; iter; iter = g_slist_next (iter)) {
device = iter->data;
info->queue = g_slist_prepend (info->queue, g_object_ref (device));
g_signal_connect (device, "notify::" NM_DEVICE_STATE,
G_CALLBACK (disconnect_state_cb), info);
/* Now disconnect the device */
nm_device_disconnect_async (device, NULL, disconnect_device_cb, info);
}
cli: show progress in --pretty mode for 'nmcli dev disconnect'
2012-05-03 11:54:52 +02:00
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
out:
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
g_slist_free (queue);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
return nmc->return_value;
}
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
static void
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
delete_device_cb (GObject *object, GAsyncResult *result, gpointer user_data)
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
{
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
NMDevice *device = NM_DEVICE (object);
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
DeviceCbInfo *info = (DeviceCbInfo *) user_data;
NmCli *nmc = info->nmc;
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
GError *error = NULL;
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
if (!nm_device_delete_finish (device, result, &error)) {
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
g_string_printf (nmc->return_text, _("Error: not all devices deleted."));
g_printerr (_("Error: Device '%s' (%s) deletion failed: %s\n"),
nm_device_get_iface (device),
nm_object_get_path (NM_OBJECT (device)),
error->message);
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
g_error_free (error);
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
nmc->return_value = NMC_RESULT_ERROR_UNKNOWN;
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
device_cb_info_finish (info, device);
} else {
if (nmc->nowait_flag)
device_cb_info_finish (info, device);
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
}
}
static NMCResultCode
cli/trivial: rename some functions for consistency Some functions that take device lists use plural form in name.
2016-06-23 18:49:39 +02:00
do_devices_delete (NmCli *nmc, int argc, char **argv)
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
{
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
NMDevice *device;
DeviceCbInfo *info = NULL;
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
GSList *queue, *iter;
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
/* Set default timeout for delete operation. */
if (nmc->timeout == -1)
nmc->timeout = 10;
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli/trivial: rename device_list() to get_device_list() It will look nicer when we have get_device().
2016-06-22 18:33:54 +02:00
queue = get_device_list (nmc, argc, argv);
cli,devices: factor out parsing of the interface list This is duplicated already and the monitor will use it.
2015-03-27 12:31:08 +01:00
if (!queue)
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
return nmc->return_value;
if (nmc->complete)
goto out;
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
queue = g_slist_reverse (queue);
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
info = g_slice_new0 (DeviceCbInfo);
info->nmc = nmc;
if (nmc->timeout > 0)
info->timeout_id = g_timeout_add_seconds (nmc->timeout, device_op_timeout_cb, info);
g_signal_connect (nmc->client, NM_CLIENT_DEVICE_REMOVED,
G_CALLBACK (device_removed_cb), info);
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
nmc->nowait_flag = (nmc->timeout == 0);
cli: turn should_wait into a semaphore Count the exit blockers. We'll want to terminate the device monitor in case all monitored devices vanish.
2015-03-27 12:39:17 +01:00
nmc->should_wait++;
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
for (iter = queue; iter; iter = g_slist_next (iter)) {
device = iter->data;
info->queue = g_slist_prepend (info->queue, g_object_ref (device));
/* Now delete the device */
nm_device_delete_async (device, NULL, delete_device_cb, info);
}
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
out:
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
g_slist_free (queue);
cli: add 'nmcli device delete <ifname>' command (rh #1034150) It only works for software devices. When called for a hardware device, an error is returned and the device is not deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1034150
2014-06-19 11:25:17 +02:00
return nmc->return_value;
}
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
static NMCResultCode
do_device_set (NmCli *nmc, int argc, char **argv)
{
#define DEV_SET_AUTOCONNECT 0
#define DEV_SET_MANAGED 1
NMDevice *device = NULL;
int i;
struct {
int idx;
gboolean value;
} values[2] = {
[DEV_SET_AUTOCONNECT] = { -1 },
[DEV_SET_MANAGED] = { -1 },
};
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
gs_free_error GError *error = NULL;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
if (argc >= 1 && g_strcmp0 (*argv, "ifname") == 0)
next_arg (nmc, &argc, &argv, NULL);
nmcli: extend `nmcli device set` command to accept explicit ifname specifier Without it, you cannot set the properties of a device named "help". Now you can with: $ nmcli device set ifname help autoconnect no
2015-09-18 10:49:19 +02:00
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
device = get_device (nmc, &argc, &argv, &error);
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
if (!device) {
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
g_string_printf (nmc->return_text, _("Error: %s."), error->message);
return error->code;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
}
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
if (!argc) {
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
g_string_printf (nmc->return_text, _("Error: No property specified."));
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
}
i = 0;
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
do {
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
gboolean flag;
cli: make subcommand dispatch do autocompletion
2016-06-24 18:07:24 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_strings (*argv, "managed", "autoconnect", NULL);
cli: make match() return boolean Coccinelle semantic patch: @@ @@ -int +gboolean matches (...); @@ expression pattern, cmd, len; @@ -int +gboolean matches (...) { ... - return memcmp (pattern, cmd, len); + return memcmp (pattern, cmd, len) == 0; } @@ expression prefix, str; @@ ( -matches (prefix, str) != 0 +!matches (prefix, str) | -matches (prefix, str) == 0 +matches (prefix, str) ) @@ expression prefix, str; @@ -(matches (prefix, str)) +matches (prefix, str) @@ expression prefix, str; @@ -(!matches (prefix, str)) +!matches (prefix, str) spatch --smpl-spacing --sp-file match.cocci --dir clients/cli/ \ --include-headers --macro-file shared/nm-utils/gsystem-local-alloc.h
2017-02-15 12:20:55 +01:00
if (matches (*argv, "managed")) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: improve error messages for 'nmcli device set' Fixes: 4dffbf8f6abe48a5677e5d702cb107f9946c824f
2015-11-25 16:39:00 +01:00
g_string_printf (nmc->return_text, _("Error: '%s' argument is missing."), *(argv-1));
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
}
cli: make subcommand dispatch do autocompletion
2016-06-24 18:07:24 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_bool (*argv);
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
if (!nmc_string_to_bool (*argv, &flag, &error)) {
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
g_string_printf (nmc->return_text, _("Error: 'managed': %s."),
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
error->message);
return NMC_RESULT_ERROR_USER_INPUT;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
}
values[DEV_SET_MANAGED].idx = ++i;
values[DEV_SET_MANAGED].value = flag;
}
cli: make match() return boolean Coccinelle semantic patch: @@ @@ -int +gboolean matches (...); @@ expression pattern, cmd, len; @@ -int +gboolean matches (...) { ... - return memcmp (pattern, cmd, len); + return memcmp (pattern, cmd, len) == 0; } @@ expression prefix, str; @@ ( -matches (prefix, str) != 0 +!matches (prefix, str) | -matches (prefix, str) == 0 +matches (prefix, str) ) @@ expression prefix, str; @@ -(matches (prefix, str)) +matches (prefix, str) @@ expression prefix, str; @@ -(!matches (prefix, str)) +!matches (prefix, str) spatch --smpl-spacing --sp-file match.cocci --dir clients/cli/ \ --include-headers --macro-file shared/nm-utils/gsystem-local-alloc.h
2017-02-15 12:20:55 +01:00
else if (matches (*argv, "autoconnect")) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: improve error messages for 'nmcli device set' Fixes: 4dffbf8f6abe48a5677e5d702cb107f9946c824f
2015-11-25 16:39:00 +01:00
g_string_printf (nmc->return_text, _("Error: '%s' argument is missing."), *(argv-1));
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
}
cli: make subcommand dispatch do autocompletion
2016-06-24 18:07:24 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_bool (*argv);
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
if (!nmc_string_to_bool (*argv, &flag, &error)) {
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
g_string_printf (nmc->return_text, _("Error: 'autoconnect': %s."),
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
error->message);
return NMC_RESULT_ERROR_USER_INPUT;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
}
values[DEV_SET_AUTOCONNECT].idx = ++i;
values[DEV_SET_AUTOCONNECT].value = flag;
}
else {
g_string_printf (nmc->return_text, _("Error: property '%s' is not known."), *argv);
cli/device: make "set" subcommand use get_devices()
2016-06-24 18:45:19 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
}
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
} while (next_arg (nmc, &argc, &argv, NULL) == 0);
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
cli: make subcommand dispatch do autocompletion
2016-06-24 18:07:24 +02:00
if (nmc->complete)
return nmc->return_value;
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
/* when multiple properties are specified, set them in the order as they
* are specified on the command line. */
if ( values[DEV_SET_AUTOCONNECT].idx >= 0
&& values[DEV_SET_MANAGED].idx >= 0
&& values[DEV_SET_MANAGED].idx < values[DEV_SET_AUTOCONNECT].idx) {
nm_device_set_managed (device, values[DEV_SET_MANAGED].value);
values[DEV_SET_MANAGED].idx = -1;
}
if (values[DEV_SET_AUTOCONNECT].idx >= 0)
nm_device_set_autoconnect (device, values[DEV_SET_AUTOCONNECT].value);
if (values[DEV_SET_MANAGED].idx >= 0)
cli: fix `nmcli device set DEV managed true` Fixes: 4dffbf8f6abe48a5677e5d702cb107f9946c824f
2015-11-06 14:51:40 +01:00
nm_device_set_managed (device, values[DEV_SET_MANAGED].value);
cli: add nmcli device set command Co-Authored-By: Thomas Haller <thaller@redhat.com>
2015-03-23 14:11:43 +01:00
return nmc->return_value;
}
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
static void
device_state (NMDevice *device, GParamSpec *pspec, NmCli *nmc)
{
NMDeviceState state = nm_device_get_state (device);
cli: move and rename TermColor and TermFormat
2017-04-04 15:12:00 +02:00
NMMetaTermColor color;
NMMetaTermFormat color_fmt;
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
char *str;
nmc_device_state_to_color (state, &color, &color_fmt);
cli: pass use_colors as separate option instead of global nmc nmc contains all options and collects output data. It is very hard to understand what it does. Start splitting it up, and pass arguments along -- as needed.
2017-03-30 12:23:26 +02:00
str = nmc_colorize (nmc->nmc_config.use_colors, color, color_fmt, "%s: %s\n",
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
nm_device_get_iface (device),
nmc_device_state_to_string (state));
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
g_print ("%s", str);
g_free (str);
}
static void
device_ac (NMDevice *device, GParamSpec *pspec, NmCli *nmc)
{
NMActiveConnection *ac = nm_device_get_active_connection (device);
const char *id = ac ? nm_active_connection_get_id (ac) : NULL;
if (!id)
return;
g_print (_("%s: using connection '%s'\n"), nm_device_get_iface (device), id);
}
static void
device_watch (NmCli *nmc, NMDevice *device)
{
nmc->should_wait++;
g_signal_connect (device, "notify::" NM_DEVICE_STATE, G_CALLBACK (device_state), nmc);
g_signal_connect (device, "notify::" NM_DEVICE_ACTIVE_CONNECTION, G_CALLBACK (device_ac), nmc);
}
static void
device_unwatch (NmCli *nmc, NMDevice *device)
{
g_signal_handlers_disconnect_by_func (device, device_state, nmc);
if (g_signal_handlers_disconnect_by_func (device, device_ac, nmc))
nmc->should_wait--;
/* Terminate if all the watched devices disappeared. */
if (!nmc->should_wait)
quit ();
}
static void
device_added (NMClient *client, NMDevice *device, NmCli *nmc)
{
g_print (_("%s: device created\n"), nm_device_get_iface (device));
device_watch (nmc, NM_DEVICE (device));
}
static void
device_removed (NMClient *client, NMDevice *device, NmCli *nmc)
{
g_print (_("%s: device removed\n"), nm_device_get_iface (device));
device_unwatch (nmc, device);
}
static NMCResultCode
cli/trivial: rename some functions for consistency Some functions that take device lists use plural form in name.
2016-06-23 18:49:39 +02:00
do_devices_monitor (NmCli *nmc, int argc, char **argv)
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
{
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
if (nmc->complete)
return nmc->return_value;
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
if (argc == 0) {
/* No devices specified. Monitor all. */
const GPtrArray *devices = nm_client_get_devices (nmc->client);
int i;
for (i = 0; i < devices->len; i++)
device_watch (nmc, g_ptr_array_index (devices, i));
/* We'll watch the device additions too, never exit. */
nmc->should_wait++;
g_signal_connect (nmc->client, NM_CLIENT_DEVICE_ADDED, G_CALLBACK (device_added), nmc);
} else {
nmcli: fix "nmcli device monitor <dev>" Call get_device_list() after next_arg(); call it only after checking that there are other arguments after "monitor". Fixes: 6a3d77fbe6271bb632e5272ef49c58e30739c38e
2017-04-11 15:39:08 +02:00
GSList *queue = get_device_list (nmc, argc, argv);
GSList *iter;
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
/* Monitor the specified devices. */
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
for (iter = queue; iter; iter = g_slist_next (iter))
device_watch (nmc, NM_DEVICE (iter->data));
g_slist_free (queue);
}
g_signal_connect (nmc->client, NM_CLIENT_DEVICE_REMOVED, G_CALLBACK (device_removed), nmc);
cli/device: add device name completion Useful with connect, disconnect, delete, monitor, show and reapply.
2016-06-23 10:49:20 +02:00
return nmc->return_value;
cli,devices: add device monitor
2015-03-27 13:11:06 +01:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static void
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
show_access_point_info (NMDevice *device, NmCli *nmc, NmcOutputData *out)
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
{
NMAccessPoint *active_ap = NULL;
const char *active_bssid = NULL;
cli: sort access point list According to signal strength, frequency and rate.
2015-02-26 11:51:07 +01:00
GPtrArray *aps;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
NmcOutputField *arr;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
if (nm_device_get_state (device) == NM_DEVICE_STATE_ACTIVATED) {
active_ap = nm_device_wifi_get_active_access_point (NM_DEVICE_WIFI (device));
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
active_bssid = active_ap ? nm_access_point_get_bssid (active_ap) : NULL;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array ((const NMMetaAbstractInfo *const*) nmc_fields_dev_wifi_list,
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
NMC_OF_FLAG_MAIN_HEADER_ADD | NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out->output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
{
APInfo info = {
.nmc = nmc,
.index = 1,
.output_flags = 0,
.active_bssid = active_bssid,
.device = nm_device_get_iface (device),
.output_data = out->output_data,
};
aps = sort_access_points (nm_device_wifi_get_access_points (NM_DEVICE_WIFI (device)));
g_ptr_array_foreach (aps, fill_output_access_point, &info);
g_ptr_array_free (aps, FALSE);
}
print_data_prepare_width (out->output_data);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli/trivial: move find_wifi_device_by_iface() up
2016-07-12 00:13:30 +02:00
/*
* Find a Wi-Fi device with 'iface' in 'devices' array. If 'iface' is NULL,
* the first Wi-Fi device is returned. 'idx' parameter is updated to the point
* where the function finished so that the function can be called repeatedly
* to get next matching device.
* Returns: found device or NULL
*/
static NMDevice *
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
find_wifi_device_by_iface (NMDevice **devices, const char *iface, int *idx)
cli/trivial: move find_wifi_device_by_iface() up
2016-07-12 00:13:30 +02:00
{
int i;
cli/device: make devices_idx argument to find_wifi_device_by_iface() optional Most callers don't use it.
2016-07-12 12:33:31 +02:00
for (i = idx ? *idx : 0; devices[i]; i++) {
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
const char *dev_iface = nm_device_get_iface (devices[i]);
cli/trivial: move find_wifi_device_by_iface() up
2016-07-12 00:13:30 +02:00
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
if (!NM_IS_DEVICE_WIFI (devices[i]))
cli/trivial: move find_wifi_device_by_iface() up
2016-07-12 00:13:30 +02:00
continue;
if (iface) {
/* If a iface was specified then use it. */
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
if (g_strcmp0 (dev_iface, iface) == 0)
cli/trivial: move find_wifi_device_by_iface() up
2016-07-12 00:13:30 +02:00
break;
} else {
/* Else return the first Wi-Fi device. */
break;
}
}
cli/device: make devices_idx argument to find_wifi_device_by_iface() optional Most callers don't use it.
2016-07-12 12:33:31 +02:00
if (idx)
*idx = i + 1;
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
return devices[i];
cli/trivial: move find_wifi_device_by_iface() up
2016-07-12 00:13:30 +02:00
}
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
/*
* Find AP on 'device' according to 'bssid' or 'ssid' parameter.
* Returns: found AP or NULL
*/
static NMAccessPoint *
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
find_ap_on_device (NMDevice *device, const char *bssid, const char *ssid, gboolean complete)
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
{
const GPtrArray *aps;
NMAccessPoint *ap = NULL;
int i;
g_return_val_if_fail (NM_IS_DEVICE_WIFI (device), NULL);
aps = nm_device_wifi_get_access_points (NM_DEVICE_WIFI (device));
for (i = 0; i < aps->len; i++) {
NMAccessPoint *candidate_ap = g_ptr_array_index (aps, i);
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
if (bssid) {
/* Parameter is BSSID */
const char *candidate_bssid = nm_access_point_get_bssid (candidate_ap);
/* Compare BSSIDs */
if (complete) {
if (g_str_has_prefix (candidate_bssid, bssid))
g_print ("%s\n", candidate_bssid);
} else if (strcmp (bssid, candidate_bssid) == 0) {
ap = candidate_ap;
break;
}
}
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
if (ssid) {
/* Parameter is SSID */
GBytes *candidate_ssid;
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
char *ssid_tmp;
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
candidate_ssid = nm_access_point_get_ssid (candidate_ap);
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
if (!candidate_ssid)
continue;
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
ssid_tmp = nm_utils_ssid_to_utf8 (g_bytes_get_data (candidate_ssid, NULL),
g_bytes_get_size (candidate_ssid));
/* Compare SSIDs */
if (complete) {
if (g_str_has_prefix (ssid_tmp, ssid))
g_print ("%s\n", ssid_tmp);
} else if (strcmp (ssid, ssid_tmp) == 0) {
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
ap = candidate_ap;
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
g_free (ssid_tmp);
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
break;
}
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
g_free (ssid_tmp);
cli/trivial: move find_ap_on_device() up
2016-07-12 14:31:25 +02:00
}
}
return ap;
}
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
static void
complete_aps (NMDevice **devices, const char *ifname,
const char *bssid_prefix, const char *ssid_prefix)
{
int devices_idx = 0;
NMDevice *device;
while ((device = find_wifi_device_by_iface (devices, ifname, &devices_idx)))
find_ap_on_device (device, bssid_prefix, ssid_prefix, TRUE);
}
void
nmc_complete_bssid (NMClient *client, const char *ifname, const char *bssid_prefix)
{
gs_free NMDevice **devices = NULL;
devices = nmc_get_devices_sorted (client);
complete_aps (devices, ifname, bssid_prefix, NULL);
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static NMCResultCode
do_device_wifi_list (NmCli *nmc, int argc, char **argv)
{
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
GError *error = NULL;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
NMDevice *device = NULL;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
NMAccessPoint *ap = NULL;
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
const char *ifname = NULL;
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
const char *bssid_user = NULL;
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
gs_free NMDevice **devices = NULL;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
const GPtrArray *aps;
APInfo *info;
int i, j;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const char *fields_str = NULL;
const NMMetaAbstractInfo *const*tmpl;
NmcOutputField *arr;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
const char *base_hdr = _("Wi-Fi scan list");
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
gs_free char *header_name = NULL;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
devices = nmc_get_devices_sorted (nmc->client);
cli/device: allow completion of the "wifi" command name That's the "wifi" string itself. The subcommands need some work.
2016-06-24 19:03:36 +02:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
while (argc > 0) {
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_strings (*argv, "ifname", "bssid", NULL);
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
if (strcmp (*argv, "ifname") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: allow next_arg() to move arguments past the last one
2012-11-16 22:43:29 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
ifname = *argv;
cli: remove spurious device names from output When running one of: nmcli device wifi list ifname wlan0 nmcli device wifi connect ... ifname wlan0 nmcli wrongly adds the device name to the output. Do the completion only when requested. Fixes: 8679793f6b711847d3209ab3cfbd95ab4a2e2488 Fixes: 1a0dfd31c4af5f3e42b4fc8363a2065d8c5a325b
2017-06-11 22:31:43 +02:00
if (argc == 1 && nmc->complete)
complete_device (devices, ifname, TRUE);
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
} else if (strcmp (*argv, "bssid") == 0 || strcmp (*argv, "hwaddr") == 0) {
/* hwaddr is deprecated and will be removed later */
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: allow next_arg() to move arguments past the last one
2012-11-16 22:43:29 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
bssid_user = *argv;
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
if (argc == 1 && nmc->complete)
complete_aps (devices, NULL, bssid_user, NULL);
} else if (!nmc->complete) {
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Unknown parameter: %s\n"), *argv);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (!nmc->required_fields || strcasecmp (nmc->required_fields, "common") == 0)
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
fields_str = NMC_FIELDS_DEV_WIFI_LIST_COMMON;
else if (!nmc->required_fields || strcasecmp (nmc->required_fields, "all") == 0) {
} else
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
fields_str = nmc->required_fields;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_wifi_list;
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
out_indices = parse_output_fields (fields_str, tmpl, FALSE, NULL, &error);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (error) {
cli: allow specifying 'group.name' syntax for '--fields' and use it for 'nmcli device show'. This allows filtering output not only for whole groups, but also for individual fields in commands that print data in groups (sections). Example: $ nmcli -f general.device,general.driver,ipv4,ipv6.address device show eth0 GENERAL.DEVICE: eth0 GENERAL.DRIVER: e1000e IP4.ADDRESS[1]: ip = 10.0.5.228/23, gw = 10.0.5.254 IP4.ADDRESS[2]: ip = 5.5.5.5/32, gw = 5.5.5.1 IP4.DNS[1]: 192.168.122.1 IP4.DNS[2]: 8.8.8.8 IP4.DOMAIN[1]: mycompany.com
2013-12-10 12:00:53 +01:00
g_string_printf (nmc->return_text, _("Error: 'device wifi': %s"), error->message);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
g_error_free (error);
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
if (nmc->complete)
return nmc->return_value;
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
if (ifname) {
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
cli/device: replace some duplicate code with find_wifi_device_by_iface()
2016-07-12 10:16:47 +02:00
device = find_wifi_device_by_iface (devices, ifname, NULL);
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
if (!device) {
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
g_string_printf (nmc->return_text, _("Error: Device '%s' not found."), ifname);
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
return NMC_RESULT_ERROR_NOT_FOUND;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
/* Main header name */
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
header_name = construct_header_name (base_hdr, ifname);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (NM_IS_DEVICE_WIFI (device)) {
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
if (bssid_user) {
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
/* Specific AP requested - list only that */
aps = nm_device_wifi_get_access_points (NM_DEVICE_WIFI (device));
libnm: change empty-GPtrArray-return semantics libnm functions that return GPtrArrays of objects had a rule that if the array was empty, they would return NULL rather than a 0-length array. As it turns out, this is just a nuisance to clients, since in most places the code for the non-empty case would end up doing the right thing for the empty case as well (and where it doesn't, we can check "array->len == 0" just as easily as "array == NULL"). So just return the 0-length array instead.
2014-08-25 10:33:53 -04:00
for (j = 0; j < aps->len; j++) {
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
char *bssid_up;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
NMAccessPoint *candidate_ap = g_ptr_array_index (aps, j);
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
const char *candidate_bssid = nm_access_point_get_bssid (candidate_ap);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
bssid_up = g_ascii_strup (bssid_user, -1);
if (!strcmp (bssid_up, candidate_bssid))
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
ap = candidate_ap;
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
g_free (bssid_up);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
if (!ap) {
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
g_string_printf (nmc->return_text, _("Error: Access point with bssid '%s' not found."),
bssid_user);
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
return NMC_RESULT_ERROR_NOT_FOUND;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
/* Add headers (field names) */
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_MAIN_HEADER_ADD | NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
info = g_malloc0 (sizeof (APInfo));
info->nmc = nmc;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
info->index = 1;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
info->output_flags = 0;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
info->active_bssid = NULL;
info->device = nm_device_get_iface (device);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
fill_output_access_point (ap, info);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, header_name, 0, &out);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
g_free (info);
} else {
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
show_access_point_info (device, nmc, &out);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, NULL, 0, &out);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
} else {
cli: better indicate when a Wi-Fi plugin might not be available (rh #1168573) * print an error message indicating NM Wi-Fi plugin may be missing, for nmcli device wifi ifname <dev-name> nmcli device wifi connect ifname <dev-name> * add NM-TYPE to 'nmcli device show' command displaying internal NM device type (like NMDeviceWifi, NMDeviceGeneric, ...) https://bugzilla.redhat.com/show_bug.cgi?id=1168573
2015-04-14 14:35:14 +02:00
if ( nm_device_get_device_type (device) == NM_DEVICE_TYPE_GENERIC
build: avoid compiler warnings about non-constant format strings (-Wformat-nonliteral)
2016-06-05 11:46:06 +02:00
&& g_strcmp0 (nm_device_get_type_description (device), "wifi") == 0) {
g_string_printf (nmc->return_text,
_("Error: Device '%s' was not recognized as a Wi-Fi device, check NetworkManager Wi-Fi plugin."),
ifname);
} else {
g_string_printf (nmc->return_text,
_("Error: Device '%s' is not a Wi-Fi device."),
ifname);
}
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
return NMC_RESULT_ERROR_UNKNOWN;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
} else {
cli: fix crash in `nmcli device wifi` with multiple wifi devices (bgo #734466) And separate details for particular devices with an empty line. https://bugzilla.gnome.org/show_bug.cgi?id=734466 Reported-by: Martin Pitt <martinpitt@gnome.org>
2014-08-18 12:41:01 +02:00
gboolean empty_line = FALSE;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
/* List APs for all devices */
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
if (bssid_user) {
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
/* Specific AP requested - list only that */
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
for (i = 0; devices[i]; i++) {
NMDevice *dev = devices[i];
cli: tighter scope NmcOutputData in do_device_wifi_list() The scope of "out" can be reduced.
2017-03-31 12:32:23 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out2);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
gs_free char *header_name2 = NULL;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (!NM_IS_DEVICE_WIFI (dev))
continue;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
/* Main header name */
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
header_name2 = construct_header_name (base_hdr, nm_device_get_iface (dev));
out2_indices = parse_output_fields (fields_str, tmpl, FALSE, NULL, NULL);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_MAIN_HEADER_ADD | NMC_OF_FLAG_FIELD_NAMES);
cli: tighter scope NmcOutputData in do_device_wifi_list() The scope of "out" can be reduced.
2017-03-31 12:32:23 +02:00
g_ptr_array_add (out2.output_data, arr);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
aps = nm_device_wifi_get_access_points (NM_DEVICE_WIFI (dev));
libnm: change empty-GPtrArray-return semantics libnm functions that return GPtrArrays of objects had a rule that if the array was empty, they would return NULL rather than a 0-length array. As it turns out, this is just a nuisance to clients, since in most places the code for the non-empty case would end up doing the right thing for the empty case as well (and where it doesn't, we can check "array->len == 0" just as easily as "array == NULL"). So just return the 0-length array instead.
2014-08-25 10:33:53 -04:00
for (j = 0; j < aps->len; j++) {
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
char *bssid_up;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
NMAccessPoint *candidate_ap = g_ptr_array_index (aps, j);
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
const char *candidate_bssid = nm_access_point_get_bssid (candidate_ap);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
bssid_up = g_ascii_strup (bssid_user, -1);
if (!strcmp (bssid_up, candidate_bssid)) {
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
ap = candidate_ap;
info = g_malloc0 (sizeof (APInfo));
info->nmc = nmc;
cli: 'dev list' adjustments Modifications are mainly for multiline mode to ease parsing - each field name is prefixed with a section name. 'dev list' now also supports printing particular sections specifed via '--fields' option.
2010-03-24 19:05:35 +01:00
info->index = 1;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
info->output_flags = 0;
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
info->active_bssid = NULL;
info->device = nm_device_get_iface (dev);
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
fill_output_access_point (ap, info);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
g_free (info);
}
cli: use nm_access_point_get_bssid() instead of nm_access_point_get_hw_address() nm_access_point_get_hw_address() is deprecated in 0.9. Also change 'hwaddr' parameter for 'nmcli dev wifi list' to 'bssid'. 'hwaddr' still works but is deprecated and not documented any more.
2011-09-06 13:37:23 +02:00
g_free (bssid_up);
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
cli: fix crash in `nmcli device wifi` with multiple wifi devices (bgo #734466) And separate details for particular devices with an empty line. https://bugzilla.gnome.org/show_bug.cgi?id=734466 Reported-by: Martin Pitt <martinpitt@gnome.org>
2014-08-18 12:41:01 +02:00
if (empty_line)
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_print ("\n"); /* Empty line between devices' APs */
cli: tighter scope NmcOutputData in do_device_wifi_list() The scope of "out" can be reduced.
2017-03-31 12:32:23 +02:00
print_data_prepare_width (out2.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out2_indices, header_name2, 0, &out2);
cli: fix crash in `nmcli device wifi` with multiple wifi devices (bgo #734466) And separate details for particular devices with an empty line. https://bugzilla.gnome.org/show_bug.cgi?id=734466 Reported-by: Martin Pitt <martinpitt@gnome.org>
2014-08-18 12:41:01 +02:00
empty_line = TRUE;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
if (!ap) {
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
g_string_printf (nmc->return_text, _("Error: Access point with bssid '%s' not found."),
bssid_user);
cli/device: add completion to wifi list subcommand
2016-07-12 10:16:24 +02:00
return NMC_RESULT_ERROR_NOT_FOUND;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
} else {
cli: sort the output of "nmcli dev" Sort the output of "nmcli dev", first by state (with connected devices first and unmanaged ones last), then by type and interface name.
2013-04-24 15:27:11 -04:00
for (i = 0; devices[i]; i++) {
NMDevice *dev = devices[i];
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out2);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
gs_free char *header_name2 = NULL;
cli: adapt the code for the new flags and printing facilities The data are added to output_data at first, and then they are printed all at once using print_data(), that takes care of proper alignment and display. The static 'width' values defined in NmcOutputField columns are not used now, but dynamically computed maximal widths override them.
2013-05-22 08:37:50 +02:00
/* Main header name */
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
header_name2 = construct_header_name (base_hdr,
nm_device_get_iface (dev));
out2_indices = parse_output_fields (fields_str, tmpl, FALSE, NULL, NULL);
cli: fix crash in `nmcli device wifi` with multiple wifi devices (bgo #734466) And separate details for particular devices with an empty line. https://bugzilla.gnome.org/show_bug.cgi?id=734466 Reported-by: Martin Pitt <martinpitt@gnome.org>
2014-08-18 12:41:01 +02:00
if (NM_IS_DEVICE_WIFI (dev)) {
if (empty_line)
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_print ("\n"); /* Empty line between devices' APs */
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
show_access_point_info (dev, nmc, &out2);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out2_indices, header_name2, 0, &out2);
cli: fix crash in `nmcli device wifi` with multiple wifi devices (bgo #734466) And separate details for particular devices with an empty line. https://bugzilla.gnome.org/show_bug.cgi?id=734466 Reported-by: Martin Pitt <martinpitt@gnome.org>
2014-08-18 12:41:01 +02:00
empty_line = TRUE;
}
cli: changing nmcli output to better fit both computer and human needs The output is basically tabular with fields (columns) presenting specific pieces of info. Each line represents a single object. It's possible to switch to multiline output using '--multiline' option. In that mode single object is presented on more lines - each field on its line. Terse mode now uses ':' as field separator. It also escapes all occurences of ':' and '\' inside field values to ease parsing. The escaping behaviour can be controlled through '--escape' option. By default, escaping is switched on in tabular mode. When using terse mode ('--terse'), '--fields' option is mandatory for specifying required fields. That helps for flexibility and backwards compatibility. Not all output is converted yet.
2010-03-18 15:39:15 +01:00
}
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
}
return nmc->return_value;
}
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
static NMCResultCode
do_device_wifi_connect_network (NmCli *nmc, int argc, char **argv)
{
NMDevice *device = NULL;
NMAccessPoint *ap = NULL;
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
NM80211ApFlags ap_flags = NM_802_11_AP_FLAGS_NONE;
NM80211ApSecurityFlags ap_wpa_flags = NM_802_11_AP_SEC_NONE;
NM80211ApSecurityFlags ap_rsn_flags = NM_802_11_AP_SEC_NONE;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
NMConnection *connection = NULL;
NMSettingConnection *s_con;
NMSettingWireless *s_wifi;
NMSettingWirelessSecurity *s_wsec;
AddAndActivateInfo *info;
const char *param_user = NULL;
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
const char *ifname = NULL;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
const char *bssid = NULL;
const char *password = NULL;
const char *con_name = NULL;
gboolean private = FALSE;
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
gboolean hidden = FALSE;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
gboolean wep_passphrase = FALSE;
GByteArray *bssid1_arr = NULL;
GByteArray *bssid2_arr = NULL;
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
gs_free NMDevice **devices = NULL;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
int devices_idx;
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
char *ssid_ask = NULL;
char *passwd_ask = NULL;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
cli: remove '--nowait' and '--timeout' options and use global '--wait' instead '--wait 0' corresponds to '--nowait' '--wait n' for n > 0 corresponds to '--timeout n'
2013-05-14 12:43:43 +02:00
/* Set default timeout waiting for operation completion. */
if (nmc->timeout == -1)
nmc->timeout = 90;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
devices = nmc_get_devices_sorted (nmc->client);
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
/* Get the first compulsory argument (SSID or BSSID) */
if (argc > 0) {
param_user = *argv;
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
bssid1_arr = nm_utils_hwaddr_atoba (param_user, ETH_ALEN);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
if (argc == 1 && nmc->complete)
complete_aps (devices, NULL, param_user, param_user);
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
} else {
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
/* nmc_do_cmd() should not call this with argc=0. */
g_assert (!nmc->complete);
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
if (nmc->ask) {
cli: use readline also in other commands (besides interactive editor) It is used for '--ask' option. The main use is for questionnaires (nmcli -a con add).
2014-05-14 09:22:50 +02:00
ssid_ask = nmc_readline (_("SSID or BSSID: "));
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
param_user = ssid_ask ? ssid_ask : "";
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
bssid1_arr = nm_utils_hwaddr_atoba (param_user, ETH_ALEN);
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
}
if (!ssid_ask) {
g_string_printf (nmc->return_text, _("Error: SSID or BSSID are missing."));
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
}
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
/* Get the rest of the parameters */
while (argc > 0) {
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
if (argc == 1 && nmc->complete) {
nmc_complete_strings (*argv, "ifname", "bssid", "password", "wep-key-type",
"name", "private", "hidden", NULL);
}
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
if (strcmp (*argv, "ifname") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: allow next_arg() to move arguments past the last one
2012-11-16 22:43:29 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
ifname = *argv;
cli: remove spurious device names from output When running one of: nmcli device wifi list ifname wlan0 nmcli device wifi connect ... ifname wlan0 nmcli wrongly adds the device name to the output. Do the completion only when requested. Fixes: 8679793f6b711847d3209ab3cfbd95ab4a2e2488 Fixes: 1a0dfd31c4af5f3e42b4fc8363a2065d8c5a325b
2017-06-11 22:31:43 +02:00
if (argc == 1 && nmc->complete)
complete_device (devices, ifname, TRUE);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
} else if (strcmp (*argv, "bssid") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: allow next_arg() to move arguments past the last one
2012-11-16 22:43:29 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
bssid = *argv;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
if (argc == 1 && nmc->complete)
complete_aps (devices, NULL, bssid, NULL);
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
bssid2_arr = nm_utils_hwaddr_atoba (bssid, ETH_ALEN);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
if (!bssid2_arr) {
g_string_printf (nmc->return_text, _("Error: bssid argument value '%s' is not a valid BSSID."),
bssid);
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
} else if (strcmp (*argv, "password") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: allow next_arg() to move arguments past the last one
2012-11-16 22:43:29 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
password = *argv;
} else if (strcmp (*argv, "wep-key-type") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: allow next_arg() to move arguments past the last one
2012-11-16 22:43:29 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_strings (*argv, "key", "phrase", NULL);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
if (strcmp (*argv, "key") == 0)
wep_passphrase = FALSE;
else if (strcmp (*argv, "phrase") == 0)
wep_passphrase = TRUE;
else {
g_string_printf (nmc->return_text,
_("Error: wep-key-type argument value '%s' is invalid, use 'key' or 'phrase'."),
*argv);
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
} else if (strcmp (*argv, "name") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: allow next_arg() to move arguments past the last one
2012-11-16 22:43:29 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
con_name = *argv;
cli: change '--private' to 'private yes|no' in 'nmcli dev wifi connect' Missing 'private' argument default to private=no, the same way as before.
2013-05-14 13:51:06 +02:00
} else if (strcmp (*argv, "private") == 0) {
GError *err_tmp = NULL;
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: change '--private' to 'private yes|no' in 'nmcli dev wifi connect' Missing 'private' argument default to private=no, the same way as before.
2013-05-14 13:51:06 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: change '--private' to 'private yes|no' in 'nmcli dev wifi connect' Missing 'private' argument default to private=no, the same way as before.
2013-05-14 13:51:06 +02:00
}
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_bool (*argv);
cli: change '--private' to 'private yes|no' in 'nmcli dev wifi connect' Missing 'private' argument default to private=no, the same way as before.
2013-05-14 13:51:06 +02:00
if (!nmc_string_to_bool (*argv, &private, &err_tmp)) {
g_string_printf (nmc->return_text, _("Error: %s: %s."), *(argv-1), err_tmp->message);
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
g_clear_error (&err_tmp);
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: change '--private' to 'private yes|no' in 'nmcli dev wifi connect' Missing 'private' argument default to private=no, the same way as before.
2013-05-14 13:51:06 +02:00
}
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
} else if (strcmp (*argv, "hidden") == 0) {
GError *err_tmp = NULL;
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
}
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_bool (*argv);
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
if (!nmc_string_to_bool (*argv, &hidden, &err_tmp)) {
g_string_printf (nmc->return_text, _("Error: %s: %s."), *(argv-1), err_tmp->message);
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
g_clear_error (&err_tmp);
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
}
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
} else if (!nmc->complete) {
cli: use g_print/g_printerr rather than printf/fprintf Use g_print() and g_printerr(), which wrap printf() and fprintf(stderr,...), but handle conversion from UTF-8 if the locale is using a different character set.
2014-09-19 16:04:40 -04:00
g_printerr (_("Unknown parameter: %s\n"), *argv);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
if (nmc->complete)
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
/* Verify SSID/BSSID parameters */
if (bssid1_arr && bssid2_arr && memcmp (bssid1_arr->data, bssid2_arr->data, ETH_ALEN)) {
g_string_printf (nmc->return_text, _("Error: BSSID to connect to (%s) differs from bssid argument (%s)."),
param_user, bssid);
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
if (!bssid1_arr && strlen (param_user) > 32) {
g_string_printf (nmc->return_text, _("Error: Parameter '%s' is neither SSID nor BSSID."), param_user);
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
/* Find a device to activate the connection on */
devices_idx = 0;
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
device = find_wifi_device_by_iface (devices, ifname, &devices_idx);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
if (!device) {
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
if (ifname)
g_string_printf (nmc->return_text, _("Error: Device '%s' is not a Wi-Fi device."), ifname);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
else
g_string_printf (nmc->return_text, _("Error: No Wi-Fi device found."));
nmc->return_value = NMC_RESULT_ERROR_UNKNOWN;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
/* For hidden SSID first scan it so that NM learns about the AP */
if (hidden) {
GVariantBuilder builder, array_builder;
GVariant *options;
GError *scan_err = NULL;
g_variant_builder_init (&builder, G_VARIANT_TYPE_VARDICT);
g_variant_builder_init (&array_builder, G_VARIANT_TYPE ("aay"));
g_variant_builder_add (&array_builder, "@ay",
g_variant_new_fixed_array (G_VARIANT_TYPE_BYTE, param_user, strlen (param_user), 1));
g_variant_builder_add (&builder, "{sv}", "ssids", g_variant_builder_end (&array_builder));
options = g_variant_builder_end (&builder);
nm_device_wifi_request_scan_options (NM_DEVICE_WIFI (device), options, NULL, &scan_err);
if (scan_err) {
g_string_printf (nmc->return_text, _("Error: Failed to scan hidden SSID: %s."),
scan_err->message);
g_clear_error (&scan_err);
nmc->return_value = NMC_RESULT_ERROR_NOT_FOUND;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
}
}
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
/* Find an AP to connect to */
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
ap = find_ap_on_device (device, bssid1_arr ? param_user : NULL,
bssid1_arr ? NULL : param_user, FALSE);
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
if (!ap && !ifname) {
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
NMDevice *dev;
/* AP not found, ifname was not specified, so try finding the AP on another device. */
while ((dev = find_wifi_device_by_iface (devices, NULL, &devices_idx)) != NULL) {
cli/device: export some useful completion functions Wi-Fi access point completion functionality and stuff useful for completing "ifname" and "ap" for "nmcli c up".
2016-07-12 14:36:53 +02:00
ap = find_ap_on_device (dev, bssid1_arr ? param_user : NULL,
bssid1_arr ? NULL : param_user, FALSE);
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
if (ap) {
device = dev;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
break;
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
}
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
}
if (!ap) {
if (!bssid1_arr)
g_string_printf (nmc->return_text, _("Error: No network with SSID '%s' found."), param_user);
else
g_string_printf (nmc->return_text, _("Error: No access point with BSSID '%s' found."), param_user);
cli: add NMC_RESULT_ERROR_NOT_FOUND and use as appropriate This will allow callers to distinguish "no such connection" from "connection failed to activate", etc. https://bugzilla.redhat.com/show_bug.cgi?id=1022256
2013-11-21 09:19:18 -05:00
nmc->return_value = NMC_RESULT_ERROR_NOT_FOUND;
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
goto finish;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
/* If there are some connection data from user, create a connection and
* fill them into proper settings. */
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
if (con_name || private || bssid2_arr || password || hidden)
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 ();
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
if (con_name || private) {
s_con = (NMSettingConnection *) nm_setting_connection_new ();
nm_connection_add_setting (connection, NM_SETTING (s_con));
/* Set user provided connection name */
if (con_name)
g_object_set (s_con, NM_SETTING_CONNECTION_ID, con_name, NULL);
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
/* Connection will only be visible to this user when 'private' is specified */
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
if (private)
nm_setting_connection_add_permission (s_con, "user", g_get_user_name (), NULL);
}
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
if (bssid2_arr || hidden) {
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
s_wifi = (NMSettingWireless *) nm_setting_wireless_new ();
nm_connection_add_setting (connection, NM_SETTING (s_wifi));
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
/* 'bssid' parameter is used to restrict the connection only to the BSSID */
if (bssid2_arr)
g_object_set (s_wifi, NM_SETTING_WIRELESS_BSSID, bssid2_arr, NULL);
/* 'hidden' parameter is used to indicate that SSID is not broadcasted */
if (hidden) {
GBytes *ssid = g_bytes_new (param_user, strlen (param_user));
g_object_set (s_wifi,
NM_SETTING_WIRELESS_SSID, ssid,
NM_SETTING_WIRELESS_HIDDEN, hidden,
NULL);
g_bytes_unref (ssid);
cli: print a warning if BSSID is given instead of SSID for hidden AP $ nmcli dev wifi connect 00:22:6B:EB:1D:CA hidden yes Warning: '00:22:6B:EB:1D:CA' should be SSID for hidden APs; but it looks like a BSSID. Error: Failed to add/activate new connection: 802-11-wireless.ssid: connection does not match access point
2015-11-09 14:33:34 +01:00
/* Warn when the provided AP identifier looks like BSSID instead of SSID */
if (bssid1_arr)
g_printerr (_("Warning: '%s' should be SSID for hidden APs; but it looks like a BSSID.\n"),
param_user);
cli: fix connecting to a hidden SSID with 'nmcli dev wifi connect' (bgo #752173) We have to set 802-11-wireless.hidden=true to be able to connect to hidden SSIDs. nmcli user indicates hidden SSID with 'hidden yes' parameter. https://bugzilla.gnome.org/show_bug.cgi?id=752173
2015-07-10 13:16:21 +02:00
}
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
}
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
/* handle password */
ap_flags = nm_access_point_get_flags (ap);
ap_wpa_flags = nm_access_point_get_wpa_flags (ap);
ap_rsn_flags = nm_access_point_get_rsn_flags (ap);
/* Set password for WEP or WPA-PSK. */
if (ap_flags & NM_802_11_AP_FLAGS_PRIVACY) {
/* Ask for missing password when one is expected and '--ask' is used */
if (!password && nmc->ask)
cli: move show_secret field to NmcConfig The show-secret property is basically a part of the current configuration, relevant during printing. It will be passed on to nmc_print(), and hence must be part of NmcConfig.
2017-04-06 15:26:23 +02:00
password = passwd_ask = nmc_readline_echo (nmc->nmc_config.show_secrets, _("Password: "));
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
if (password) {
if (!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 ();
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
s_wsec = (NMSettingWirelessSecurity *) nm_setting_wireless_security_new ();
nm_connection_add_setting (connection, NM_SETTING (s_wsec));
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
if (ap_wpa_flags == NM_802_11_AP_SEC_NONE && ap_rsn_flags == NM_802_11_AP_SEC_NONE) {
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
/* WEP */
nm_setting_wireless_security_set_wep_key (s_wsec, 0, password);
g_object_set (G_OBJECT (s_wsec),
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_TYPE,
wep_passphrase ? NM_WEP_KEY_TYPE_PASSPHRASE: NM_WEP_KEY_TYPE_KEY,
NULL);
cli: wifi: connect with PSK when the AP supports WPA-PSK and WPA-EAP 'nmcli device wifi connect' only supports WEP and WPA-PSK at the moment, but not WPA-EAP. If the AP supports both WPA-PSK and WPA-EAP, nmcli doesn't add the PSK to the connection, causing a connection failure. Fix this. https://bugzilla.redhat.com/show_bug.cgi?id=1492064
2017-09-16 11:59:46 +02:00
} else if ( (ap_wpa_flags & NM_802_11_AP_SEC_KEY_MGMT_PSK)
|| (ap_rsn_flags & NM_802_11_AP_SEC_KEY_MGMT_PSK)) {
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
/* WPA PSK */
g_object_set (s_wsec, NM_SETTING_WIRELESS_SECURITY_PSK, password, NULL);
}
}
}
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
// FIXME: WPA-Enterprise is not supported yet.
// We are not able to determine and fill all the parameters for
// 802.1X authentication automatically without user providing
// the data. Adding nmcli options for the 8021x setting would
// clutter the command. However, that could be solved later by
// implementing add/edit connections support for nmcli.
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
/* nowait_flag indicates user input. should_wait says whether quit in start().
* We have to delay exit after add_and_activate_cb() is called, even if
* the user doesn't want to wait, in order to give NM time to check our
* permissions. */
cli: remove '--nowait' and '--timeout' options and use global '--wait' instead '--wait 0' corresponds to '--nowait' '--wait n' for n > 0 corresponds to '--timeout n'
2013-05-14 12:43:43 +02:00
nmc->nowait_flag = (nmc->timeout == 0);
cli: turn should_wait into a semaphore Count the exit blockers. We'll want to terminate the device monitor in case all monitored devices vanish.
2015-03-27 12:39:17 +01:00
nmc->should_wait++;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
info = g_malloc0 (sizeof (AddAndActivateInfo));
info->nmc = nmc;
info->device = device;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
info->hotspot = FALSE;
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
nm_client_add_and_activate_connection_async (nmc->client,
connection,
device,
nm_object_get_path (NM_OBJECT (ap)),
NULL,
add_and_activate_cb,
info);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
cli/device: add completion to wifi connect subcommand
2016-07-12 09:45:23 +02:00
finish:
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
if (bssid1_arr)
g_byte_array_free (bssid1_arr, TRUE);
if (bssid2_arr)
g_byte_array_free (bssid2_arr, TRUE);
cli: 'device wifi connect' - ask for SSID/BSSID and password if not provided when '--ask' parameter is specified
2012-11-22 14:32:45 +01:00
g_free (ssid_ask);
g_free (passwd_ask);
cli: add 'connect to SSID' feature: nmcli dev wifi connect <SSID> ... This is similar to clicking a Wi-Fi network's SSID in a GUI applet. The command does this: - creates new connection (fills the user data specified via options, the connection is then completed with default parameters by NM) - and then activates the connection on a Wi-Fi device. WPA-Enterprise is not supported as it requires a plethora of parameters and they can't be obtained automatically. Also, the created connection uses 'auto' IP method, which means that if the Wi-Fi network doesn't support DHCP, the connection will albeit be created, however the activation will fail (IP configuration won't be available).
2012-04-28 22:32:21 +01:00
return nmc->return_value;
}
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
static GBytes *
generate_ssid_for_hotspot (const char *ssid)
{
GBytes *ssid_bytes;
char *hotspot_ssid = NULL;
if (!ssid) {
hotspot_ssid = g_strdup_printf ("Hotspot-%s", g_get_host_name ());
if (strlen (hotspot_ssid) > 32)
hotspot_ssid[32] = '\0';
ssid = hotspot_ssid;
}
ssid_bytes = g_bytes_new (ssid, strlen (ssid));
g_free (hotspot_ssid);
return ssid_bytes;
}
#define WPA_PASSKEY_SIZE 8
static void
generate_wpa_key (char *key, size_t len)
{
guint i;
g_return_if_fail (key);
g_return_if_fail (len > WPA_PASSKEY_SIZE);
/* generate a 8-chars ASCII WPA key */
for (i = 0; i < WPA_PASSKEY_SIZE; i++) {
int c;
c = g_random_int_range (33, 126);
/* too many non alphanumeric characters are hard to remember for humans */
while (!g_ascii_isalnum (c))
c = g_random_int_range (33, 126);
key[i] = (gchar) c;
}
key[WPA_PASSKEY_SIZE] = '\0';
}
static void
generate_wep_key (char *key, size_t len)
{
int i;
const char *hexdigits = "0123456789abcdef";
g_return_if_fail (key);
g_return_if_fail (len > 10);
/* generate a 10-digit hex WEP key */
for (i = 0; i < 10; i++) {
int digit;
digit = g_random_int_range (0, 16);
key[i] = hexdigits[digit];
}
key[10] = '\0';
}
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
static gboolean
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
set_wireless_security_for_hotspot (NMSettingWirelessSecurity *s_wsec,
const char *wifi_mode,
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
NMDeviceWifiCapabilities caps,
const char *password,
cli: add '--show-password' option for 'nmcli device wifi hotspot' It is useful to show nmcli-generated hotspot password (if a user does not provide his own password). Without the option the user would have to look into the generated profile in order to find out the password.
2015-10-14 12:05:32 +02:00
gboolean show_password,
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
GError **error)
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
{
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
char generated_key[11];
const char *key;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
const char *key_mgmt;
if (g_strcmp0 (wifi_mode, NM_SETTING_WIRELESS_MODE_AP) == 0) {
if (caps & NM_WIFI_DEVICE_CAP_RSN) {
nm_setting_wireless_security_add_proto (s_wsec, "rsn");
nm_setting_wireless_security_add_pairwise (s_wsec, "ccmp");
nm_setting_wireless_security_add_group (s_wsec, "ccmp");
key_mgmt = "wpa-psk";
} else if (caps & NM_WIFI_DEVICE_CAP_WPA) {
nm_setting_wireless_security_add_proto (s_wsec, "wpa");
nm_setting_wireless_security_add_pairwise (s_wsec, "tkip");
nm_setting_wireless_security_add_group (s_wsec, "tkip");
key_mgmt = "wpa-psk";
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
} else
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
key_mgmt = "none";
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
} else
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
key_mgmt = "none";
if (g_strcmp0 (key_mgmt, "wpa-psk") == 0) {
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
/* use WPA */
if (password) {
if (!nm_utils_wpa_psk_valid (password)) {
g_set_error (error, NMCLI_ERROR, 0, _("'%s' is not valid WPA PSK"), password);
return FALSE;
}
key = password;
} else {
generate_wpa_key (generated_key, sizeof (generated_key));
key = generated_key;
}
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
g_object_set (s_wsec,
NM_SETTING_WIRELESS_SECURITY_KEY_MGMT, key_mgmt,
NM_SETTING_WIRELESS_SECURITY_PSK, key,
NULL);
} else {
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
/* use WEP */
if (password) {
if (!nm_utils_wep_key_valid (password, NM_WEP_KEY_TYPE_KEY)) {
g_set_error (error, NMCLI_ERROR, 0,
_("'%s' is not valid WEP key (it should be 5 or 13 ASCII chars)"),
password);
return FALSE;
}
key = password;
} else {
generate_wep_key (generated_key, sizeof (generated_key));
key = generated_key;
}
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
g_object_set (s_wsec,
NM_SETTING_WIRELESS_SECURITY_KEY_MGMT, key_mgmt,
NM_SETTING_WIRELESS_SECURITY_WEP_KEY0, key,
NM_SETTING_WIRELESS_SECURITY_WEP_KEY_TYPE, NM_WEP_KEY_TYPE_KEY,
NULL);
}
cli: add '--show-password' option for 'nmcli device wifi hotspot' It is useful to show nmcli-generated hotspot password (if a user does not provide his own password). Without the option the user would have to look into the generated profile in order to find out the password.
2015-10-14 12:05:32 +02:00
if (show_password)
g_print (_("Hotspot password: %s\n"), key);
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
return TRUE;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
static NMCResultCode
do_device_wifi_hotspot (NmCli *nmc, int argc, char **argv)
{
AddAndActivateInfo *info;
const char *ifname = NULL;
const char *con_name = NULL;
char *default_name = NULL;
const char *ssid = NULL;
const char *wifi_mode;
const char *band = NULL;
const char *channel = NULL;
unsigned long channel_int;
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
const char *password = NULL;
cli: add '--show-password' option for 'nmcli device wifi hotspot' It is useful to show nmcli-generated hotspot password (if a user does not provide his own password). Without the option the user would have to look into the generated profile in order to find out the password.
2015-10-14 12:05:32 +02:00
gboolean show_password = FALSE;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
NMDevice *device = NULL;
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
gs_free NMDevice **devices = NULL;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
NMDeviceWifiCapabilities caps;
NMConnection *connection = NULL;
NMSettingConnection *s_con;
NMSettingWireless *s_wifi;
NMSettingWirelessSecurity *s_wsec;
NMSettingIPConfig *s_ip4, *s_ip6;
nmcli: Add Proxy support to nmcli nmcli fixed to allow users edit proxy properties.
2016-08-16 05:56:00 +05:30
NMSettingProxy *s_proxy;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
GBytes *ssid_bytes;
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
GError *error = NULL;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
/* Set default timeout waiting for operation completion. */
if (nmc->timeout == -1)
nmc->timeout = 60;
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
devices = nmc_get_devices_sorted (nmc->client);
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
while (argc > 0) {
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
if (argc == 1 && nmc->complete) {
nmc_complete_strings (*argv, "ifname", "con-name", "ssid", "band",
"channel", "password", NULL);
}
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
if (strcmp (*argv, "ifname") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
ifname = *argv;
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
if (argc == 1 && nmc->complete)
complete_device (devices, ifname, TRUE);
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
} else if (strcmp (*argv, "con-name") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
con_name = *argv;
} else if (strcmp (*argv, "ssid") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
ssid = *argv;
if (strlen (ssid) > 32) {
g_string_printf (nmc->return_text, _("Error: ssid is too long."));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
} else if (strcmp (*argv, "band") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
band = *argv;
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_strings (band, "a", "bg", NULL);
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
if (strcmp (band, "a") && strcmp (band, "bg")) {
g_string_printf (nmc->return_text, _("Error: band argument value '%s' is invalid; use 'a' or 'bg'."),
band);
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
} else if (strcmp (*argv, "channel") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
channel = *argv;
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
} else if (strcmp (*argv, "password") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
}
password = *argv;
cli: add global --show-secrets option - it controls echoing passwords input on terminal - it replaces --show-secrets in 'nmcli connection show', which is deprecated now - it replaces --show-password in 'nmcli device wifi hotspot', which is deprecated now
2015-12-04 13:29:42 +01:00
/* --show-password is deprecated in favour of global --show-secrets option */
/* Keep it here for backwards compatibility */
cli: add '--show-password' option for 'nmcli device wifi hotspot' It is useful to show nmcli-generated hotspot password (if a user does not provide his own password). Without the option the user would have to look into the generated profile in order to find out the password.
2015-10-14 12:05:32 +02:00
} else if (nmc_arg_is_option (*argv, "show-password")) {
show_password = TRUE;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
} else {
g_string_printf (nmc->return_text, _("Error: Unknown parameter %s."), *argv);
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
cli: move show_secret field to NmcConfig The show-secret property is basically a part of the current configuration, relevant during printing. It will be passed on to nmc_print(), and hence must be part of NmcConfig.
2017-04-06 15:26:23 +02:00
show_password = nmc->nmc_config.show_secrets || show_password;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
if (nmc->complete)
return nmc->return_value;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
/* Verify band and channel parameters */
if (!channel) {
if (g_strcmp0 (band, "bg") == 0)
channel = "1";
if (g_strcmp0 (band, "a") == 0)
channel = "7";
}
if (channel) {
if (!band) {
g_string_printf (nmc->return_text, _("Error: channel requires band too."));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
if ( !nmc_string_to_uint (channel, TRUE, 1, 5825, &channel_int)
|| !nm_utils_wifi_is_channel_valid (channel_int, band)) {
g_string_printf (nmc->return_text, _("Error: channel '%s' not valid for band '%s'."),
channel, band);
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
}
/* Find Wi-Fi device. When no ifname is provided, the first Wi-Fi is used. */
cli/device: make devices_idx argument to find_wifi_device_by_iface() optional Most callers don't use it.
2016-07-12 12:33:31 +02:00
device = find_wifi_device_by_iface (devices, ifname, NULL);
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
if (!device) {
if (ifname)
g_string_printf (nmc->return_text, _("Error: Device '%s' is not a Wi-Fi device."), ifname);
else
g_string_printf (nmc->return_text, _("Error: No Wi-Fi device found."));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_UNKNOWN;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
/* Check device supported mode */
caps = nm_device_wifi_get_capabilities (NM_DEVICE_WIFI (device));
if (caps & NM_WIFI_DEVICE_CAP_AP)
wifi_mode = NM_SETTING_WIRELESS_MODE_AP;
else if (caps & NM_WIFI_DEVICE_CAP_ADHOC)
wifi_mode = NM_SETTING_WIRELESS_MODE_ADHOC;
else {
g_string_printf (nmc->return_text, _("Error: Device '%s' supports neither AP nor Ad-Hoc mode."),
nm_device_get_iface (device));
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_UNKNOWN;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
}
/* Create a connection with appropriate parameters */
connection = nm_simple_connection_new ();
s_con = (NMSettingConnection *) nm_setting_connection_new ();
nm_connection_add_setting (connection, NM_SETTING (s_con));
if (!con_name)
con_name = default_name = nmc_unique_connection_name (nm_client_get_connections (nmc->client), "Hotspot");
g_object_set (s_con,
NM_SETTING_CONNECTION_ID, con_name,
NM_SETTING_CONNECTION_AUTOCONNECT, FALSE,
NULL);
g_free (default_name);
s_wifi = (NMSettingWireless *) nm_setting_wireless_new ();
nm_connection_add_setting (connection, NM_SETTING (s_wifi));
ssid_bytes = generate_ssid_for_hotspot (ssid);
g_object_set (s_wifi, NM_SETTING_WIRELESS_MODE, wifi_mode,
NM_SETTING_WIRELESS_SSID, ssid_bytes,
NULL);
g_bytes_unref (ssid_bytes);
if (channel)
g_object_set (s_wifi,
NM_SETTING_WIRELESS_CHANNEL, (guint32) channel_int,
NM_SETTING_WIRELESS_BAND, band,
NULL);
s_wsec = (NMSettingWirelessSecurity *) nm_setting_wireless_security_new ();
nm_connection_add_setting (connection, NM_SETTING (s_wsec));
cli: add '--show-password' option for 'nmcli device wifi hotspot' It is useful to show nmcli-generated hotspot password (if a user does not provide his own password). Without the option the user would have to look into the generated profile in order to find out the password.
2015-10-14 12:05:32 +02:00
if (!set_wireless_security_for_hotspot (s_wsec, wifi_mode, caps, password, show_password, &error)) {
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
g_object_unref (connection);
g_string_printf (nmc->return_text, _("Error: Invalid 'password': %s."), error->message);
g_clear_error (&error);
cli/device: add completion to wifi hotspot subcommand
2016-07-12 11:37:18 +02:00
return NMC_RESULT_ERROR_UNKNOWN;
cli: add 'password' option for 'nmcli device wifi hotspot' It allows user provided password to be used to secure the hotspot. Otherwise, nmcli will generate a suitable password.
2015-10-13 17:48:06 +02:00
}
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
s_ip4 = (NMSettingIPConfig *) nm_setting_ip4_config_new ();
nm_connection_add_setting (connection, NM_SETTING (s_ip4));
g_object_set (s_ip4, NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP4_CONFIG_METHOD_SHARED, NULL);
s_ip6 = (NMSettingIPConfig *) nm_setting_ip6_config_new ();
nm_connection_add_setting (connection, NM_SETTING (s_ip6));
g_object_set (s_ip6, NM_SETTING_IP_CONFIG_METHOD, NM_SETTING_IP6_CONFIG_METHOD_IGNORE, NULL);
nmcli: Add Proxy support to nmcli nmcli fixed to allow users edit proxy properties.
2016-08-16 05:56:00 +05:30
s_proxy = (NMSettingProxy *) nm_setting_proxy_new ();
nm_connection_add_setting (connection, NM_SETTING (s_proxy));
proxy: cleanup handling of proxy-method The numeric value of NM_SETTING_PROXY_METHOD_NONE should be zero, as that is the more natural default. Also, cast all uses of the enum values in g_object_set() to (int).
2016-10-04 16:32:09 +02:00
g_object_set (s_proxy, NM_SETTING_PROXY_METHOD, (int) NM_SETTING_PROXY_METHOD_NONE, NULL);
nmcli: Add Proxy support to nmcli nmcli fixed to allow users edit proxy properties.
2016-08-16 05:56:00 +05:30
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
/* Activate the connection now */
nmc->nowait_flag = (nmc->timeout == 0);
cli: use should_wait consistently It's a semaphore, not a boolean.
2016-06-22 17:44:39 +02:00
nmc->should_wait++;
cli: add 'nmcli device wifi hotspot' command Synopsis: nmcli device wifi hotspot [ifname <ifname>] [con-name <name>] [ssid <SSID>] [band a|bg] [channel <channel>]
2015-10-02 22:46:53 +02:00
info = g_malloc0 (sizeof (AddAndActivateInfo));
info->nmc = nmc;
info->device = device;
info->hotspot = TRUE;
nm_client_add_and_activate_connection_async (nmc->client,
connection,
device,
NULL,
NULL,
add_and_activate_cb,
info);
return nmc->return_value;
}
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
static void
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
request_rescan_cb (GObject *object, GAsyncResult *result, gpointer user_data)
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
{
NmCli *nmc = (NmCli *) user_data;
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
GError *error = NULL;
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
nm_device_wifi_request_scan_finish (NM_DEVICE_WIFI (object), result, &error);
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
if (error) {
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
g_string_printf (nmc->return_text, _("Error: %s."), error->message);
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
nmc->return_value = NMC_RESULT_ERROR_UNKNOWN;
libnm: make sync/async APIs more GLib-like Make synchronous APIs take GCancellables, and make asynchronous APIs use GAsyncReadyCallbacks and have names ending in "_async", with "_finish" functions to retrieve the results. Also, make nm_client_activate_connection_finish(), nm_client_add_and_activate_finish(), and nm_remote_settings_add_connection_finish() be (transfer full) rather than (transfer none), because the refcounting semantics become slightly confusing in some edge cases otherwise.
2014-09-11 16:27:13 -04:00
g_error_free (error);
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
}
quit ();
}
static NMCResultCode
cli: rename 'nmcli dev wifi scan' to 'nmcli dev wifi rescan' (bgo #698241)
2013-04-22 18:39:08 +02:00
do_device_wifi_rescan (NmCli *nmc, int argc, char **argv)
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
{
NMDevice *device;
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
const char *ifname = NULL;
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
GPtrArray *ssids;
cli/device: make find_wifi_device_by_iface use sorted device array ...as opposed to GPtrArray. The point is to be able to have a single device list that could be later reused with complete_device().
2016-07-12 14:45:36 +02:00
gs_free NMDevice **devices = NULL;
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
GVariantBuilder builder, array_builder;
GVariant *options;
const char *ssid;
int i;
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
ssids = g_ptr_array_new ();
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
devices = nmc_get_devices_sorted (nmc->client);
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
/* Get the parameters */
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
while (argc > 0) {
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_strings (*argv, "ifname", "ssid", NULL);
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
if (strcmp (*argv, "ifname") == 0) {
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
if (ifname) {
g_string_printf (nmc->return_text, _("Error: '%s' cannot repeat."), *(argv-1));
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
goto finish;
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
}
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
goto finish;
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
}
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
ifname = *argv;
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
if (argc == 1 && nmc->complete)
complete_device (devices, ifname, TRUE);
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
} else if (strcmp (*argv, "ssid") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
nmc->return_value = NMC_RESULT_ERROR_USER_INPUT;
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
goto finish;
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
}
g_ptr_array_add (ssids, *argv);
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
} else if (!nmc->complete)
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
g_printerr (_("Unknown parameter: %s\n"), *argv);
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
}
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
if (nmc->complete)
goto finish;
cli: rename 'iface' --> 'ifname' in all commands
2013-04-18 14:54:01 +02:00
/* Find Wi-Fi device to scan on. When no ifname is provided, the first Wi-Fi is used. */
cli/device: make devices_idx argument to find_wifi_device_by_iface() optional Most callers don't use it.
2016-07-12 12:33:31 +02:00
device = find_wifi_device_by_iface (devices, ifname, NULL);
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
if (!device) {
cli: fix an bug when reporting an error $ nmcli dev wifi rescan ifname eteee (process:21211): libnm-CRITICAL **: nm_device_get_device_type: assertion 'NM_IS_DEVICE (self)' failed Error: Device 'eteee' is not a Wi-Fi device. Fixes: 823df334eda48a8e2cec897d6123f7b2158c12ff
2015-08-13 09:48:12 +02:00
if (ifname)
g_string_printf (nmc->return_text, _("Error: Device '%s' is not a Wi-Fi device."), ifname);
else
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
g_string_printf (nmc->return_text, _("Error: No Wi-Fi device found."));
nmc->return_value = NMC_RESULT_ERROR_UNKNOWN;
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
goto finish;
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
}
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
if (ssids->len) {
g_variant_builder_init (&builder, G_VARIANT_TYPE_VARDICT);
g_variant_builder_init (&array_builder, G_VARIANT_TYPE ("aay"));
for (i = 0; i < ssids->len; i++) {
ssid = g_ptr_array_index (ssids, i);
g_variant_builder_add (&array_builder, "@ay",
g_variant_new_fixed_array (G_VARIANT_TYPE_BYTE, ssid, strlen (ssid), 1));
}
g_variant_builder_add (&builder, "{sv}", "ssids", g_variant_builder_end (&array_builder));
options = g_variant_builder_end (&builder);
nm_device_wifi_request_scan_options_async (NM_DEVICE_WIFI (device), options,
NULL, request_rescan_cb, nmc);
} else
nm_device_wifi_request_scan_async (NM_DEVICE_WIFI (device),
NULL, request_rescan_cb, nmc);
cli: turn should_wait into a semaphore Count the exit blockers. We'll want to terminate the device monitor in case all monitored devices vanish.
2015-03-27 12:39:17 +01:00
nmc->should_wait++;
cli/device: add completion to wifi rescan subcommand
2016-07-12 10:14:53 +02:00
finish:
cli: add 'ssid' parameter for 'nmcli device wifi rescan' 'ssid' can repeat when more SSIDs should be scanned, e.g. $ nmcli dev wifi rescan ssid "hidden cafe" ssid AP12 ssid "my home Wi-Fi" Bash completion fixed by thaller@redhat.com
2015-07-14 09:50:56 +02:00
g_ptr_array_free (ssids, FALSE);
cli: add 'nmcli device wifi scan' command This requests NetworkManager to perform scanning for Wi-Fi networks. NetworkManager scans periodically, but this command can be used by impatient users who don't want to wait for the next scan. Direct D-Bus API call example: gdbus call --system -d org.freedesktop.NetworkManager \ -o /org/freedesktop/NetworkManager/Devices/2 \ -m org.freedesktop.NetworkManager.Device.Wireless.RequestScan {}
2012-11-19 17:10:56 +01:00
return nmc->return_value;
}
cli/device: convert wifi subcommand to nmc_do_cmd()
2016-06-24 19:03:33 +02:00
static NMCCommand device_wifi_cmds[] = {
cli: use nmc_do_cmd to get the client and check if the daemon is running The makes use of asynchronous client initialization, making things a bit faster and reduces code duplication too.
2016-08-31 20:52:48 +02:00
{ "list", do_device_wifi_list, NULL, TRUE, TRUE },
{ "connect", do_device_wifi_connect_network, NULL, TRUE, TRUE },
{ "hotspot", do_device_wifi_hotspot, NULL, TRUE, TRUE },
{ "rescan", do_device_wifi_rescan, NULL, TRUE, TRUE },
{ NULL, do_device_wifi_list, NULL, TRUE, TRUE },
cli/device: convert wifi subcommand to nmc_do_cmd()
2016-06-24 19:03:33 +02:00
};
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
static NMCResultCode
do_device_wifi (NmCli *nmc, int argc, char **argv)
{
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: make it possible to call sub-commands with client obtained asynchronously
2016-08-31 21:04:33 +02:00
nmc_do_cmd (nmc, device_wifi_cmds, *argv, argc, argv);
return nmc->return_value;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
static int
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
show_device_lldp_list (NMDevice *device, NmCli *nmc, const char *fields_str, int *counter)
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
{
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const NMMetaAbstractInfo *const*tmpl;
NmcOutputField *arr;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
GPtrArray *neighbors;
const char *str;
int i;
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
NMC_OUTPUT_DATA_DEFINE_SCOPED (out);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
gs_free char *header_name = NULL;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
neighbors = nm_device_get_lldp_neighbors (device);
if (!neighbors || !neighbors->len)
return 0;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
tmpl = (const NMMetaAbstractInfo *const*) nmc_fields_dev_lldp_list;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
/* Main header name */
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
header_name = construct_header_name (_("Device LLDP neighbors"),
nm_device_get_iface (device));
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
out_indices = parse_output_fields (fields_str, (const NMMetaAbstractInfo *const*) nmc_fields_dev_lldp_list, FALSE, NULL, NULL);
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_MAIN_HEADER_ADD | NMC_OF_FLAG_FIELD_NAMES);
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
for (i = 0; i < neighbors->len; i++) {
NMLldpNeighbor *neighbor = neighbors->pdata[i];
guint value;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
arr = nmc_dup_fields_array (tmpl, NMC_OF_FLAG_SECTION_PREFIX);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
set_val_str (arr, 0, g_strdup_printf ("NEIGHBOR[%d]", (*counter)++));
set_val_strc (arr, 1, nm_device_get_iface (device));
if (nm_lldp_neighbor_get_attr_string_value (neighbor, NM_LLDP_ATTR_CHASSIS_ID, &str))
set_val_strc (arr, 2, str);
if (nm_lldp_neighbor_get_attr_string_value (neighbor, NM_LLDP_ATTR_PORT_ID, &str))
set_val_strc (arr, 3, str);
if (nm_lldp_neighbor_get_attr_string_value (neighbor, NM_LLDP_ATTR_PORT_DESCRIPTION, &str))
set_val_strc (arr, 4, str);
if (nm_lldp_neighbor_get_attr_string_value (neighbor, NM_LLDP_ATTR_SYSTEM_NAME, &str))
set_val_strc (arr, 5, str);
if (nm_lldp_neighbor_get_attr_string_value (neighbor, NM_LLDP_ATTR_SYSTEM_DESCRIPTION, &str))
set_val_strc (arr, 6, str);
if (nm_lldp_neighbor_get_attr_uint_value (neighbor, NM_LLDP_ATTR_SYSTEM_CAPABILITIES, &value))
set_val_str (arr, 7, g_strdup_printf ("%u (%s)", value, nmc_parse_lldp_capabilities (value)));
if (nm_lldp_neighbor_get_attr_uint_value (neighbor, NM_LLDP_ATTR_IEEE_802_1_PVID, &value))
set_val_str (arr, 8, g_strdup_printf ("%u", value));
if (nm_lldp_neighbor_get_attr_uint_value (neighbor, NM_LLDP_ATTR_IEEE_802_1_PPVID, &value))
set_val_str (arr, 9, g_strdup_printf ("%u", value));
if (nm_lldp_neighbor_get_attr_uint_value (neighbor, NM_LLDP_ATTR_IEEE_802_1_PPVID_FLAGS, &value))
set_val_str (arr, 10, g_strdup_printf ("%u", value));
if (nm_lldp_neighbor_get_attr_uint_value (neighbor, NM_LLDP_ATTR_IEEE_802_1_VID, &value))
set_val_str (arr, 11, g_strdup_printf ("%u", value));
if (nm_lldp_neighbor_get_attr_string_value (neighbor, NM_LLDP_ATTR_IEEE_802_1_VLAN_NAME, &str))
set_val_strc (arr, 12, str);
if (nm_lldp_neighbor_get_attr_string_value (neighbor, NM_LLDP_ATTR_DESTINATION, &str))
set_val_strc (arr, 13, str);
if (nm_lldp_neighbor_get_attr_uint_value (neighbor, NM_LLDP_ATTR_CHASSIS_ID_TYPE, &value))
set_val_strc (arr, 14, g_strdup_printf ("%u", value));
if (nm_lldp_neighbor_get_attr_uint_value (neighbor, NM_LLDP_ATTR_PORT_ID_TYPE, &value))
set_val_strc (arr, 15, g_strdup_printf ("%u", value));
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
g_ptr_array_add (out.output_data, arr);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
}
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
print_data_prepare_width (out.output_data);
cli: pass arguments for print_data separately of NmcOutputData Don't pass on large structs of input arguments. It only convolutes which arguments are passed, and where they come from.
2017-03-31 13:21:47 +02:00
print_data (&nmc->nmc_config, out_indices, header_name, 0, &out);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
return neighbors->len;
}
cli/device: convert lldp subcommand to nmc_do_cmd()
2016-06-24 19:03:11 +02:00
static NMCResultCode
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
do_device_lldp_list (NmCli *nmc, int argc, char **argv)
{
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
NMDevice *device = NULL;
gs_free_error GError *error = NULL;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
const char *fields_str = NULL;
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
int counter = 0;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
gs_unref_array GArray *out_indices = NULL;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
while (argc > 0) {
cli/device: add "lldp list" subcommand completion
2016-06-24 19:03:14 +02:00
if (argc == 1 && nmc->complete)
nmc_complete_strings (*argv, "ifname", NULL);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
if (strcmp (*argv, "ifname") == 0) {
nmcli: fix nmcli parsing Fixes the parsing for correct and incorrect commands. e.g.: nmcli connection down id <conn> nmcli -t ... Fixes: 16902a2be66866aa541bb56ccb98c765ac21d1a2
2017-03-29 12:02:14 +02:00
argc--;
argv++;
cli: use next_arg() on argument boundary Shift argc and argc manually between argument ant its value and use next_arg() between arguments everywhere. Whill be useful to parse global arguments.
2017-03-24 14:00:25 +01:00
if (!argc) {
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
g_string_printf (nmc->return_text, _("Error: %s argument is missing."), *(argv-1));
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
}
device = get_device (nmc, &argc, &argv, &error);
if (!device) {
g_string_printf (nmc->return_text, _("Error: %s."), error->message);
return error->code;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
}
} else {
cli: remove some dead code Coverity points out.
2016-02-12 15:43:19 +01:00
g_string_printf (nmc->return_text, _("Error: invalid extra argument '%s'."), *argv);
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
}
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
}
if (!nmc->required_fields || strcasecmp (nmc->required_fields, "common") == 0)
fields_str = NMC_FIELDS_DEV_LLDP_LIST_COMMON;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
else if (!nmc->required_fields || strcasecmp (nmc->required_fields, "all") == 0) {
} else
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
fields_str = nmc->required_fields;
cli: split tracking of meta data out of NmcOutputField When generating output data, nmcli iterates over a list of property-descriptors (nmc_fields_ip4_config), creates an intermediate array (output_data) and finally prints it. However, previously both the meta data (nmc_fields_ip4_config) and the intermediate format use the same type NmcOutputField. This means, certain fields are relevant to describe a property, and other fields are output/formatting fields. Split this up. Now, the meta data is tracked in form of an NMMetaAbstractInfo lists. This separates the information about properties from intermediate steps during creation of the output. Note that currently functions like print_ip4_config() still have all the knowledge about how to generate the output. That is wrong, instead, the meta data (NMMetaAbstractInfo) should describe how to create the output and then all those functions could be replaced. This means, later we want to add more knowledge to the NMMetaAbstractInfo, so it is important to keep them separate from NmcOutputField.
2017-03-31 19:18:16 +02:00
out_indices = parse_output_fields (fields_str, (const NMMetaAbstractInfo *const*) nmc_fields_dev_lldp_list, FALSE, NULL, &error);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
if (error) {
g_string_printf (nmc->return_text, _("Error: 'device lldp list': %s"), error->message);
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
return NMC_RESULT_ERROR_USER_INPUT;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
}
cli/device: add "lldp list" subcommand completion
2016-06-24 19:03:14 +02:00
if (nmc->complete)
return nmc->return_value;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
if (device) {
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
show_device_lldp_list (device, nmc, fields_str, &counter);
} else {
cli/devices: export device_state_to_color() Also, get rid of the ColorInfo structure. It was not too nice anyway.
2016-07-01 21:45:43 +02:00
NMDevice **devices = nmc_get_devices_sorted (nmc->client);
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
int i;
cli: don't track output data in global NmCli structure We should not violate the global data to track the output data while it is constructed and printed. Most of the time, we actually clear the output data anyway -- either before constructing it, or after printing it. In some cases we didn't, but I think that is a bug. It's really hard to keep track of this. The output data should belong to a certain scope and get destroyed afterwards. Passing it around is very confusing. Don't do that.
2017-03-30 14:56:19 +02:00
for (i = 0; devices[i]; i++)
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
show_device_lldp_list (devices[i], nmc, fields_str, &counter);
cli/device: make "lldp list" subcommand use get_device()
2016-06-24 19:22:36 +02:00
g_free (devices);
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
}
return nmc->return_value;
}
cli/device: convert lldp subcommand to nmc_do_cmd()
2016-06-24 19:03:11 +02:00
static NMCCommand device_lldp_cmds[] = {
cli: use nmc_do_cmd to get the client and check if the daemon is running The makes use of asynchronous client initialization, making things a bit faster and reduces code duplication too.
2016-08-31 20:52:48 +02:00
{ "list", do_device_lldp_list, NULL, TRUE, TRUE },
{ NULL, do_device_lldp_list, NULL, TRUE, TRUE },
cli/device: convert lldp subcommand to nmc_do_cmd()
2016-06-24 19:03:11 +02:00
};
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
static NMCResultCode
do_device_lldp (NmCli *nmc, int argc, char **argv)
{
cli/device: use nmc_do_cmd()
2016-06-23 12:19:10 +02:00
if (!nmc->mode_specified)
cli: pass configuration options separately from NmCli structure The NmCli structure is passed around everywhere and contains all the state of the program. It is very hard to follow which parts are used where. Split out more configuration options to a NmcConfig field. This field is mostly immutable, and modified at particular places that now can be easily found.
2017-03-30 12:45:41 +02:00
nmc->nmc_config_mutable.multiline_output = TRUE; /* multiline mode is default for 'device lldp' */
cli/device: use nmc_do_cmd()
2016-06-23 12:19:10 +02:00
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: make it possible to call sub-commands with client obtained asynchronously
2016-08-31 21:04:33 +02:00
nmc_do_cmd (nmc, device_lldp_cmds, *argv, argc, argv);
return nmc->return_value;
cli: add command for displaying LLDP neighbors The list of LLDP neighbors is available through the D-Bus interface and libnm already provides functions to retrieve it; make the list available through nmcli as well. Sample output: $ nmcli device lldp NEIGHBOR[0].DEVICE: eth0 NEIGHBOR[0].CHASSIS-ID: 00:13:21:58:CA:42 NEIGHBOR[0].PORT-ID: 1 NEIGHBOR[0].PORT-DESCRIPTION: 1 NEIGHBOR[0].SYSTEM-NAME: ProCurve Switch 2600-8-PWR NEIGHBOR[0].SYSTEM-DESCRIPTION: ProCurve J8762A Switch 2600-8-PWR, revision H.08.89 NEIGHBOR[0].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) NEIGHBOR[1].DEVICE: eth2 NEIGHBOR[1].CHASSIS-ID: 00:01:30:F8:AD:A2 NEIGHBOR[1].PORT-ID: 1/1 NEIGHBOR[1].PORT-DESCRIPTION: Summit300-48-Port 1001 NEIGHBOR[1].SYSTEM-NAME: Summit300-48 NEIGHBOR[1].SYSTEM-DESCRIPTION: Summit300-48 - Version 7.4e.1 (Build 5) NEIGHBOR[1].SYSTEM-CAPABILITIES: 20 (mac-bridge,router) https://bugzilla.gnome.org/show_bug.cgi?id=757307
2015-11-10 14:06:02 +01:00
}
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
static gboolean
is_single_word (const char* line)
{
size_t n1, n2, n3;
n1 = strspn (line, " \t");
n2 = strcspn (line+n1, " \t\0") + n1;
n3 = strspn (line+n2, " \t");
if (n3 == 0)
return TRUE;
else
return FALSE;
}
cli: use readline also in other commands (besides interactive editor) It is used for '--ask' option. The main use is for questionnaires (nmcli -a con add).
2014-05-14 09:22:50 +02:00
static char **
cli: switch to the new-style readline typedefs (bgo #731294) The old Function, CPFunction, CPPFunction and VFunction typedefs are considered deprecated starting from readline 4.2 (though were not marked as such): http://lists.gnu.org/archive/html/bug-bash/2000-10/msg00078.html Compatibility typedefs have been in place up to readline 6.2 but were removed with the 6.3 release thus causing build to break. http://lists.gnu.org/archive/html/bug-readline/2014-04/msg00018.html https://bugzilla.gnome.org/show_bug.cgi?id=731294
2014-06-06 11:22:24 +02:00
nmcli_device_tab_completion (const char *text, int start, int end)
cli: use readline also in other commands (besides interactive editor) It is used for '--ask' option. The main use is for questionnaires (nmcli -a con add).
2014-05-14 09:22:50 +02:00
{
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
char **match_array = NULL;
cli: switch to the new-style readline typedefs (bgo #731294) The old Function, CPFunction, CPPFunction and VFunction typedefs are considered deprecated starting from readline 4.2 (though were not marked as such): http://lists.gnu.org/archive/html/bug-bash/2000-10/msg00078.html Compatibility typedefs have been in place up to readline 6.2 but were removed with the 6.3 release thus causing build to break. http://lists.gnu.org/archive/html/bug-readline/2014-04/msg00018.html https://bugzilla.gnome.org/show_bug.cgi?id=731294
2014-06-06 11:22:24 +02:00
rl_compentry_func_t *generator_func = NULL;
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
cli: use readline also in other commands (besides interactive editor) It is used for '--ask' option. The main use is for questionnaires (nmcli -a con add).
2014-05-14 09:22:50 +02:00
/* Disable readline's default filename completion */
rl_attempted_completion_over = 1;
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
if (g_strcmp0 (rl_prompt, PROMPT_INTERFACE) == 0) {
/* Disable appending space after completion */
rl_completion_append_character = '\0';
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
if (!is_single_word (rl_line_buffer))
return NULL;
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
cli: (trivial) move gen_func_ifnames() from devices.c to common.c and rename it to nmc_rl_gen_func()
2016-09-15 01:20:53 +02:00
generator_func = nmc_rl_gen_func_ifnames;
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
} else if (g_strcmp0 (rl_prompt, PROMPT_INTERFACES) == 0) {
cli: (trivial) move gen_func_ifnames() from devices.c to common.c and rename it to nmc_rl_gen_func()
2016-09-15 01:20:53 +02:00
generator_func = nmc_rl_gen_func_ifnames;
cli: allow multiple devices for 'nmcli device disconnect/delete' (bgo #746097) Allow disconnecting and deleting multiple interfaces at a time. It is much more user friendly. TAB and bash completions are supported as well. https://bugzilla.gnome.org/show_bug.cgi?id=746097
2015-03-11 19:10:58 +01:00
}
cli: implement TAB completion for connection and interface names For commands like nmcli -a dev disconnect nmcli -a con up ...
2014-05-15 12:52:26 +02:00
if (generator_func)
match_array = rl_completion_matches (text, generator_func);
return match_array;
cli: use readline also in other commands (besides interactive editor) It is used for '--ask' option. The main use is for questionnaires (nmcli -a con add).
2014-05-14 09:22:50 +02:00
}
cli/device: use nmc_do_cmd()
2016-06-23 12:19:10 +02:00
static const NMCCommand device_cmds[] = {
cli: use nmc_do_cmd to get the client and check if the daemon is running The makes use of asynchronous client initialization, making things a bit faster and reduces code duplication too.
2016-08-31 20:52:48 +02:00
{ "status", do_devices_status, usage_device_status, TRUE, TRUE },
{ "show", do_device_show, usage_device_show, TRUE, TRUE },
{ "connect", do_device_connect, usage_device_connect, TRUE, TRUE },
{ "reapply", do_device_reapply, usage_device_reapply, TRUE, TRUE },
{ "disconnect", do_devices_disconnect, usage_device_disconnect, TRUE, TRUE },
{ "delete", do_devices_delete, usage_device_delete, TRUE, TRUE },
{ "set", do_device_set, usage_device_set, TRUE, TRUE },
{ "monitor", do_devices_monitor, usage_device_monitor, TRUE, TRUE },
{ "wifi", do_device_wifi, usage_device_wifi, FALSE, FALSE },
{ "lldp", do_device_lldp, usage_device_lldp, FALSE, FALSE },
{ "modify", do_device_modify, usage_device_modify, TRUE, TRUE },
{ NULL, do_devices_status, usage, TRUE, TRUE },
cli/device: use nmc_do_cmd()
2016-06-23 12:19:10 +02:00
};
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
NMCResultCode
do_devices (NmCli *nmc, int argc, char **argv)
{
nmcli: allow cmd specific option parsing in next_arg() func Options dependant on specific commands (e.g., nmcli connection show --active) are now allowed to be processed by the next_arg() function. This would allow autocompletion to expand options belonging to specific command first, and then global ones. Note that global options ("--ask" and "--show-secrets") will be auto-completed everywhere but only if at least a '-' is passed. Command specific ones (--temporary, --active, --order) will be auto-completed only after the command they belongs to but without requiring the user to pass a heading '-'. Example: 'nmcli connection show -a' will expand '-a' into '--active', but 'nmcli connection add -a` will expand '-a' into '--ask' (as it is a global option) This commit fixes also autocompletion for: nmcli connection modify --temporary
2017-03-30 16:09:46 +02:00
next_arg (nmc, &argc, &argv, NULL);
cli: add 'nmcli agent' command (bgo #739568) Synopsis: nmcli agent { secret | polkit | all } The command runs separate NetworkManager secret agent or session polkit agent, or both. It is useful when - no other secret agent is available (such as GUI nm-applet, gnome-shell, KDE applet) - no other polkit agent is available (such as polkit-gnome-authentication-agent-1, polkit-kde-authentication-agent-1 or lxpolkit) https://bugzilla.gnome.org/show_bug.cgi?id=739568
2014-10-30 15:45:43 +01:00
/* Register polkit agent */
nmc_start_polkit_agent_start_try (nmc);
cli: switch to the new-style readline typedefs (bgo #731294) The old Function, CPFunction, CPPFunction and VFunction typedefs are considered deprecated starting from readline 4.2 (though were not marked as such): http://lists.gnu.org/archive/html/bug-bash/2000-10/msg00078.html Compatibility typedefs have been in place up to readline 6.2 but were removed with the 6.3 release thus causing build to break. http://lists.gnu.org/archive/html/bug-readline/2014-04/msg00018.html https://bugzilla.gnome.org/show_bug.cgi?id=731294
2014-06-06 11:22:24 +02:00
rl_attempted_completion_function = (rl_completion_func_t *) nmcli_device_tab_completion;
cli: use readline also in other commands (besides interactive editor) It is used for '--ask' option. The main use is for questionnaires (nmcli -a con add).
2014-05-14 09:22:50 +02:00
cli: make it possible to call sub-commands with client obtained asynchronously
2016-08-31 21:04:33 +02:00
nmc_do_cmd (nmc, device_cmds, *argv, argc, argv);
return nmc->return_value;
cli: add initial pieces of nmcli
2010-02-25 09:52:30 -08:00
}
cli: add nmcli monitor https://bugzilla.redhat.com/show_bug.cgi?id=1034158
2015-03-27 13:07:43 +01:00
void
monitor_devices (NmCli *nmc)
{
cli/trivial: rename some functions for consistency Some functions that take device lists use plural form in name.
2016-06-23 18:49:39 +02:00
do_devices_monitor (nmc, 0, NULL);
cli: add nmcli monitor https://bugzilla.redhat.com/show_bug.cgi?id=1034158
2015-03-27 13:07:43 +01:00
}
Reference in a new issue Copy permalink