summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* libnma/vpn-password-dialog: add nma_vpn_password_dialog_run_and_print()lr/external-auth-dialogLubomir Rintel2018-07-264-4/+219
| | | | | | | | | | It should avoid code duplication in VPN plugins that implement external authentication while already utilizing the NMAVpnPasswordDialog. The auth-dialogs shall just do _set_password_*key() and then call nma_vpn_password_dialog_run_and_print() in place of nma_vpn_password_dialog_run_and_block().
* libnma/vpn-password-dialog: use actual types for widgets bound from a templateLubomir Rintel2018-07-261-40/+40
| | | | | | | | It is mostly not useful to have the typed as GtkWidget and cast it to their real types when needed, becasue the layout work where GtkWidgets are convenient is mostly done by GtkBuilder. This is what other templated widget classes do as well.
* libnma-docs: fix the URLLubomir Rintel2018-07-261-1/+1
|
* gitlab: build libnma site documentationLubomir Rintel2018-07-262-2/+29
| | | | https://gitlab.gnome.org/GNOME/network-manager-applet/merge_requests/14
* merge: branch 'lr/fixes'Lubomir Rintel2018-07-2656-262/+544
|\ | | | | | | https://gitlab.gnome.org/GNOME/network-manager-applet/merge_requests/11
| * editor/trivial: move utils_fake_return_key() to editorLubomir Rintel2018-07-2610-18/+73
| | | | | | | | | | It relies on Gdk trickery that will not be possible in future. Has no place in libnma.
| * glade: replace GtkHBox and GtkVBox with GtkBoxLubomir Rintel2018-07-2629-86/+172
| | | | | | | | Both long deprecated.
| * glade: get rid of GtkContainer.border-widthLubomir Rintel2018-07-2632-50/+200
| | | | | | | | | | | | | | | | | | | | | | | | | | It's gone from Gtk4. We mostly use it to pad out the pages in the editor and on a couple other places (where the use is probably incorrect). The migration manual suggests that a margin of a child should be used instead. This indeed works for us. However, we need to set four properties from each side, which is not nice. An alternative would be to use a CSS class, butthere's no sensible class in the default theme for the job and it's perhaps not worth adding a custom style sheet for us.
| * glade: drop use of GtkCheckButton.xalignLubomir Rintel2018-07-265-15/+14
| | | | | | | | It's long deprecated and gone from Gtk4.
| * glade: get rid of the GtkMisc.ypad propertyLubomir Rintel2018-07-261-1/+1
| | | | | | | | It's long deprecated and gone from Gtk4.
| * glade: drop GtkImage.yalign and GtkLabel.yalign usesLubomir Rintel2018-07-263-4/+3
| | | | | | | | Valign does the same thing while not being deprecated and dropped in Gtk4.
| * glade: use margin_start/margin_end instead of left/rightLubomir Rintel2018-07-266-27/+27
| | | | | | | | | | | | | | | | | | The latter is deprecated in favor of the former. The difference in behavior is that they have opposite meanings in RTL scripts, which is probaby good (well, at least for the labels). Notably, this future-proofs us for Gtk4 that drops the deprecated properties.
| * libnma/wifi-dialog: fix assertion failure on changing a deviceLubomir Rintel2018-07-261-5/+2
| | | | | | | | | | connection_combo_init() is also called when the device combo is changed. Don't assume the connection is NULL; it is not going to be.
| * libnma/wifi-dialog: don't set GtkContainer.border-widthLubomir Rintel2018-07-262-2/+4
| | | | | | | | | | The property will be gone in Gtk4. Instead set marigins on the child widgets.
| * libnma/wifi-dialog: properly escape the SSIDLubomir Rintel2018-07-261-3/+3
| | | | | | | | The SSIDs with [<>] in it would mess up with the markup.
| * libnma: drop use of GtkWidget.no-show-all propertyLubomir Rintel2018-07-263-19/+8
| | | | | | | | | | It's going away in Gtk4. We don't do show_all() anyway -- just hide the widgets by default.
| * utils: drop an useless use of gtk_widget_show_all()Lubomir Rintel2018-07-261-1/+1
| |
| * wireless-security: avoid passing NULL to nma_cert_chooser_set_*_password()Lubomir Rintel2018-07-262-3/+8
| | | | | | | | | | | | This never worked for the file picker, since it tripped an assertion on setting NULL to the GtkEntry. Don't do that and guard the PKCS#11 picker too.
| * all: make the labelling style consistentLubomir Rintel2018-07-264-28/+28
|/ | | | | | | There were some ommissions in the commits below: Fixes: 004ac2a23e185712b7d827528bf8e7b4c260ce53 Fixes: e1a0b880ad3484a70ef2a9832984e0b3d592e448
* password-dialog: preserve mnemonicLubomir Rintel2018-07-261-4/+3
| | | | | | | | | | The password labels as specified in the .ui file have "use_underline" property set. However it gets reset upon gtk_label_set_text(). Let's preserve it. This fixes an ugliness with VPNC plugin and perhaps more. https://gitlab.gnome.org/GNOME/network-manager-applet/merge_requests/13
* libnma/cert-chooser: add some gtk-doc annotationsLubomir Rintel2018-07-261-4/+4
|
* bluetooth: fix an assert failure on creation cancellationlr/hold-importLubomir Rintel2018-07-241-1/+2
| | | | | | $ nm-connection-editor --create --type bluetooth <ESC> GLib-GObject-CRITICAL: g_object_unref: assertion 'G_IS_OBJECT (object)' failed
* connection-list: attempt a VPN import firstLubomir Rintel2018-07-241-25/+44
| | | | | | | | | | | On nm-connection-editor --import, nm_connection_list_create() is called with an empty ctype (G_TYPE_INVALID). This is not good -- on a successful import the first known type (which happens to be Bluetooth) would match. Let's treat this as "any connection type" -- attempt a VPN import opportunistically and if it succeeds, continute with a ctype of NM_TYPE_SETTING_VPN. Otherwise just bail out.
* connection-editor: hold GApplication while the import dialog is runningLubomir Rintel2018-07-245-46/+90
| | | | | | | | | | This is done from new callbacks of nm_connection_list_create() and nm_connection_list_add(). In order for this to work, import_vpn_from_file_cb() had to be fixed to always invoke its callback. In turn, vpn_connection_from_file() had to be changed to pass its error result up to its callers instead of presenting a potentially redundant error message itself.
* build: set -Wall before we start probing the extra warning optionsLubomir Rintel2018-07-241-1/+1
| | | | | | Some warnings depends on others: -Wformat-security won't work without -Wformat. With -Wall we're confident enough that we have important warnings enabled and in any case we're going to enable it anyway.
* build: enable more warnings on autogenLubomir Rintel2018-07-241-1/+1
| | | | This is the same as NetworkManager does. Allows us to catch bugs earlier.
* docs: update copyright yearlr/import-holdLubomir Rintel2018-07-231-1/+2
|
* Update Turkish translationMuhammet Kara2018-07-201-999/+1691
|
* Update Turkish translationMuhammet Kara2018-07-201-895/+717
|
* merge: branch 'lr/vpn-add-idle'Lubomir Rintel2018-07-131-49/+10
|\ | | | | | | https://gitlab.gnome.org/GNOME/network-manager-applet/merge_requests/12
| * connection-editor: don't defer creation of vpn connection to idlelr/vpn-add-idleLubomir Rintel2018-07-121-49/+9
| | | | | | | | | | | | | | | | | | | | | | Now that GtkApplication makes good decisions about when to quit (when no windows are present), the magic that deferred creation of vpn type selector is useless. Worse even, it makes GtkApplication believe that we're done and shuts it down before an editor window is added to it. https://gitlab.gnome.org/GNOME/network-manager-applet/issues/9 Reported-by: Jan Tojnar
| * connection-editor: ignore --help in main() option contextLubomir Rintel2018-07-121-0/+1
|/ | | | | | | It's solely used for parsing --keep-above and doesn't know anything about actual options and thus prints irrelevant help. Skip that and let the GtkApplication::command-line signal handler parse it and print help instead.
* Update Brazilian Portuguese translationRafael Fontenelle2018-07-051-407/+338
|
* Update German translationMario Blättermann2018-07-051-270/+297
|
* c-e: fix leak in update_relabel_list_filename() (clear_name_if_present())Thomas Haller2018-06-261-1/+1
|
* release: bump version to 1.8.15 (development)1.8.15-devLubomir Rintel2018-06-252-2/+2
|
* release: bump version to 1.8.141.8.14Lubomir Rintel2018-06-252-2/+2
|
* release: update NEWSLubomir Rintel2018-06-251-0/+15
|
* build: appindicator and xembed are not mutually exclusiveLubomir Rintel2018-06-252-2/+2
|
* build: install the appdata to a non-legacy locationLubomir Rintel2018-06-252-2/+2
| | | | | | | | The current AppStream specification prefers /usr/share/metainfo eveyone who relies on this (that's distros, not the end users) already supports the new location. https://gitlab.gnome.org/GNOME/network-manager-applet/issues/2
* Update Catalan translationJordi Mas2018-06-241-796/+878
|
* editor: add label reference to "Interface name" in Team tabFrancesco Giudici2018-06-211-1/+5
| | | | https://bugzilla.redhat.com/show_bug.cgi?id=1390638
* merge: branch 'lr/app-windows'Lubomir Rintel2018-06-1932-324/+203
|\ | | | | | | https://gitlab.gnome.org/GNOME/network-manager-applet/merge_requests/5
| * editor: fix --keep-above parsinglr/app-windowsLubomir Rintel2018-06-191-5/+2
| | | | | | | | | | | | The other options should be ignored when we look for --keep-above. g_option_context_parse() removes it from argv[], thus we can also revert 88e4dcab27e6.
| * editor/connection-list: drop editing-done signalLubomir Rintel2018-06-193-23/+7
| | | | | | | | It's of no use anymore now that GtkApplication decides when to quit.
| * editor: register editor windows as application windowsLubomir Rintel2018-06-191-25/+29
| | | | | | | | | | | | Tracking all windows as application windows makes it possible for the GtkApplication to terminate at the correct point -- when no windows are present.
| * editor/connection-list: add new-editor signalLubomir Rintel2018-06-192-4/+31
| | | | | | | | | | This indicates that a new editor window has been opened, for the purposes of tracking the windows as GtkApplication windows.
| * editor/connection-editor: propagate the new-editor signalLubomir Rintel2018-06-192-1/+19
| |
| * editor/connection-editor: drop signal slotsLubomir Rintel2018-06-192-5/+1
| | | | | | | | They're offensive.
| * editor/connection-editor: use defines for signal namesLubomir Rintel2018-06-195-6/+8
| | | | | | | | Brings joy to Thomas while grepping through code.
View Lorry Controller Status