]> Pileus Git - ~andy/gtk/log
~andy/gtk
11 years agolabel: Fix ellipsize and wrap being set
Benjamin Otte [Fri, 2 Nov 2012 20:49:32 +0000 (21:49 +0100)]
label: Fix ellipsize and wrap being set

The label code assumed that Pango treats this as "wrap to as much space
as possible and then ellipsize all the lines", but for Pango, ellipsize
takes precedence over wrap. So do the same thing in GtkLabel.

Also updated is the reftest that checked this behavior.

11 years agoreftests: Improve reftest performance
Benjamin Otte [Fri, 2 Nov 2012 20:38:16 +0000 (21:38 +0100)]
reftests: Improve reftest performance

Get rid of all the event boxes in this test. Event boxes need GDK
windows which cost a lot of performance when running the test and they
clip the label output.

Getting rid of the clipping also shows 2 bugs in this test that weren't
visible before. Those will be fixed in a followup patch.

11 years agoreftests: Make label-sizing tests use better CSS
Benjamin Otte [Fri, 2 Nov 2012 18:16:46 +0000 (19:16 +0100)]
reftests: Make label-sizing tests use better CSS

ow that labels can have backgrounds, just use label backgrounds

11 years agosizerequest: Export _gtk_widget_compute_size_for_orientation()
Benjamin Otte [Fri, 2 Nov 2012 12:40:00 +0000 (13:40 +0100)]
sizerequest: Export _gtk_widget_compute_size_for_orientation()

and add an "ignore_size_groups" flag to it. This way we can use it for
size group shenanigans.

11 years agosizegroup: Move GtkSizeGroupMode to gtkenums.h
Benjamin Otte [Fri, 2 Nov 2012 12:39:19 +0000 (13:39 +0100)]
sizegroup: Move GtkSizeGroupMode to gtkenums.h

This is in preparation for the next patch, which would otherwise lead to
conflicts.

11 years agosizerequest: Cache sizes without size groups
Benjamin Otte [Thu, 1 Nov 2012 23:50:54 +0000 (00:50 +0100)]
sizerequest: Cache sizes without size groups

We compute on-demand for size groups anyway, so we can (in theory, this
patch doesn't do that yet) get around costly cache blowing when
invalidating single widgets of a size group this way.

11 years agosizegroups: Use is_visible() instead of get_mapped() for visibility
Benjamin Otte [Thu, 1 Nov 2012 23:44:30 +0000 (00:44 +0100)]
sizegroups: Use is_visible() instead of get_mapped() for visibility

The current approach of using gtk_widget_get_mapped() is broken:
The usual steps taken when showing a window are:
(1) request the sizes
(2) allocate the sizes
(3) show the window in the allocated size

Showing the window with a random size between steps (1) and (2) would of
course
result in extra work and potential flickering when the widgets get
resized to
their proper sizes.

However, as GtkSizeGroup::ignore-hidden uses gtk_widget_get_mapped() to
determine visibility for a widget, the following will happen:
(1) the widget will request a 0 size
(2) the widget will be allocated a 0 size
(3) the widget will be too small when it is shown

gtk_widget_get_visible() however is set in advance. Note that toggling
visibility also causes a gtk-widget_queue_resize() call already so we
take care of changes in here automatically.

11 years agoAPI: Add gtk_widget_is_visible()
Benjamin Otte [Thu, 1 Nov 2012 23:42:45 +0000 (00:42 +0100)]
API: Add gtk_widget_is_visible()

This is a recursive gtk_widget_get_visible(): Returns TRUE if the widget
and all its parents are visible.

11 years agosizegroup: Handle hfw in size groups
Benjamin Otte [Thu, 1 Nov 2012 23:33:42 +0000 (00:33 +0100)]
sizegroup: Handle hfw in size groups

11 years agosizegroup: Add a function for clarity
Benjamin Otte [Thu, 1 Nov 2012 23:28:10 +0000 (00:28 +0100)]
sizegroup: Add a function for clarity

... and restructure code to accomodate that function.

11 years agosizegroup: Don't keep groups around everywhere
Benjamin Otte [Thu, 1 Nov 2012 23:18:11 +0000 (00:18 +0100)]
sizegroup: Don't keep groups around everywhere

The code is only interested in the actual widgets that belong together,
not in the groups. So just don't return the groups.

11 years agosizegroup: Check ignore_hidden flag when adding groups
Benjamin Otte [Thu, 1 Nov 2012 23:12:48 +0000 (00:12 +0100)]
sizegroup: Check ignore_hidden flag when adding groups

Instead of only checking the ignore_hidden flag when getting the
preferred sizes, respect it already when constructing the list of
widgets. This way, widgets don't queue resizes for groups they're
ignored in anyway.

11 years agosizegroup: Use for loops
Benjamin Otte [Thu, 1 Nov 2012 23:06:02 +0000 (00:06 +0100)]
sizegroup: Use for loops

For loops to loop over lists look nicer and actually do the right thing
with "break" and "continue" statements. So they are vastly preferred to
while loops.

11 years agosizegroups: Restructure code
Benjamin Otte [Thu, 1 Nov 2012 22:58:23 +0000 (23:58 +0100)]
sizegroups: Restructure code

This way, we do the checks at the start of the effected function, not
before calling it.

11 years agosizegroup: Don't cache the sizes anymore
Benjamin Otte [Thu, 1 Nov 2012 22:41:38 +0000 (23:41 +0100)]
sizegroup: Don't cache the sizes anymore

This simplifies code and because sizes are cached by the widgets
themselves, it's not a large performance problem (unless people use huge
amounts of widgets in a single size group, but who does that?

11 years agoMerge branch 'bgo687196-filesystemmodel-crash'
Federico Mena Quintero [Fri, 2 Nov 2012 20:13:02 +0000 (14:13 -0600)]
Merge branch 'bgo687196-filesystemmodel-crash'

11 years agobgo#687196 - Fix model corruption during file removal
Federico Mena Quintero [Thu, 1 Nov 2012 23:48:23 +0000 (17:48 -0600)]
bgo#687196 - Fix model corruption during file removal

The main problem is that we were emitting the row-deleted signal for the model in the middle
of the process that actually deletes the row from the model (remove the row from the array,
update the model->file_lookup hash table, etc.).  In the model's caller, one of the row-deleted
callbacks was requesting an iter, which caused the model to revalidate itself - but it did
this while it was in an inconsistent state.  This led to an assertion failure later when the
model resorted itself.

The fix in remove_file() is like this:

* The filteredness/visibility of the deleted node is not updated.  The
  node will simply be gone; we don't need to update those values at
  all.

* We invalidate just the node that is being deleted.

* The model->file_lookup hash table is not completely nuked; instead,
  we carefully adjust its indices.

* The row-deleted signal is only emitted at the very end, when
  deletion is complete and the model is consistent.

Many thanks to William Hua for doing the detective work on this bug!

Signed-off-by: Federico Mena Quintero <federico@gnome.org>
11 years agoMake freeze_updates() and thaw_updates() static functions
Federico Mena Quintero [Thu, 1 Nov 2012 23:21:47 +0000 (17:21 -0600)]
Make freeze_updates() and thaw_updates() static functions

They were in the semi-public API of GtkFileSystemModel, but never actually used outside of it.

Signed-off-by: Federico Mena Quintero <federico@gnome.org>
11 years agoComments on how the filtering and sorting processes work
Federico Mena Quintero [Thu, 1 Nov 2012 23:19:10 +0000 (17:19 -0600)]
Comments on how the filtering and sorting processes work

Signed-off-by: Federico Mena Quintero <federico@gnome.org>
11 years agosizegroup: Always at least use widget's size
Benjamin Otte [Thu, 1 Nov 2012 19:22:35 +0000 (20:22 +0100)]
sizegroup: Always at least use widget's size

When widgets were hidden, they were otherwise assigned a 0 size.

11 years agosettings: Reset all styles when the enable-animations settings changes
Benjamin Otte [Thu, 1 Nov 2012 11:23:42 +0000 (12:23 +0100)]
settings: Reset all styles when the enable-animations settings changes

https://bugzilla.gnome.org/show_bug.cgi?id=686021

11 years agoRaleigh: Fix spinners with disabled animations
Benjamin Otte [Thu, 1 Nov 2012 11:16:06 +0000 (12:16 +0100)]
Raleigh: Fix spinners with disabled animations

When animations are disabled, active and inactive spinners should look
different.

https://bugzilla.gnome.org/show_bug.cgi?id=686021

11 years agocssimage: Implement some equal functions
Benjamin Otte [Wed, 31 Oct 2012 23:22:39 +0000 (00:22 +0100)]
cssimage: Implement some equal functions

11 years agoRemove argument to _gtk_file_system_model_update_file() that should not be part of...
Federico Mena Quintero [Thu, 1 Nov 2012 02:54:02 +0000 (20:54 -0600)]
Remove argument to _gtk_file_system_model_update_file() that should not be part of the internal API

Signed-off-by: Federico Mena Quintero <federico@gnome.org>
11 years agoRename gtk_tree_path_new_from_node() to tree_path_new_from_node()
Federico Mena Quintero [Thu, 1 Nov 2012 02:21:47 +0000 (20:21 -0600)]
Rename gtk_tree_path_new_from_node() to tree_path_new_from_node()

This is a function internal to the file system model; let's not pollute the gtk_tree_path namespace.

Also, make the 'i' variable into 'r' as it refers to a row index, not a file-array index (for
consistency with the docs and the rest of the code).

Signed-off-by: Federico Mena Quintero <federico@gnome.org>
11 years agoicon-theme: support loading symbolic GFileIcons from generic URIs
Cosimo Cecchi [Sun, 28 Oct 2012 19:01:30 +0000 (15:01 -0400)]
icon-theme: support loading symbolic GFileIcons from generic URIs

Right now we support loading and recoloring symbolic GFileIcons, but
only if the underlying GFile has a local path. This breaks when the
GFileIcon is loaded from a GResource, which is a reasonable option for an
application that wants to ship a custom symbolic icon.

This patch changes GtkIconInfo to store a GFile together with the file
path, and changes the symbolic icon lookup code to use the GFile URI,
which transparently makes the code work also for GResources.

https://bugzilla.gnome.org/show_bug.cgi?id=687059

11 years agothemingbackground: Remove struct members
Benjamin Otte [Sat, 27 Oct 2012 01:22:03 +0000 (03:22 +0200)]
themingbackground: Remove struct members

... and put them in the only function they are used in.

11 years agoreftests: Add reftest for fractional border sizes
Benjamin Otte [Sat, 27 Oct 2012 01:17:43 +0000 (03:17 +0200)]
reftests: Add reftest for fractional border sizes

11 years agothemingengine: Draw fradctional border sizes
Benjamin Otte [Sat, 27 Oct 2012 01:16:35 +0000 (03:16 +0200)]
themingengine: Draw fradctional border sizes

11 years agoreftests: Fix linear-gradient reftest
Benjamin Otte [Sat, 27 Oct 2012 00:40:14 +0000 (02:40 +0200)]
reftests: Fix linear-gradient reftest

... and add it to the Makefile

11 years agothemingbackground: Remove GtkThemingBackgroundLayer
Benjamin Otte [Sat, 27 Oct 2012 00:25:36 +0000 (02:25 +0200)]
themingbackground: Remove GtkThemingBackgroundLayer

The struct was just the index. So just pass the index around instead of
a full struct.

11 years agothemingbackground: Use get_box() for background image size
Benjamin Otte [Sat, 27 Oct 2012 00:23:23 +0000 (02:23 +0200)]
themingbackground: Use get_box() for background image size

11 years agothemingbackground: Introduce gtk_theming_background_get_box()
Benjamin Otte [Sat, 27 Oct 2012 00:16:42 +0000 (02:16 +0200)]
themingbackground: Introduce gtk_theming_background_get_box()

to query the different clip boxes used by the background drawing code.
Use this function to query these boxes when clipping.

11 years agothemingbackground: Add content_box variable
Benjamin Otte [Sat, 27 Oct 2012 00:07:43 +0000 (02:07 +0200)]
themingbackground: Add content_box variable

... to go with border_box and padding_box.

11 years agothemingbackground: Get rid of flags variable
Benjamin Otte [Fri, 26 Oct 2012 23:55:41 +0000 (01:55 +0200)]
themingbackground: Get rid of flags variable

11 years agothemingbackground: Move image variable
Benjamin Otte [Fri, 26 Oct 2012 23:52:08 +0000 (01:52 +0200)]
themingbackground: Move image variable

... from the Layer struct onto the stack of the only function using it.

11 years agothemingbackground: Restructure code some more
Benjamin Otte [Fri, 26 Oct 2012 23:39:30 +0000 (01:39 +0200)]
themingbackground: Restructure code some more

Move variable initialization outside the first code with side effects.
This allows adding some more early returns, including one for code that
used to trigger g_return_if_fail() in certain corner cases.

11 years agothemingbackground: Restructure code
Benjamin Otte [Fri, 26 Oct 2012 23:35:00 +0000 (01:35 +0200)]
themingbackground: Restructure code

Make if statements encompassing the whole function into early returns.
The rest of the diff is reindenting.

11 years agocssimage: Add a warning for drawing empty images
Benjamin Otte [Fri, 26 Oct 2012 23:31:59 +0000 (01:31 +0200)]
cssimage: Add a warning for drawing empty images

width and height of an image must be > 0 for the image to get drawn.
It's up to the code further up to ensure that this is not happening.

11 years agoreftests: Add a reftest for recent commit
Benjamin Otte [Fri, 26 Oct 2012 21:38:19 +0000 (23:38 +0200)]
reftests: Add a reftest for recent commit

Check that a computed background-size of 0 is treated as 0.

11 years agocssvalue: Compute "background-size: 0 0" properly
Benjamin Otte [Fri, 26 Oct 2012 21:36:39 +0000 (23:36 +0200)]
cssvalue: Compute "background-size: 0 0" properly

Previously a computed value of 0 was treated as "auto", which is wrong.

11 years agostylecontext: Don't use bg image in gtk_style_context_set_background()
Benjamin Otte [Fri, 26 Oct 2012 21:33:40 +0000 (23:33 +0200)]
stylecontext: Don't use bg image in gtk_style_context_set_background()

Old code tried to use the "background-image" proeprty for setting the
default image background. While this used to work in the early days of
GTK3, today it is grossly misleading as the backgronud image may be
resized, repositioned and semi-translucent which causes very weird
artifacts when rendering.

So we use the background-color only instead.

11 years agocssimage: Add an equal vfunc
Benjamin Otte [Thu, 11 Oct 2012 10:10:20 +0000 (12:10 +0200)]
cssimage: Add an equal vfunc

No implementations for it exist yet.

11 years agomenuitem: Draw background unconditionally
Benjamin Otte [Thu, 11 Oct 2012 09:48:32 +0000 (11:48 +0200)]
menuitem: Draw background unconditionally

11 years agosettings: Use _gtk_css_provider_load_named()
Benjamin Otte [Wed, 19 Sep 2012 14:41:19 +0000 (16:41 +0200)]
settings: Use _gtk_css_provider_load_named()

This way we create one provider per settings object instead of stuffing
it into a global unchanging never-deleting hash table.
Also, we now reload the theme when instructed instead of keeping the old
loaded (and possibly stale) data forever.

https://bugzilla.gnome.org/show_bug.cgi?id=683896

11 years agocssprovider: Move fallback code into _gtk_css_provider_load_named()
Benjamin Otte [Wed, 19 Sep 2012 14:30:27 +0000 (16:30 +0200)]
cssprovider: Move fallback code into _gtk_css_provider_load_named()

This makes sure the full theme loading logic resides in one function and
isn't scattered around.

As a side-effect, the hash table kept by gtk_css_provider_get_named()
will now be populated with fallback themes. This will not be a problem
after the next commit though.

11 years agocssprovider: Export gtk_css_provider_load_named) function
Benjamin Otte [Wed, 19 Sep 2012 14:15:05 +0000 (16:15 +0200)]
cssprovider: Export gtk_css_provider_load_named) function

... and document it.
For now, the function is only exported internally.

11 years agocssprovider: Split out theme loading function
Benjamin Otte [Wed, 19 Sep 2012 14:07:39 +0000 (16:07 +0200)]
cssprovider: Split out theme loading function

Split maintaining the global themes hash table and the theme loading
code into two functions.
This also fixes leaking the provider when loading a theme from a builtin
resource.

11 years agoUpdated Galician translations
Fran Diéguez [Wed, 31 Oct 2012 09:52:47 +0000 (10:52 +0100)]
Updated Galician translations

11 years agoUpdated Marathi Translations
Sandeep Sheshrao Shedmake [Wed, 31 Oct 2012 05:09:01 +0000 (10:39 +0530)]
Updated Marathi Translations

11 years agoUpdated German translation
Wolfgang Stoeggl [Tue, 30 Oct 2012 18:45:59 +0000 (19:45 +0100)]
Updated German translation

11 years ago[l10n] Updated Estonian translation
Mattias Põldaru [Tue, 30 Oct 2012 13:38:09 +0000 (15:38 +0200)]
[l10n] Updated Estonian translation

11 years agoUpdated Slovenian translation
Matej Urbančič [Mon, 29 Oct 2012 21:33:43 +0000 (22:33 +0100)]
Updated Slovenian translation

11 years agoUpdated Lithuanian translation
Aurimas Černius [Mon, 29 Oct 2012 20:07:22 +0000 (22:07 +0200)]
Updated Lithuanian translation

11 years agowayland: update to work with stable libxkbcommon
Ran Benita [Fri, 26 Oct 2012 00:52:30 +0000 (02:52 +0200)]
wayland: update to work with stable libxkbcommon

libxkbcommon has had some changes to its API. However, it now has a
stable release (0.2.0), so this makes the necessary changes, and
replaces all uses of the deprecated API.

Signed-off-by: Ran Benita <ran234@gmail.com>
11 years agoUpdated Spanish translation
Daniel Mustieles [Mon, 29 Oct 2012 12:07:29 +0000 (13:07 +0100)]
Updated Spanish translation

11 years agoUpdated Indonesian translation
Andika Triwidada [Mon, 29 Oct 2012 11:45:22 +0000 (18:45 +0700)]
Updated Indonesian translation

11 years agoUpdated gujarati file
Sweta Kothari [Mon, 29 Oct 2012 10:58:12 +0000 (16:28 +0530)]
Updated gujarati file

11 years agoUpdated Serbian translation
Мирослав Николић [Mon, 29 Oct 2012 10:48:33 +0000 (11:48 +0100)]
Updated Serbian translation

11 years agoFix compiling for win32
Hib Eris [Sun, 28 Oct 2012 19:05:23 +0000 (20:05 +0100)]
Fix compiling for win32

https://bugzilla.gnome.org/show_bug.cgi?id=687066

11 years agotexthandles: Add an extra style class to the cursor-mode handle
Carlos Garnacho [Fri, 26 Oct 2012 15:57:03 +0000 (17:57 +0200)]
texthandles: Add an extra style class to the cursor-mode handle

Themes may want to render handles differently depending on whether
the widget is in selection mode (2 handles enclosing a selection) or
cursor mode (one handle pointing out the insertion cursor).

11 years agotexthandles: set input shape on handles' window
Carlos Garnacho [Fri, 26 Oct 2012 15:48:26 +0000 (17:48 +0200)]
texthandles: set input shape on handles' window

This improves both interaction and theming, as it allows
arbitrary handle shapes while just being draggable from
the visible areas.

This way themes can set up handles with the hotspot visually
displaced from the horizontal center, as long as the hotspot
lies centered in the image/svg asset.

11 years agotexthandle: Fix shape setup on non-composited environments
Carlos Garnacho [Fri, 26 Oct 2012 15:46:40 +0000 (17:46 +0200)]
texthandle: Fix shape setup on non-composited environments

The check on the handle to be drawn on the mask was based on the yet to
be set priv->windows pointers, pass explicitly the handle position to
have the shape correctly initialized on non-composited environments

11 years agoAssamese translation updatedas.po
Nilamdyuti Goswami [Fri, 26 Oct 2012 08:53:33 +0000 (14:23 +0530)]
Assamese translation updatedas.po

11 years agoFix typo from f2ab3af
John Ralls [Thu, 25 Oct 2012 17:58:47 +0000 (10:58 -0700)]
Fix typo from f2ab3af

11 years agoUpdated Latvian translation
Rūdolfs Mazurs [Wed, 24 Oct 2012 12:14:37 +0000 (15:14 +0300)]
Updated Latvian translation

11 years agoconfigure.ac: Look for wayland-client 1.0.0
Kristian Høgsberg [Wed, 24 Oct 2012 02:37:02 +0000 (22:37 -0400)]
configure.ac: Look for wayland-client 1.0.0

11 years agoAdd GDK_VERSION_3_8
Matthias Clasen [Tue, 23 Oct 2012 21:36:17 +0000 (17:36 -0400)]
Add GDK_VERSION_3_8

This macro is needed for the new AVAILABLE_IN and DEPRECATED_IN
macros.

11 years agoUpdated Greek translation
Tom Tryfonidis [Tue, 23 Oct 2012 16:54:51 +0000 (19:54 +0300)]
Updated Greek translation

11 years agoBump version
Matthias Clasen [Tue, 23 Oct 2012 00:45:02 +0000 (20:45 -0400)]
Bump version

11 years ago3.7.0
Matthias Clasen [Tue, 23 Oct 2012 00:44:24 +0000 (20:44 -0400)]
3.7.0

11 years agoBe robust against unrealized windows in GtkWindowAccessible
Matthias Clasen [Mon, 22 Oct 2012 23:42:41 +0000 (19:42 -0400)]
Be robust against unrealized windows in GtkWindowAccessible

Based on a patch by Albert Astals Cid,
https://bugzilla.gnome.org/show_bug.cgi?id=686152

11 years agoForgotten fixup
Matthias Clasen [Mon, 22 Oct 2012 23:41:14 +0000 (19:41 -0400)]
Forgotten fixup

This was meant to be included in the previous commit :-(

11 years agotreeview: Add support for styling the dragged header
Stefano Facchini [Thu, 27 Sep 2012 17:07:42 +0000 (19:07 +0200)]
treeview: Add support for styling the dragged header

https://bugzilla.gnome.org/show_bug.cgi?id=684980

11 years agotreeview: Move the dragged header in the headers window
Stefano Facchini [Thu, 27 Sep 2012 16:08:44 +0000 (18:08 +0200)]
treeview: Move the dragged header in the headers window

https://bugzilla.gnome.org/show_bug.cgi?id=684980

11 years agoUse named union for _GtkSymbolicColor in gtk/gtksymboliccolor.c
Matthias Clasen [Mon, 22 Oct 2012 22:58:32 +0000 (18:58 -0400)]
Use named union for _GtkSymbolicColor in gtk/gtksymboliccolor.c

Patch by Richard Lloyd,
https://bugzilla.gnome.org/show_bug.cgi?id=686366

11 years agoscrollbar: remove unused variable
Cosimo Cecchi [Mon, 22 Oct 2012 22:39:42 +0000 (18:39 -0400)]
scrollbar: remove unused variable

My bad.

11 years agonotebook: return TRUE for drag-motion event when over tabs
Cosimo Cecchi [Thu, 20 Sep 2012 00:56:26 +0000 (20:56 -0400)]
notebook: return TRUE for drag-motion event when over tabs

The GtkNotebook drag-motion event handler may install a timeout when
hovering over a tab, in order to switch to it.
On the other hand it's desirable for applications to use the empty tab
area as a drop target, so the drag-motion handler returns FALSE
(also in case it installs the switch tab timeout), as explained in
https://bugzilla.gnome.org/show_bug.cgi?id=350665.

Unfortunately, applications can use the tab label widget (or a child
of it) as a different drop target area, and install their own
drag-motion handler there.
In this scenario, the timeout will still be installed by GtkNotebook's
handler, but since it returns FALSE, it will never get the matching
drag-leave event, causing it to trigger also when the mouse pointer
moved elsewhere before it expired.

Fix this by returning TRUE from drag-motion when the event is over a
tab. Note that this makes automatic tab switching not work anymore when
drag and drop is handled in the tab label widget; applications are
expected to also handle tab switching if desired in such a case.

https://bugzilla.gnome.org/show_bug.cgi?id=684415

11 years agonotebook: consolidate code to remove the switch tab timer
Cosimo Cecchi [Wed, 19 Sep 2012 21:01:00 +0000 (17:01 -0400)]
notebook: consolidate code to remove the switch tab timer

https://bugzilla.gnome.org/show_bug.cgi?id=684415

11 years ago[l10n] Updated Estonian translation
Mattias Põldaru [Mon, 22 Oct 2012 19:01:01 +0000 (22:01 +0300)]
[l10n] Updated Estonian translation

11 years agoscale: update style properties on GtkRange at init
Cosimo Cecchi [Mon, 22 Oct 2012 17:13:11 +0000 (13:13 -0400)]
scale: update style properties on GtkRange at init

Same fix as in ef027c93d49e2e31ccb8c181d0096c705c141906, but for
the GtkScale subclass of GtkRange.

https://bugzilla.gnome.org/show_bug.cgi?id=686280

11 years agoscrollbar: update style properties on GtkRange at init
Cosimo Cecchi [Mon, 22 Oct 2012 15:42:55 +0000 (11:42 -0400)]
scrollbar: update style properties on GtkRange at init

GtkScrollbar used to rely on style-updated being emitted every time
after the widget was created in order to set the right values from its
style properties on GtkRange.
Nowadays we try to be smarter and avoid emitting style-updated at
creation time, so we need to manually initialize the GtkRange values.

This fixes a regression from 35e36b9fe506143f722a00f26a86143f9c95b740.

https://bugzilla.gnome.org/show_bug.cgi?id=686280

11 years agoFix out of source tree building for gdk/broadway
Jean Parpaillon [Mon, 15 Oct 2012 14:16:51 +0000 (14:16 +0000)]
Fix out of source tree building for gdk/broadway

This patch fix compilation when building out of source tree (gdk/broadway dir).

Signed-off-by: Colin Walters <walters@verbum.org>
11 years agoUpdated Slovak translation
Pavol Šimo [Sat, 20 Oct 2012 18:46:03 +0000 (19:46 +0100)]
Updated Slovak translation

11 years agoUpdated Galician translations
Fran Diéguez [Sat, 20 Oct 2012 11:09:24 +0000 (13:09 +0200)]
Updated Galician translations

11 years agodocs: fix typo on 'Getting Started with Gtk+' page
Tim-Philipp Müller [Sat, 20 Oct 2012 10:25:02 +0000 (11:25 +0100)]
docs: fix typo on 'Getting Started with Gtk+' page

11 years agoUpdated Lithuanian translation
Aurimas Černius [Fri, 19 Oct 2012 17:45:59 +0000 (20:45 +0300)]
Updated Lithuanian translation

11 years agoUpdate Czech translation
Petr Kovar [Fri, 19 Oct 2012 11:25:50 +0000 (13:25 +0200)]
Update Czech translation

11 years agowayland: Update to reflect protocol changes
Scott Moreau [Wed, 17 Oct 2012 02:38:41 +0000 (20:38 -0600)]
wayland: Update to reflect protocol changes

11 years agoUpdate Galician translations
Fran Diéguez [Thu, 18 Oct 2012 08:53:55 +0000 (10:53 +0200)]
Update Galician translations

11 years agoUpdated Slovenian translation
Matej Urbančič [Thu, 18 Oct 2012 06:16:39 +0000 (08:16 +0200)]
Updated Slovenian translation

11 years agoFix duplicate columns in filechooser entry completion
Matthias Clasen [Thu, 18 Oct 2012 01:21:09 +0000 (21:21 -0400)]
Fix duplicate columns in filechooser entry completion

This partically reverts commit
331bba1ad6c5d9535fc5a827b91019f5824e25a7, which broke documented
behaviour.

11 years agomenubutton: don't refer to the non-existant menu property
Cosimo Cecchi [Wed, 17 Oct 2012 21:02:53 +0000 (17:02 -0400)]
menubutton: don't refer to the non-existant menu property

7c6454246e7fa181d9e16f3520bc6d3765168f40 removed the property, but
forgot to change the name in a g_object_notify().

11 years agoUpdated Serbian translation
Мирослав Николић [Wed, 17 Oct 2012 20:24:36 +0000 (22:24 +0200)]
Updated Serbian translation

11 years agoUpdated Indonesian translation
Andika Triwidada [Wed, 17 Oct 2012 16:53:04 +0000 (23:53 +0700)]
Updated Indonesian translation

11 years agoUpdated Spanish translation
Daniel Mustieles [Wed, 17 Oct 2012 16:05:57 +0000 (18:05 +0200)]
Updated Spanish translation

11 years agolevel-bar: Fix typo
Stefano Facchini [Wed, 17 Oct 2012 14:09:08 +0000 (16:09 +0200)]
level-bar: Fix typo

11 years agoscrolledwindow: set GDK_EXPOSURE_MASK on the overshoot window
Cosimo Cecchi [Wed, 17 Oct 2012 04:59:59 +0000 (00:59 -0400)]
scrolledwindow: set GDK_EXPOSURE_MASK on the overshoot window

Currently we use gtk_style_context_set_background() when the state flags
change in order to propagate the background color to the overshoot
window, but this is actually only needed because the window doesn't get
expose events, since we always draw a full background in draw().
This also fixes some problems when the GdkWindow of the scrolled
window's child is composited, as seen in oxygen-gtk3.

https://bugzilla.gnome.org/show_bug.cgi?id=686265

11 years agolevel-bar: add support for RTL locales
Stefano Facchini [Tue, 18 Sep 2012 12:19:01 +0000 (14:19 +0200)]
level-bar: add support for RTL locales

https://bugzilla.gnome.org/show_bug.cgi?id=684288

11 years agolevel-bar: add an "inverted" property like GtkProgressBar
Stefano Facchini [Tue, 18 Sep 2012 12:05:45 +0000 (14:05 +0200)]
level-bar: add an "inverted" property like GtkProgressBar

https://bugzilla.gnome.org/show_bug.cgi?id=684288