gtk4.git
12 years agoUpdated Belarusian translation.
Ihar Hrachyshka [Thu, 12 Sep 2013 17:14:11 +0000 (20:14 +0300)]
Updated Belarusian translation.

12 years agoUpdated Hungarian translation
Gabor Kelemen [Thu, 12 Sep 2013 15:26:48 +0000 (17:26 +0200)]
Updated Hungarian translation

12 years agoUpdated Latvian translation
Rūdolfs Mazurs [Thu, 12 Sep 2013 11:01:16 +0000 (14:01 +0300)]
Updated Latvian translation

12 years agoa11y: Remove default description from GtkSwitch
Alejandro Piñeiro [Wed, 11 Sep 2013 16:37:26 +0000 (18:37 +0200)]
a11y: Remove default description from GtkSwitch

https://bugzilla.gnome.org/show_bug.cgi?id=707926

12 years agoUpdated Russian translation
Yuri Myasoedov [Thu, 12 Sep 2013 09:08:28 +0000 (13:08 +0400)]
Updated Russian translation

12 years agoUpdated slovak translation
Pavol Klačanský [Wed, 11 Sep 2013 20:26:37 +0000 (22:26 +0200)]
Updated slovak translation

12 years agoUpdated Slovenian translation
Matej Urbančič [Wed, 11 Sep 2013 19:11:15 +0000 (21:11 +0200)]
Updated Slovenian translation

12 years agowayland: Fix make dist
Rico Tzschichholz [Wed, 11 Sep 2013 15:20:55 +0000 (17:20 +0200)]
wayland: Fix make dist

In addition to d34335e51c98efebe010bd4a202fe2a4b2fb7d06

12 years agoUpdated Thai translation.
Theppitak Karoonboonyanan [Wed, 11 Sep 2013 10:35:45 +0000 (17:35 +0700)]
Updated Thai translation.

12 years agol10n: Update Japanese translation
Jiro Matsuzawa [Wed, 11 Sep 2013 10:09:39 +0000 (19:09 +0900)]
l10n: Update Japanese translation

12 years agoUpdated Thai translation.
Theppitak Karoonboonyanan [Wed, 11 Sep 2013 09:47:21 +0000 (16:47 +0700)]
Updated Thai translation.

12 years agoFinnish translation update
Timo Jyrinki [Wed, 11 Sep 2013 07:04:12 +0000 (10:04 +0300)]
Finnish translation update

12 years agoFinnish translation update
Timo Jyrinki [Wed, 11 Sep 2013 06:50:30 +0000 (09:50 +0300)]
Finnish translation update

12 years agoUpdated Czech translation
Marek Černocký [Wed, 11 Sep 2013 06:22:27 +0000 (08:22 +0200)]
Updated Czech translation

12 years agoa11y: not using default name for switch
Alejandro Piñeiro [Tue, 10 Sep 2013 17:29:56 +0000 (19:29 +0200)]
a11y: not using default name for switch

A proper name or label is required. In the same way, if the
widget is labelled, ATs uses to expose both the label and the
name, making the final output not really user-friendly.

https://bugzilla.gnome.org/show_bug.cgi?id=707872

12 years agoMake logos in GtkAboutDialog bigger
Matthias Clasen [Wed, 11 Sep 2013 02:45:06 +0000 (22:45 -0400)]
Make logos in GtkAboutDialog bigger

When available, choose a size of up to 128 in GtkAboutDialog.
A 48x48 icon just looks too small in there.

12 years agoAdd missing introspection annotations
Phil Clayton [Thu, 5 Sep 2013 00:08:41 +0000 (01:08 +0100)]
Add missing introspection annotations

https://bugzilla.gnome.org/show_bug.cgi?id=707519

12 years agolistbox: Set activate_signal on listboxrow class
Alexander Larsson [Tue, 10 Sep 2013 12:12:07 +0000 (14:12 +0200)]
listbox: Set activate_signal on listboxrow class

keyboard navigation didn't support activation since we moved
the keynav to the child row widgets. We fix this by adding a
activate signal handler for the row and setting
widget_class->activate_signal to it.

https://bugzilla.gnome.org/show_bug.cgi?id=707778

12 years agogdk: event: add get_window() method
Lionel Landwerlin [Tue, 10 Sep 2013 10:20:28 +0000 (11:20 +0100)]
gdk: event: add get_window() method

https://bugzilla.gnome.org/show_bug.cgi?id=707844

12 years agoMSVC Builds: Fix gengir Projects
Chun-wei Fan [Tue, 10 Sep 2013 07:51:23 +0000 (15:51 +0800)]
MSVC Builds: Fix gengir Projects

Like the install projects that were fixed few days ago, the gengir projects
did not have info on the intermediate and output directories as a result of
the split up of the property sheets.  Fix this by including the appropriate
property sheet in the gtk-gengir property sheet so that we can avoid
confusing messages from Visual Studio on whether to reload the gengir
project as it was modified, at least on 2008.

12 years agoUpdated Slovenian translation
Matej Urbančič [Mon, 9 Sep 2013 19:00:35 +0000 (21:00 +0200)]
Updated Slovenian translation

12 years agoUpdated Slovenian translation
Matej Urbančič [Mon, 9 Sep 2013 18:36:26 +0000 (20:36 +0200)]
Updated Slovenian translation

12 years agoPunjabi Translation updated by Aman
A S Alam [Mon, 9 Sep 2013 18:27:33 +0000 (13:27 -0500)]
Punjabi Translation updated by Aman

12 years agoRevert "pango: Handle case where pango_layout_get_text() returns NULL"
Benjamin Otte [Mon, 9 Sep 2013 15:19:28 +0000 (17:19 +0200)]
Revert "pango: Handle case where pango_layout_get_text() returns NULL"

This reverts commit 70ac2b24c3d400c854d8a9e0937afa076f9dbe7f.

It turns out the correct fix is to make pango_layout_get_text() not
return NULL. This has been done, so we can drop this patch.

I won't bump the Pango dependency in configure.ac for this as I don't
consider the crash critical enough.

12 years agopango: Handle case where pango_layout_get_text() returns NULL
Benjamin Otte [Mon, 9 Sep 2013 14:50:55 +0000 (16:50 +0200)]
pango: Handle case where pango_layout_get_text() returns NULL

Fixes crashers with accessibility

https://bugzilla.gnome.org/show_bug.cgi?id=707659

12 years agoPixel cache: Add no-pixel-cache debug flag
Alexander Larsson [Mon, 9 Sep 2013 09:07:50 +0000 (11:07 +0200)]
Pixel cache: Add no-pixel-cache debug flag

This is useful for benchmarking comparisons, and to verify if bugs
happend with or without the pixel cache.

12 years agotextview: use pixelcache rendered area to inform invalidation region.
Christian Hergert [Fri, 6 Sep 2013 22:52:33 +0000 (15:52 -0700)]
textview: use pixelcache rendered area to inform invalidation region.

Use the pixelcache rendered area to inform what part of the cache should
be invalidated upon changes to the underlying textlayout.

By rendering the background to the pixelcache, we can avoid the need to
use RGBA content.

Also, we're using the pixel cache on the text windows bin_window (see
gtk_text_view_get_window) so we need to register the invalidation handler
on that, otherwise the region passed to the invalidate handler will get
clipped to the visible region.

https://bugzilla.gnome.org/show_bug.cgi?id=707244

12 years agoAdd a test for icons in menu models
Matthias Clasen [Mon, 9 Sep 2013 02:37:50 +0000 (22:37 -0400)]
Add a test for icons in menu models

testmenubutton now features an icon.

12 years agoAdd a testcase for custom levels in GtkLevelBar
Matthias Clasen [Sun, 8 Sep 2013 20:39:49 +0000 (16:39 -0400)]
Add a testcase for custom levels in GtkLevelBar

This is related to https://bugzilla.gnome.org/show_bug.cgi?id=707695

12 years agoUpdated Norwegian bokmål translation
Kjartan Maraas [Sun, 8 Sep 2013 17:40:51 +0000 (19:40 +0200)]
Updated Norwegian bokmål translation

12 years agoTajik translation updated
Victor Ibragimov [Sun, 8 Sep 2013 12:21:16 +0000 (17:21 +0500)]
Tajik translation updated

12 years agoTajik translation updated
Victor Ibragimov [Sun, 8 Sep 2013 12:18:11 +0000 (17:18 +0500)]
Tajik translation updated

12 years agoTajik translation updated
Victor Ibragimov [Sun, 8 Sep 2013 12:06:57 +0000 (17:06 +0500)]
Tajik translation updated

12 years agoUpdated Galician translations
Fran Diéguez [Sat, 7 Sep 2013 16:51:24 +0000 (18:51 +0200)]
Updated Galician translations

12 years agopixelcache: add getter for extra width and height.
Christian Hergert [Fri, 6 Sep 2013 22:49:50 +0000 (15:49 -0700)]
pixelcache: add getter for extra width and height.

12 years agotextview: only set invalidate handler for text window.
Christian Hergert [Fri, 6 Sep 2013 20:58:37 +0000 (13:58 -0700)]
textview: only set invalidate handler for text window.

12 years agoreftests: I can't count in CSS
Benjamin Otte [Fri, 6 Sep 2013 22:23:38 +0000 (00:23 +0200)]
reftests: I can't count in CSS

nth-child() is 1-indexed, not 0-indexed.

It doesn't matter for this test really, but better do it right to not confuse
poor developers who wonder why the first image is highlighted when nth-child(0)
clearly states "none".

12 years agoreftests: Check that -gtk-image-effect is inherited
Benjamin Otte [Fri, 6 Sep 2013 22:21:20 +0000 (00:21 +0200)]
reftests: Check that -gtk-image-effect is inherited

12 years agogdk_pixbuf_get_from_source: only reuse surfaces with matching format
Claudio Saavedra [Fri, 6 Sep 2013 15:44:57 +0000 (18:44 +0300)]
gdk_pixbuf_get_from_source: only reuse surfaces with matching format

Otherwise the conversion functions will return rubbish.

https://bugzilla.gnome.org/show_bug.cgi?id=707445

12 years agoFix config dir <> data dir confusion
Matthias Clasen [Fri, 6 Sep 2013 13:42:01 +0000 (09:42 -0400)]
Fix config dir <> data dir confusion

We added code to look for settings.ini in system config dirs,
and then proceeded to move it to /usr/share/gtk-3.0 :-(. So,
look in that location as well.

12 years agoUpdated Traditional Chinese translation(Hong Kong and Taiwan)
Chao-Hsiung Liao [Fri, 6 Sep 2013 08:26:34 +0000 (16:26 +0800)]
Updated Traditional Chinese translation(Hong Kong and Taiwan)

12 years agoUpdated Brazilian Portuguese translationproofread by Enrico Nicoletto
Rafael Ferreira [Fri, 6 Sep 2013 01:28:48 +0000 (22:28 -0300)]
Updated Brazilian Portuguese translationproofread by Enrico Nicoletto

12 years agoreftests: Add test for just fixed Polari crasher
Benjamin Otte [Fri, 6 Sep 2013 01:04:44 +0000 (03:04 +0200)]
reftests: Add test for just fixed Polari crasher

12 years agocss: Change currentColor handling
Benjamin Otte [Fri, 6 Sep 2013 01:01:37 +0000 (03:01 +0200)]
css: Change currentColor handling

We now pass NULL when the current color should be the default value of
the "color" property and we haven't looked up any value yet. This way we
don't need to look it up all the time and more importantly we can
resolve the default color, which is required because it's a
GtkCssColorValue and not a GtkCssRgbaValue.

Fixes assertions triggering at Polari startup.

12 years agocssimage: Remove unused variables
Benjamin Otte [Fri, 6 Sep 2013 01:01:10 +0000 (03:01 +0200)]
cssimage: Remove unused variables

12 years agoDon't use deprecated GSimpleActionGroup api
Matthias Clasen [Thu, 5 Sep 2013 22:50:23 +0000 (18:50 -0400)]
Don't use deprecated GSimpleActionGroup api

Instead, use the corresponding GActionMap api.

12 years agoDon't call deprecated combobox api
Matthias Clasen [Thu, 5 Sep 2013 22:45:16 +0000 (18:45 -0400)]
Don't call deprecated combobox api

gtk_combo_box_set_title is not needed anymore, since we
dropped the tearoff support.

12 years ago[gdkcairo] Fix color premultiplication
Behdad Esfahbod [Thu, 5 Sep 2013 22:42:45 +0000 (18:42 -0400)]
[gdkcairo] Fix color premultiplication

https://bugzilla.gnome.org/show_bug.cgi?id=513812

12 years agoSync with changes in the cairo device-scale handling
Alexander Larsson [Thu, 5 Sep 2013 17:30:23 +0000 (19:30 +0200)]
Sync with changes in the cairo device-scale handling

The version of device scale that landed in upstream cairo
master already inherits the device scale in cairo_create_similar,
so no need to do that in gtk anymore.

12 years agoreftests: Add a reftest for the cell renderer fix
Benjamin Otte [Thu, 5 Sep 2013 16:47:29 +0000 (18:47 +0200)]
reftests: Add a reftest for the cell renderer fix

The reftest is a bit flaky because it compares cell renderers with
GtkImage and therefor an icon view with a GtkBox, but it's the best I
can come up with.

https://bugzilla.gnome.org/show_bug.cgi?id=702423

12 years agocellrendererpixbuf: Don't eat states
Benjamin Otte [Thu, 5 Sep 2013 16:45:13 +0000 (18:45 +0200)]
cellrendererpixbuf: Don't eat states

Even when not following states, blacklist the states we don't follow
instead of whitelisting the ones we do. This way, we don't accidentally
eat new ones like the text direction flags.

Fixes a bug where text direction wasn't available when rendering stock
items.

https://bugzilla.gnome.org/show_bug.cgi?id=702423

12 years agocellrendererpixbuf: Follow state even when insensitive
Benjamin Otte [Thu, 5 Sep 2013 16:42:53 +0000 (18:42 +0200)]
cellrendererpixbuf: Follow state even when insensitive

12 years agocellrendererpixbuf: Always follow state
Benjamin Otte [Thu, 5 Sep 2013 14:26:46 +0000 (16:26 +0200)]
cellrendererpixbuf: Always follow state

... not just in PRELIT | SELECTED.

12 years agoUpdated Polish translation
Piotr Drąg [Thu, 5 Sep 2013 12:40:56 +0000 (14:40 +0200)]
Updated Polish translation

12 years agoUpdated Spanish translation
Daniel Mustieles [Thu, 5 Sep 2013 09:17:52 +0000 (11:17 +0200)]
Updated Spanish translation

12 years ago[l10n] Updated Italian translation.
Milo Casagrande [Thu, 5 Sep 2013 08:41:57 +0000 (10:41 +0200)]
[l10n] Updated Italian translation.

12 years agoVisual C++ Builds: Fix "install" Projects
Chun-wei Fan [Thu, 5 Sep 2013 06:33:22 +0000 (14:33 +0800)]
Visual C++ Builds: Fix "install" Projects

Due to the split up of the property sheets, the install projects did not
have info on the Intermediate and Output Paths, which caused confusing
messages from Visual Studio to show up upon completing build+"install" and
closing Visual Studio on whether to reload the install project, at least on
Visual Studio 2008.

Also clean up the Visual Studio 2008 install project a bit.

Include the property sheet which defines these properties to fix this.

12 years agoTrivial formatting fixes
Matthias Clasen [Thu, 5 Sep 2013 00:59:56 +0000 (20:59 -0400)]
Trivial formatting fixes

12 years agoFix a typo in an error message
Jiro Matsuzawa [Wed, 4 Sep 2013 16:39:16 +0000 (01:39 +0900)]
Fix a typo in an error message

https://bugzilla.gnome.org/show_bug.cgi?id=707488

12 years agoGtkListBox: ref_sink new object in set_header and set_adjustment
Alexander Larsson [Wed, 4 Sep 2013 19:00:23 +0000 (21:00 +0200)]
GtkListBox: ref_sink new object in set_header and set_adjustment

These are taking ownership of the object in the hierarchy, so should
sink any floating refs to make it easy to use in C code.

12 years agoUpdated slovak translation
Pavol Klačanský [Wed, 4 Sep 2013 19:00:20 +0000 (21:00 +0200)]
Updated slovak translation

12 years agogdkwindow-x11: Multiply opaque region rects by the window scale
Jasper St. Pierre [Wed, 4 Sep 2013 18:45:01 +0000 (14:45 -0400)]
gdkwindow-x11: Multiply opaque region rects by the window scale

This fixes rendering issues under HiDPI

12 years agoUpdate French translation
Alexandre Franke [Wed, 4 Sep 2013 14:44:55 +0000 (16:44 +0200)]
Update French translation

12 years agogdkframeclockidle: Add explicit fallthrough markers
Jasper St. Pierre [Tue, 3 Sep 2013 20:58:12 +0000 (16:58 -0400)]
gdkframeclockidle: Add explicit fallthrough markers

So people reading the code don't get confused.

12 years agoaccessible: Fix test runner
Jasper St. Pierre [Tue, 3 Sep 2013 20:51:03 +0000 (16:51 -0400)]
accessible: Fix test runner

This was an obvious typo.

12 years agogtkwidget: Fix issues with baseline alignment comparison
Jasper St. Pierre [Tue, 3 Sep 2013 20:47:21 +0000 (16:47 -0400)]
gtkwidget: Fix issues with baseline alignment comparison

12 years agobuild: Fix srcdir != builddir with wayland-client-protocol.h
Colin Walters [Tue, 3 Sep 2013 21:48:30 +0000 (17:48 -0400)]
build: Fix srcdir != builddir with wayland-client-protocol.h

In the gnome-ostree model builddir contains all generated files not in
git (unless the build system explicitly overrides that).  Here the
wayland-client-protocol.h was in $(builddir)/wayland, so we need to
find it using our already extant -I$(top_builddir)/gdk, rather than
relying on same-directory lookup.

12 years agoUpdated Korean translation
Changwoo Ryu [Tue, 3 Sep 2013 16:54:03 +0000 (01:54 +0900)]
Updated Korean translation

12 years agowayland: set the wm_class on toplevel windows
Giovanni Campagna [Tue, 3 Sep 2013 09:38:35 +0000 (11:38 +0200)]
wayland: set the wm_class on toplevel windows

Before mapping the window, set the title and class, to allow
application tracking by gnome-shell.

https://bugzilla.gnome.org/show_bug.cgi?id=707129

12 years agogtk-shell: extend the protocol with shell capabilities
Giovanni Campagna [Tue, 3 Sep 2013 07:54:53 +0000 (09:54 +0200)]
gtk-shell: extend the protocol with shell capabilities

Add the concept of shell capabilities, which allow the compositor
to advertise support for the app menu and the global menubar,
which are then propagated as GdkSettings.

https://bugzilla.gnome.org/show_bug.cgi?id=707129

12 years agowayland: restore support for the application menu
Giovanni Campagna [Fri, 30 Aug 2013 11:56:45 +0000 (13:56 +0200)]
wayland: restore support for the application menu

If the compositor supports the gtk-shell interface, use it to
export the application ID, dbus name and paths that can be used
for the application menu.

https://bugzilla.gnome.org/show_bug.cgi?id=707129

12 years agowayland: add support for a private gtk-shell protocol
Giovanni Campagna [Fri, 30 Aug 2013 11:55:37 +0000 (13:55 +0200)]
wayland: add support for a private gtk-shell protocol

This protocol will be used by mutter-wayland and gtk to replace
the _GTK X11 properties for DBus names/paths.

https://bugzilla.gnome.org/show_bug.cgi?id=707129

12 years agoUpdated Spanish translation
Daniel Mustieles [Tue, 3 Sep 2013 14:28:10 +0000 (16:28 +0200)]
Updated Spanish translation

12 years agoUpdated Spanish translation
Daniel Mustieles [Tue, 3 Sep 2013 14:21:15 +0000 (16:21 +0200)]
Updated Spanish translation

12 years agoBump version
Matthias Clasen [Mon, 2 Sep 2013 18:51:01 +0000 (14:51 -0400)]
Bump version

12 years ago3.9.14
Matthias Clasen [Mon, 2 Sep 2013 17:45:59 +0000 (13:45 -0400)]
3.9.14

12 years agoBump glib version requirement - GPropertyAction is used on examples
Gustavo Noronha Silva [Sun, 1 Sep 2013 00:12:09 +0000 (21:12 -0300)]
Bump glib version requirement - GPropertyAction is used on examples

https://bugzilla.gnome.org/show_bug.cgi?id=707201

12 years agoAdded Scottish Gaelic translation
GunChleoc [Sun, 1 Sep 2013 15:00:31 +0000 (09:00 -0600)]
Added Scottish Gaelic translation

12 years agoAdded Scottish Gaelic translation
GunChleoc [Sun, 1 Sep 2013 14:58:11 +0000 (08:58 -0600)]
Added Scottish Gaelic translation

12 years agotextview: Use guint since we specify ": 1"
Paolo Borelli [Sun, 1 Sep 2013 09:43:05 +0000 (11:43 +0200)]
textview: Use guint since we specify ": 1"

12 years agoa11y: Check whether a widget is mapped before querying its parent
Mike Gorse [Fri, 30 Aug 2013 14:13:39 +0000 (09:13 -0500)]
a11y: Check whether a widget is mapped before querying its parent

Call gtk_widget_get_mapped() in a couple of places before looking at the
widget's parent, since it might be set to a widget that has been
finalized, causing an invalid read.

12 years agopathbar: fix positioning of the down slider button
Cosimo Cecchi [Sun, 25 Aug 2013 03:17:09 +0000 (20:17 -0700)]
pathbar: fix positioning of the down slider button

So that it's always linked with the rest of the pathbar.

https://bugzilla.gnome.org/show_bug.cgi?id=706722

12 years agopathbar: remove unused spacing private member
Cosimo Cecchi [Sun, 25 Aug 2013 02:55:23 +0000 (19:55 -0700)]
pathbar: remove unused spacing private member

It's always zero.

https://bugzilla.gnome.org/show_bug.cgi?id=706722

12 years agoGtkPathBar: Don't make current dir bold
Matthias Clasen [Sat, 31 Aug 2013 21:20:50 +0000 (17:20 -0400)]
GtkPathBar: Don't make current dir bold

This is closer to the nautilus path bar.

https://bugzilla.gnome.org/show_bug.cgi?id=706451

12 years agoUpdated Polish translation
Piotr Drąg [Sat, 31 Aug 2013 21:23:32 +0000 (23:23 +0200)]
Updated Polish translation

12 years agoFix up a11y tests again
Matthias Clasen [Sat, 31 Aug 2013 21:06:44 +0000 (17:06 -0400)]
Fix up a11y tests again

This change was caused by the recently reverted patch.

12 years agoRevert "a11y: Check whether a widget is mapped before querying its parent"
Matthias Clasen [Sat, 31 Aug 2013 20:16:26 +0000 (16:16 -0400)]
Revert "a11y: Check whether a widget is mapped before querying its parent"

This reverts commit 7e3db6fdd36960fca0ae0e7e078b448b5f8b220a.

It broke the testsuite.

12 years agogtkwindow: Don't set an opaque region if the window is app-paintable
Jasper St. Pierre [Sat, 31 Aug 2013 19:52:56 +0000 (15:52 -0400)]
gtkwindow: Don't set an opaque region if the window is app-paintable

This means the background is effectively unused. This fixes bad drawing
errors with cc-rr-labeller and potentially also notify-osd.

12 years agowin32: use the input locale to decide the default input method
Aleksander Morgado [Sat, 31 Aug 2013 17:58:12 +0000 (13:58 -0400)]
win32: use the input locale to decide the default input method

GTK+ tries to automatically assign the best input module based on the
'system locale'. In the specific case of the IME input method, it will
be the default for the whole GTK+ application if the system locale is
either Japanese (ja), Korean (ko) or Chinese (zh). Other defaults are
equally applicable, e.g. if system locale is Catalan (ca), the special
'Cedilla' input module is chosen.

System locale can be changed (e.g. Win7) through the following sequence
(reboot required):
  Control Panel
    Region and Language
      Administrative
        Language for non-Unicode Programs
          Change system locale...

The problem with this behaviour is that changing the 'default input
language' (e.g. from English to Japanese+IME) doesn't affect the GTK+
application. Therefore, I can have an English system locale (where GTK+
will choose Simple IM by default) but then have Japanese+IME as input
language.

Default input language can be changed (e.g. Win7) through the following
sequence (no reboot required):
  System locale can be changed (e.g. Win7) through:
    Control Panel
      Region and Language
        Keyboards and Languages
          Keyboards and other input languages
            Change keyboards...

Default input language can also be changed using the language bar directly.

So, instead of using the system-wide default locale to decide which input
method to use as default, better use the input language specified by the
user, which may be the same as the system-wide default locale, or different.
Following the previous example, with an English system locale and a
Japanese+IME input language, the default input method will now be IME
instead of Simple, which is closer to what's expected by the user.

This change only affects the application during startup; i.e. if the user
changes the input language while the application is running, we wouldn't be
changing the default input method to use. We could do this processing the
WM_INPUTLANGCHANGE messages, though.

https://bugzilla.gnome.org/show_bug.cgi?id=700428

12 years agogdkwindow-x11: Correct math for filling in the _NET_WM_OPAQUE_REGION property
Jasper St. Pierre [Sat, 31 Aug 2013 17:21:09 +0000 (13:21 -0400)]
gdkwindow-x11: Correct math for filling in the _NET_WM_OPAQUE_REGION property

This fixes incorrect rendering under mutter and other WMs that implement
_NET_WM_OPAQUE_REGION.

12 years agoGdkWaylandWindow: Fix memory leak
Pavel Vasin [Wed, 21 Aug 2013 11:09:29 +0000 (15:09 +0400)]
GdkWaylandWindow: Fix memory leak

https://bugzilla.gnome.org/show_bug.cgi?id=706493

12 years agoGtkClipboardWayland: Fix memory leak
Pavel Vasin [Wed, 21 Aug 2013 10:59:45 +0000 (14:59 +0400)]
GtkClipboardWayland: Fix memory leak

https://bugzilla.gnome.org/show_bug.cgi?id=706493

12 years agocss shadow: Fix memory leak
Pavel Vasin [Wed, 21 Aug 2013 10:56:21 +0000 (14:56 +0400)]
css shadow: Fix memory leak

https://bugzilla.gnome.org/show_bug.cgi?id=706493

12 years agoprintsettings: Add missing docs for output constants
Arnel A. Borja [Mon, 26 Aug 2013 09:52:59 +0000 (17:52 +0800)]
printsettings: Add missing docs for output constants

GTK_PRINT_SETTINGS_OUTPUT_DIR and GTK_PRINT_SETTINGS_OUTPUT_BASENAME
doesn't have documentation.

https://bugzilla.gnome.org/show_bug.cgi?id=706802

12 years agoFix an age-old typo
Matthias Clasen [Sat, 31 Aug 2013 16:23:54 +0000 (12:23 -0400)]
Fix an age-old typo

GtkIconView was not properly propagating key release events,
since forever.

https://bugzilla.gnome.org/show_bug.cgi?id=706740

12 years agoUpdate Kazakh translation
Baurzhan Muftakhidinov [Sat, 31 Aug 2013 11:15:18 +0000 (17:15 +0600)]
Update Kazakh translation

12 years agoRevert "Deprecate and ignore gtk-entry-password-hint-timeout"
Matthias Clasen [Sat, 31 Aug 2013 04:14:58 +0000 (00:14 -0400)]
Revert "Deprecate and ignore gtk-entry-password-hint-timeout"

This reverts commit 4b5a389e88af7e7a1fa9e33294642dcfbc2832ec.

This change caused considerable concern about accidental
leaking of passwords, see e.g.

https://bugzilla.gnome.org/show_bug.cgi?id=706563
https://bugzilla.gnome.org/show_bug.cgi?id=706873
https://bugzilla.redhat.com/show_bug.cgi?id=994237

We may have to do something else for password entries, such
as the windows-style 'peekabo' icon.

12 years agoDon't mark a space as translatable
Matthias Clasen [Sat, 31 Aug 2013 03:44:04 +0000 (23:44 -0400)]
Don't mark a space as translatable

There is no point in translating a string that consists just
of a single space. Pointed out in

https://bugzilla.gnome.org/show_bug.cgi?id=706778

12 years agoUpdated Irish translation
Seán de Búrca [Sat, 31 Aug 2013 03:03:09 +0000 (21:03 -0600)]
Updated Irish translation

12 years agoFix up a11y tests for recent changes
Matthias Clasen [Sat, 31 Aug 2013 02:24:09 +0000 (22:24 -0400)]
Fix up a11y tests for recent changes

The fix for child properties made the a11y dump for the assistant
case change. After close inspection, the new output is what the
code intends to produce, so update the expected result.