]> Pileus Git - ~andy/gtk/log
~andy/gtk
13 years agoUpdated Norwegian bokmål translation.
Kjartan Maraas [Sun, 6 Feb 2011 18:14:53 +0000 (19:14 +0100)]
Updated Norwegian bokmål translation.

13 years agoNo more gtk_plug_*/gtk_socket_* symbols on win32
Hans Breuer [Sun, 6 Feb 2011 14:05:35 +0000 (15:05 +0100)]
No more gtk_plug_*/gtk_socket_* symbols on win32

13 years agowin32: no more GdkNativeWindow and related changes
Hans Breuer [Sun, 6 Feb 2011 14:03:32 +0000 (15:03 +0100)]
win32: no more GdkNativeWindow and related changes

 - replace GdkNativeWindow with HWND, remove type casts
 - no more GdkDisplayClass::get_drag_protocol but GdkWindowImpl::get_drag_protocol
 - remove *_client_message*()

13 years agowin32: msvc build update
Hans Breuer [Sun, 30 Jan 2011 21:06:52 +0000 (22:06 +0100)]
win32: msvc build update

13 years agogtk: gtk_cell_renderer_get_state(): check the widget's effective sensitivity
Michael Natterer [Sun, 6 Feb 2011 13:47:58 +0000 (14:47 +0100)]
gtk: gtk_cell_renderer_get_state(): check the widget's effective sensitivity

13 years agogtk: gtk_cell_renderer_get_state(): a row can be both selected and insensitive
Michael Natterer [Sun, 6 Feb 2011 13:43:55 +0000 (14:43 +0100)]
gtk: gtk_cell_renderer_get_state(): a row can be both selected and insensitive

so pull the SELECTED state flag out of the !sensitive branch. Also,
don't make FOCUSED depend on SELECTED here, it's up to the widget to
decide whether or not that is possible.

13 years agoUpdated Polish translation
Piotr Drąg [Sun, 6 Feb 2011 13:11:57 +0000 (14:11 +0100)]
Updated Polish translation

13 years agogtk: make the .trough background-color darker() as it was in GTK+ 2.x
Michael Natterer [Sun, 6 Feb 2011 12:48:42 +0000 (13:48 +0100)]
gtk: make the .trough background-color darker() as it was in GTK+ 2.x

13 years agos/state_changed/state_flags_changed/ in GtkComboBox
Carlos Garnacho [Sun, 6 Feb 2011 01:52:51 +0000 (02:52 +0100)]
s/state_changed/state_flags_changed/ in GtkComboBox

13 years agoHandle insensitive cell renderer flags in gtk_cell_renderer_get_state()
Carlos Garnacho [Sun, 6 Feb 2011 01:51:20 +0000 (02:51 +0100)]
Handle insensitive cell renderer flags in gtk_cell_renderer_get_state()

13 years agoGtkStyleContext is meant to have a GdkScreen
Carlos Garnacho [Sun, 6 Feb 2011 01:40:06 +0000 (02:40 +0100)]
GtkStyleContext is meant to have a GdkScreen

It is used to get the default providers, without them
the style context can't do much. A check for NULL screen
is done before any sensitive call to
gtk_style_context_set_screen(), in the hope that any widget
will open the display before doing anything related to
styling. Fixes bug #641429, reported by Bastien Nocera.

13 years agogtk: bail out in render_background_internal() if the extents are <= 0
Michael Natterer [Sun, 6 Feb 2011 10:13:42 +0000 (11:13 +0100)]
gtk: bail out in render_background_internal() if the extents are <= 0

otherwise we run into scaling the cairo_t into a borken state that's
not fixable by cairo_restore().

13 years agogtk: remove function gtk_entry_queue_draw() and call gtk_widget_queue_draw()
Michael Natterer [Sat, 5 Feb 2011 22:52:19 +0000 (23:52 +0100)]
gtk: remove function gtk_entry_queue_draw() and call gtk_widget_queue_draw()

the old function was a leftover from when GtkEntry had a window
and was now invalidating the entire toplevel on each entry change.

13 years agogtk: get rid of two cases of invalidating the spinbutton when not needed
Michael Natterer [Sat, 5 Feb 2011 22:42:37 +0000 (23:42 +0100)]
gtk: get rid of two cases of invalidating the spinbutton when not needed

13 years agoUpdated Galician translations
Fran Diéguez [Sat, 5 Feb 2011 23:26:36 +0000 (00:26 +0100)]
Updated Galician translations

13 years agoAvoid tons of warnings from stylecontext initialization without screen
Matthias Clasen [Sat, 5 Feb 2011 16:21:21 +0000 (11:21 -0500)]
Avoid tons of warnings from stylecontext initialization without screen

This reverts part of commit 4dc9b29.

13 years agoDemonstrate the problem with a workaround
Matthias Clasen [Sat, 5 Feb 2011 07:17:00 +0000 (02:17 -0500)]
Demonstrate the problem with a workaround

13 years agoEnsure gtk_render_background()/frame() yield the same shape
Carlos Garnacho [Sat, 5 Feb 2011 12:53:04 +0000 (13:53 +0100)]
Ensure gtk_render_background()/frame() yield the same shape

On rounded boxes the background and the frame wouldn't fit in
perfectly, leaving some visual noise in the corners. This is now
gone.

13 years agoUse cairo_push_group() to mix backgrounds during transitions
Carlos Garnacho [Sat, 5 Feb 2011 12:44:30 +0000 (13:44 +0100)]
Use cairo_push_group() to mix backgrounds during transitions

It also helps remove the cairo_clip() call to achieve the rounded
shape.

13 years agogtk: don't style GtkLabel:selected's color separately
Michael Natterer [Sat, 5 Feb 2011 13:10:18 +0000 (14:10 +0100)]
gtk: don't style GtkLabel:selected's color separately

or labels in selected state will not have a different color. Reviewed
by Carlos.

13 years agoDon't leak a cell area context
Matthias Clasen [Sat, 5 Feb 2011 07:03:35 +0000 (02:03 -0500)]
Don't leak a cell area context

13 years agoMigration guide additions
Matthias Clasen [Sat, 5 Feb 2011 06:56:19 +0000 (01:56 -0500)]
Migration guide additions

Some information about plugs and sockets, and event filters.

13 years agoAdd a testcase for cell positions
Matthias Clasen [Sat, 5 Feb 2011 06:24:21 +0000 (01:24 -0500)]
Add a testcase for cell positions

13 years agogtk-demo: Use symbolic icon that exists
Bastien Nocera [Sat, 5 Feb 2011 02:22:17 +0000 (02:22 +0000)]
gtk-demo: Use symbolic icon that exists

For the Images test.

13 years agoSimple test for display-less widget creation
Matthias Clasen [Sat, 5 Feb 2011 03:06:25 +0000 (22:06 -0500)]
Simple test for display-less widget creation

13 years agoDon't assume a screen exists before realize()
Matthias Clasen [Sat, 5 Feb 2011 03:01:40 +0000 (22:01 -0500)]
Don't assume a screen exists before realize()

Fix for https://bugzilla.gnome.org/show_bug.cgi?id=641429

13 years agoImprove GtkApplication migration guide
Matthias Clasen [Sat, 5 Feb 2011 02:21:14 +0000 (21:21 -0500)]
Improve GtkApplication migration guide

As pointed out in bug 641517, the examples did not actually
do the same.

13 years agoFix a memory leak in GtkCSSProvider
Matthias Clasen [Sat, 5 Feb 2011 02:13:56 +0000 (21:13 -0500)]
Fix a memory leak in GtkCSSProvider

Found by Felix Riemann
https://bugzilla.gnome.org/show_bug.cgi?id=641558

13 years agoXI2: translate group state back into core state
Matthias Clasen [Sat, 5 Feb 2011 02:08:41 +0000 (21:08 -0500)]
XI2: translate group state back into core state

This is necessary to make the key event translation work
as expected.

https://bugzilla.gnome.org/show_bug.cgi?id=641367

13 years agoPass group state to _gdk_x11_device_xi2_translate_state
Matthias Clasen [Sat, 5 Feb 2011 02:01:52 +0000 (21:01 -0500)]
Pass group state to _gdk_x11_device_xi2_translate_state

This will be necessary to make groups work with XI2.

13 years agoShare code to fill GdkEventKey.string
Matthias Clasen [Sat, 5 Feb 2011 01:58:41 +0000 (20:58 -0500)]
Share code to fill GdkEventKey.string

This code was duplicated between the core and XI2 device manager
implementations.

13 years agoUpdated Arabic translation
Khaled Hosny [Fri, 4 Feb 2011 22:31:14 +0000 (00:31 +0200)]
Updated Arabic translation

13 years agoAdd an empty English translation for properties
Matthias Clasen [Fri, 4 Feb 2011 22:18:24 +0000 (17:18 -0500)]
Add an empty English translation for properties

Otherwise the build is unhappy about en in LINGUAS

13 years agoAdd an English translation
Matthias Clasen [Fri, 4 Feb 2011 20:35:44 +0000 (15:35 -0500)]
Add an English translation

Currently, this just includes translations for keysyms.

13 years agoStrip XF86 prefix from untranslated keysyms
Matthias Clasen [Fri, 4 Feb 2011 20:34:36 +0000 (15:34 -0500)]
Strip XF86 prefix from untranslated keysyms

When we don't have translations, at least remove the
ugly XF86 prefix.

13 years agoAdd translation entries for more common keysyms
Matthias Clasen [Fri, 4 Feb 2011 20:33:36 +0000 (15:33 -0500)]
Add translation entries for more common keysyms

In particular, the XF86 keysyms can benefit from a translation
that strips the prefix.

13 years agoRemove debug spew
Matthias Clasen [Fri, 4 Feb 2011 19:05:22 +0000 (14:05 -0500)]
Remove debug spew

13 years agoXI2: don't translate button releases as scroll events
Matthias Clasen [Fri, 4 Feb 2011 17:51:46 +0000 (12:51 -0500)]
XI2: don't translate button releases as scroll events

This fixes scroll-by-2 sympoms seen in combo boxes.
https://bugzilla.gnome.org/show_bug.cgi?id=639846

13 years agoRemove latest bits of GtkItemFactory
Javier Jardón [Fri, 4 Feb 2011 16:58:59 +0000 (16:58 +0000)]
Remove latest bits of GtkItemFactory

13 years agoFixed gtk_recent_chooser_menu_set_current_filter to update the local pointer.
Tristan Van Berkom [Fri, 4 Feb 2011 15:58:40 +0000 (00:58 +0900)]
Fixed gtk_recent_chooser_menu_set_current_filter to update the local pointer.

Before this patch, unsetting the filter manually before disposing
the recent chooser menu (or unsetting the filter twice), would cause
warnings (or invalid memory accesses).

13 years agoUpdated Hebrew translation.
Yaron Shahrabani [Fri, 4 Feb 2011 07:10:23 +0000 (09:10 +0200)]
Updated Hebrew translation.

13 years agoProtect against tree_view being NULL
Matthias Clasen [Thu, 3 Feb 2011 04:54:38 +0000 (23:54 -0500)]
Protect against tree_view being NULL

This apparently happens when setting the "visible" property
from ui files.
https://bugzilla.gnome.org/show_bug.cgi?id=640801

13 years agoAdded set_transient_for() implementation for GdkOffscreenWindow
Tristan Van Berkom [Thu, 3 Feb 2011 14:50:10 +0000 (23:50 +0900)]
Added set_transient_for() implementation for GdkOffscreenWindow

Just a noop that is better than crashing in the case this is called
(it can be called for a toplevel GtkWindow that is parented into
another widget by setting gtk_widget_set_parent_window to an offscreen
window).

13 years ago[GI] Add method annotation for gtk_grab_add and gtk_grab_remove.
Laszlo Pandy [Thu, 3 Feb 2011 14:20:23 +0000 (15:20 +0100)]
[GI] Add method annotation for gtk_grab_add and gtk_grab_remove.

13 years ago[GI] Add method annotations to remaining gtk_drag_* functions.
Laszlo Pandy [Thu, 3 Feb 2011 14:09:08 +0000 (15:09 +0100)]
[GI] Add method annotations to remaining gtk_drag_* functions.

https://bugzilla.gnome.org/show_bug.cgi?id=639945

13 years agoFixing GtkColorButton to redraw itself when gtk_color_button_set_rgba() is called.
Tristan Van Berkom [Thu, 3 Feb 2011 14:02:15 +0000 (23:02 +0900)]
Fixing GtkColorButton to redraw itself when gtk_color_button_set_rgba() is called.

13 years agoAdded default implementations of various GtkCellArea vfuncs.
Tristan Van Berkom [Thu, 3 Feb 2011 10:03:49 +0000 (19:03 +0900)]
Added default implementations of various GtkCellArea vfuncs.

Having default implementations on abstract classes that do nothing
or print errors make life easier for language bindings implementors,
inside GTK+ this makes little or no difference.

13 years agoAdded default implementation of GtkCellRendererClass.get_request_mode()
Tristan Van Berkom [Thu, 3 Feb 2011 10:01:55 +0000 (19:01 +0900)]
Added default implementation of GtkCellRendererClass.get_request_mode()

Having default implementations on abstract classes that do nothing
or print errors make life easier for language bindings implementors,
inside GTK+ this makes little or no difference.

13 years agoDon't g_object_unref cairo surfaces
Matthias Clasen [Thu, 3 Feb 2011 03:05:26 +0000 (22:05 -0500)]
Don't g_object_unref cairo surfaces

This was causing warnings when dragging text from labels.

13 years agoUpdated Polish translation
Piotr Drąg [Wed, 2 Feb 2011 17:00:19 +0000 (18:00 +0100)]
Updated Polish translation

13 years agoUpdated Polish translation
Piotr Drąg [Wed, 2 Feb 2011 16:59:47 +0000 (17:59 +0100)]
Updated Polish translation

13 years ago[GI] Annotate gtk_drag_{dest,source}* as methods
Martin Pitt [Wed, 2 Feb 2011 16:51:48 +0000 (17:51 +0100)]
[GI] Annotate gtk_drag_{dest,source}* as methods

As these do not start with gtk_widget_*, the GI scanner does not recognize them
as methods. With the recently added (method) annotation [1] it is now possible
to mark them as proper methods of GtkWidget.

https://bugzilla.gnome.org/show_bug.cgi?id=639945

[1] http://git.gnome.org/browse/gobject-introspection/commit/?id=09bca85dd4

13 years agoUpdated Korean translation
Changwoo Ryu [Wed, 2 Feb 2011 15:28:37 +0000 (00:28 +0900)]
Updated Korean translation

13 years ago[GI] Mark mis-detected constructors as such
Martin Pitt [Wed, 2 Feb 2011 14:11:06 +0000 (15:11 +0100)]
[GI] Mark mis-detected constructors as such

constructors which take an object of the same class as its first argument are
mis-detected as method call with "self" argument by the GIR scanner. Using the
new (constructor) annotation from bug 561264, mark some of them as proper
constuctors, so that you can call them with NULL as first argument from
bindings; in particular, this fixes gdk_window_new() and the
gtk_radio_button_new_with*() constructors.

13 years agoFix the build with --enable-gtk2-dependency
Matthias Clasen [Wed, 2 Feb 2011 06:30:21 +0000 (01:30 -0500)]
Fix the build with --enable-gtk2-dependency

13 years agoFixed GtkCellLayout buildable implementation to set cell properties explicitly set...
Tristan Van Berkom [Wed, 2 Feb 2011 06:20:42 +0000 (15:20 +0900)]
Fixed GtkCellLayout buildable implementation to set cell properties explicitly set to "" (i.e. explicitly set to nothing).

13 years agoFixed GtkContainer buildable implementation to set packing properties explicitly...
Tristan Van Berkom [Wed, 2 Feb 2011 06:19:39 +0000 (15:19 +0900)]
Fixed GtkContainer buildable implementation to set packing properties explicitly set to "" (i.e. explicitly set to nothing).

13 years agoDon't build pixbuf engine, still
Matthias Clasen [Wed, 2 Feb 2011 06:00:51 +0000 (01:00 -0500)]
Don't build pixbuf engine, still

This was accidentally re-enabled together with the wimp.

13 years agoFix another GtkFixed regression, in gtk_fixed_forall()
Matthias Clasen [Wed, 2 Feb 2011 05:40:02 +0000 (00:40 -0500)]
Fix another GtkFixed regression, in gtk_fixed_forall()

b3f6f67c changed the loop from while() to for() in gtk_fixed_forall(),
but that's wrong since the callback can have side-effects on the list,
in case the current child gets removed. And that's the case when the
widget is destroyed.

Patch by Vincent Untz
https://bugzilla.gnome.org/show_bug.cgi?id=641196

13 years agoBump version
Matthias Clasen [Wed, 2 Feb 2011 05:36:41 +0000 (00:36 -0500)]
Bump version

13 years agoUpdate quartz to reflect deletion of GdkNativeWindow and client_message functions.
John Ralls [Tue, 1 Feb 2011 21:53:00 +0000 (13:53 -0800)]
Update quartz to reflect deletion of GdkNativeWindow and client_message functions.

13 years agoAdd another bug ref
Matthias Clasen [Tue, 1 Feb 2011 19:11:20 +0000 (14:11 -0500)]
Add another bug ref

13 years agoFix crash in gtk_fixed_remove()
Vincent Untz [Tue, 1 Feb 2011 19:01:49 +0000 (20:01 +0100)]
Fix crash in gtk_fixed_remove()

b3f6f67c changed the loop from while() to for(), but the line to move to
the next child was kept, so we could go past the end of the list.

https://bugzilla.gnome.org/show_bug.cgi?id=641176

13 years agoRemove accidentally exported symbol
Matthias Clasen [Tue, 1 Feb 2011 18:54:27 +0000 (13:54 -0500)]
Remove accidentally exported symbol

13 years agoUpdates for 2.99.3
Matthias Clasen [Tue, 1 Feb 2011 18:39:41 +0000 (13:39 -0500)]
Updates for 2.99.3

13 years agoAPI: gdk: Get rid of GdkNativeWindow
Benjamin Otte [Tue, 1 Feb 2011 17:31:44 +0000 (18:31 +0100)]
API: gdk: Get rid of GdkNativeWindow

Also get rid of the GDK_NATIVE_WINDOW_POINTER define.

13 years agox11: Use Window isntead of GdkNativeWindow
Benjamin Otte [Tue, 1 Feb 2011 17:29:20 +0000 (18:29 +0100)]
x11: Use Window isntead of GdkNativeWindow

13 years agoAPI: gdk: Change get_drag_window() API
Benjamin Otte [Tue, 1 Feb 2011 17:11:35 +0000 (18:11 +0100)]
API: gdk: Change get_drag_window() API

The previous function gdk_drag_get_protocol_for_display() took native
window handles, so it had to be changed. Because it didn't do what it
was named to do (it didn't return a protocol even though it was named
get_protocol) and because it doesn't operate on the display anymore but
on the actual window, it's now called gdk_window_get_drag_protocol().

13 years agoAPI: gdk: Make GDK_DRAG_PROTOCOL_NONE equal to 0
Benjamin Otte [Tue, 1 Feb 2011 16:29:33 +0000 (17:29 +0100)]
API: gdk: Make GDK_DRAG_PROTOCOL_NONE equal to 0

THis way, we can do comparisons like "if (protocol)"

13 years agoAPI: gdk: Make GdkEventSelection.requestor a GdkWindow
Benjamin Otte [Tue, 1 Feb 2011 16:22:04 +0000 (17:22 +0100)]
API: gdk: Make GdkEventSelection.requestor a GdkWindow

instead of a GdkNativeWindow. Also change gdk_selection_notify() API to
take a GdkWindow to match this change.

13 years agox11: Use gdk_x11_window_foreign_new_for_display() unconditionally
Benjamin Otte [Tue, 1 Feb 2011 15:36:23 +0000 (16:36 +0100)]
x11: Use gdk_x11_window_foreign_new_for_display() unconditionally

gdk_x11_window_lookup_window() calls are done by this function alreayd,
so no need to do them manually.

13 years agoFix a warning
Matthias Clasen [Tue, 1 Feb 2011 14:48:32 +0000 (09:48 -0500)]
Fix a warning

13 years agoRemove a dead member
Matthias Clasen [Tue, 1 Feb 2011 14:44:33 +0000 (09:44 -0500)]
Remove a dead member

The client_pointer field in GdkX11DeviceManagerXI2 is not used anywhere.

13 years agoMake xim work with XI2
Matthias Clasen [Tue, 1 Feb 2011 14:42:53 +0000 (09:42 -0500)]
Make xim work with XI2

We cannot totally ignore core key events, since XFilterEvent generates
those at times and relies on them having an effect.
https://bugzilla.gnome.org/show_bug.cgi?id=640965

13 years agoMake treeview expanders work reliably
Matthias Clasen [Tue, 1 Feb 2011 06:28:45 +0000 (01:28 -0500)]
Make treeview expanders work reliably

It turns out that the gtk_grab_remove() can trigger a do_prelight()
call, which may end up changing prelight_node, and then the state
gets messed up. Moving the grab removal until after we're done with
button_pressed_node and prlight_node makes expanders work reliably.

One thing that is still not right is that the expander doesn't get
prelight again after the animation is done, if you manage to release
without any additional motion events.

https://bugzilla.gnome.org/show_bug.cgi?id=641039

13 years agogdk: Remove GdkEventClient
Benjamin Otte [Tue, 1 Feb 2011 05:30:27 +0000 (06:30 +0100)]
gdk: Remove GdkEventClient

... and all APIs making use of it.

That code like it hasn't been touched in years, Google codesearch
didn't find any users and most importantly it's a horrendous API, so
let's just make it die instead of having to port it over to
non-GdkNativeWindow usage, which would be required for multi-backend
GDK.

http://mail.gnome.org/archives/gtk-devel-list/2011-January/msg00049.html

13 years agotests: Remove refreh button from testtoolbar test
Benjamin Otte [Tue, 1 Feb 2011 05:29:51 +0000 (06:29 +0100)]
tests: Remove refreh button from testtoolbar test

It didn't work and did that by using APIs that are about to go away.

13 years agotestgtk: Remove ability to reload all rc files
Benjamin Otte [Tue, 1 Feb 2011 05:27:46 +0000 (06:27 +0100)]
testgtk: Remove ability to reload all rc files

That ability didn't work for a while anyway, so removing client messages
doesn't hurt.

13 years agotests: Remove clientmessage test
Benjamin Otte [Tue, 1 Feb 2011 05:25:57 +0000 (06:25 +0100)]
tests: Remove clientmessage test

Client messages will not be handled by GDK3 anymore, so no need to test
them.

13 years agox11: Do dnd event handling via gdk_window_add_filter()
Benjamin Otte [Tue, 1 Feb 2011 05:12:51 +0000 (06:12 +0100)]
x11: Do dnd event handling via gdk_window_add_filter()

13 years agox11: Use macros for byte order to set byte order
Benjamin Otte [Tue, 1 Feb 2011 05:04:01 +0000 (06:04 +0100)]
x11: Use macros for byte order to set byte order

No need to run sophisticated functions to guess it.

13 years agox11: Register wm_protocols function with gdk_window_add_filter()
Benjamin Otte [Tue, 1 Feb 2011 04:31:41 +0000 (05:31 +0100)]
x11: Register wm_protocols function with gdk_window_add_filter()

The client message code is about to go away, so we need to use a
different way to get at ClientMessage output.

13 years agox11: Fill in event->any.window before calling filter functions
Benjamin Otte [Tue, 1 Feb 2011 04:55:10 +0000 (05:55 +0100)]
x11: Fill in event->any.window before calling filter functions

a) We advertise it for GdkFilterFunc
b) It's necessary to differentiate between different event types with
   multi-backend GTK.

13 years agox11: Handle motif client message filter just like xdnd filters
Benjamin Otte [Tue, 1 Feb 2011 03:35:49 +0000 (04:35 +0100)]
x11: Handle motif client message filter just like xdnd filters

13 years agox11: Use event->any.window instead of any random member
Benjamin Otte [Tue, 1 Feb 2011 02:53:21 +0000 (03:53 +0100)]
x11: Use event->any.window instead of any random member

13 years agoAPI: gtk: Remove client_event vfunc from GtkWidget
Benjamin Otte [Mon, 31 Jan 2011 22:18:47 +0000 (23:18 +0100)]
API: gtk: Remove client_event vfunc from GtkWidget

Also remove the (de-facto unused) implementation from GtkWindow that was
only used for sockets/plugs.

13 years agoDeal gracefully with unowned selections
Matthias Clasen [Tue, 1 Feb 2011 00:35:51 +0000 (19:35 -0500)]
Deal gracefully with unowned selections

If a selection looses its owner without a replacement, owner is
None, which we did not handle very well.

https://bugzilla.gnome.org/show_bug.cgi?id=641042

13 years agoFix several bugs handling GtkTrayIcon symbolic colors
Dan Winship [Tue, 1 Feb 2011 00:13:29 +0000 (19:13 -0500)]
Fix several bugs handling GtkTrayIcon symbolic colors

https://bugzilla.gnome.org/show_bug.cgi?id=641059

13 years agoClarify models vs callbacks in entry completion docs
Matthias Clasen [Mon, 31 Jan 2011 23:49:20 +0000 (18:49 -0500)]
Clarify models vs callbacks in entry completion docs

See https://bugzilla.gnome.org/show_bug.cgi?id=641073

13 years agoDon't use GTK_IS_PLUG for non-X11 backends
Kristian Rietveld [Mon, 31 Jan 2011 23:26:04 +0000 (00:26 +0100)]
Don't use GTK_IS_PLUG for non-X11 backends

13 years agoCheck whether mask is NULL
Kristian Rietveld [Mon, 31 Jan 2011 22:30:56 +0000 (23:30 +0100)]
Check whether mask is NULL

13 years agoOnly include gtktrayicon.h when building for X11
Kristian Rietveld [Mon, 31 Jan 2011 22:10:49 +0000 (23:10 +0100)]
Only include gtktrayicon.h when building for X11

GtkTrayIcon depends on GtkPlug which is X11-specific.  This fixes the
build for Quartz.

13 years agoBandaid fix for icon view subclassing
Matthias Clasen [Mon, 31 Jan 2011 22:34:37 +0000 (17:34 -0500)]
Bandaid fix for icon view subclassing

I've decided that it is isn't feasible to make cell areas runtime-settable
in the time we have left before 3.0, therefore, I'm going with the
approach to allow init() functions to instantiate the default cell area
and issue a warning if a construct property is ignored.

This is not ideal, but it keeps existing icon view and combo box
subclasses working.

https://bugzilla.gnome.org/show_bug.cgi?id=639139

13 years agoClarify key-{press,release}-event docs.
Stefan Kost [Mon, 31 Jan 2011 21:07:34 +0000 (23:07 +0200)]
Clarify key-{press,release}-event docs.

Mention key-repeat in key-press and fix a copy'n'paste in key-release docs.

13 years agoFixed GtkToolPalette ->forall implementation to be gtk_widget_destroy friendly
Tristan Van Berkom [Mon, 31 Jan 2011 19:57:05 +0000 (04:57 +0900)]
Fixed GtkToolPalette ->forall implementation to be gtk_widget_destroy friendly

The forall() loop was buggy as it was skipping items in the list when
the current item gets removed from the groups array as a result of
calling the callback (causing memory leaks).

13 years agoAdded "shift" modifier versions of keybings for GtkFileChooser "folder-up"/"folder...
Tristan Van Berkom [Sat, 29 Jan 2011 11:06:25 +0000 (20:06 +0900)]
Added "shift" modifier versions of keybings for GtkFileChooser "folder-up"/"folder-down" bindings.

Also updated documentation, closes bug 322926

13 years agoUpdated Spanish translation
Daniel Mustieles [Mon, 31 Jan 2011 18:33:55 +0000 (19:33 +0100)]
Updated Spanish translation

13 years agoFix a segfault in gtk_statusbar_remove_all
Matthias Clasen [Mon, 31 Jan 2011 16:32:19 +0000 (11:32 -0500)]
Fix a segfault in gtk_statusbar_remove_all

https://bugzilla.gnome.org/show_bug.cgi?id=640487

13 years agocss-provider: Fix up the Switch style
Emmanuele Bassi [Wed, 29 Dec 2010 03:09:14 +0000 (03:09 +0000)]
css-provider: Fix up the Switch style

As it was with the old style, when the Switch is in the "on" state the
trough should be using the selected background color.

13 years agoSpeed up expander animation
Matthias Clasen [Mon, 31 Jan 2011 15:51:09 +0000 (10:51 -0500)]
Speed up expander animation

Go back to 200ms, which is what we had in GTK+ 2.x.
Themes can change this, of course.