]>
Pileus Git - ~andy/gtk/log
Benjamin Otte [Thu, 5 May 2011 15:30:58 +0000 (17:30 +0200)]
label: Compute label offset properly for wrapped justified labels
label offsets were computed wrong for wrapped or ellipsized labels that
were right- or center-justified and had excess space available.
Benjamin Otte [Thu, 5 May 2011 03:16:38 +0000 (05:16 +0200)]
reftests: Fix filename in makefile
Benjamin Otte [Thu, 5 May 2011 03:16:09 +0000 (05:16 +0200)]
reftests: Add a test that ensures toplevels and popups render identical
Benjamin Otte [Thu, 5 May 2011 03:13:34 +0000 (05:13 +0200)]
reftests: Ensure we got a draw event before quitting the main loop
Benjamin Otte [Thu, 5 May 2011 02:29:56 +0000 (04:29 +0200)]
reftests: Add a hack to make toplevels work in various WMs
If we have a toplevel, and not a popup window, do wait an additional
0.5s to give the WM/server enough time to actually create the window.
This is a hack and there should be a better solution. But it works.
Please use POPUP windows for tests unless the test must use toplevel
windows.
Benjamin Otte [Thu, 5 May 2011 02:27:38 +0000 (04:27 +0200)]
reftests: Plug a leak
Benjamin Otte [Wed, 4 May 2011 23:09:45 +0000 (01:09 +0200)]
reftests: Add test for box expand handling
Benjamin Otte [Wed, 4 May 2011 18:12:44 +0000 (20:12 +0200)]
reftests: Add a test for grid expand flags
Javier Jardón [Mon, 2 May 2011 15:13:50 +0000 (16:13 +0100)]
gtk/gtktrayicon-x11.c: Use GdkRGBA instead GdkColor
Matthias Clasen [Thu, 5 May 2011 11:46:03 +0000 (07:46 -0400)]
GtkButtonBox: add a non-homogeneous child property
This makes it easier to add things like lock buttons in
dialogs. Part of #626457.
Matthias Clasen [Thu, 5 May 2011 11:26:55 +0000 (07:26 -0400)]
Fix box-packing test
Matthias Clasen [Thu, 5 May 2011 02:58:32 +0000 (22:58 -0400)]
Expand box-packing reftest
Compare vexpand behavior, too.
Matthias Clasen [Thu, 5 May 2011 00:38:38 +0000 (20:38 -0400)]
Add a box-packing reftest
This test compares GtkHBox expand handling with GtkGrid+GtkWidget::hexpand,
to ensure that replacing GtkHBox by GtkGrid works as intended.
Daniel Mustieles [Wed, 4 May 2011 20:46:27 +0000 (22:46 +0200)]
Updated Spanish translation
Javier Jardón [Mon, 2 May 2011 16:42:18 +0000 (17:42 +0100)]
gdk/x11/gdkcursor-x11.c: Use GdkRGBA instead GdkColor
Kristian Rietveld [Wed, 4 May 2011 05:31:39 +0000 (07:31 +0200)]
GdkWindow: remove unused variable
Matthias Clasen [Tue, 3 May 2011 23:35:48 +0000 (19:35 -0400)]
Convert the commit message into a README
Not everybody thinks git grep is a great doc reader...
Carlos Garnacho [Tue, 3 May 2011 22:01:10 +0000 (00:01 +0200)]
GdkWindow: Fix gdk_window_set_[device_]cursor() issues with root/foreign windows
It could be the case that gdk_window_set_cursor() is called on
pointers not yet known to the device tracking code in GdkDisplay,
so update the cursor on all master pointers.
The code actually updating the cursor for the given window has
been refactored out to gdk_window_set_cursor_internal(), used
in gdk_window_set_device_cursor() as well, which makes it handle
root/foreign windows too.
https://bugzilla.gnome.org/show_bug.cgi?id=649313
Colin Walters [Tue, 3 May 2011 16:33:31 +0000 (12:33 -0400)]
gtkdnd: Add API that takes GIcon
This will be useful for me to port gnome-panel to gnome-menus 4, which
uses GAppInfo, which in turn returns icon references as GIcon.
https://bugzilla.gnome.org/show_bug.cgi?id=649295
Colin Walters [Tue, 3 May 2011 15:04:56 +0000 (11:04 -0400)]
gtkimageprivate: Drop unused theme_change_id
Benjamin Otte [Tue, 3 May 2011 12:03:25 +0000 (14:03 +0200)]
reftests: Add a --output option to the test runner
This argument specifies where to dump images instead of /tmp. It's not
hooked up to the test runner, so that one will continue to dump into
/tmp.
Benjamin Otte [Tue, 3 May 2011 10:05:45 +0000 (12:05 +0200)]
reftests: Add test for low mnemonics
Low mnemonics were invisible previously. This tests the fix from
9bb145a677c3c9fa3266ae6999414f441016d93f
Benjamin Otte [Tue, 3 May 2011 09:18:04 +0000 (11:18 +0200)]
reftests: Test that nonresizable windows get the right size
We want the default size, not the minimum one. This tests the fixes from
4f23dc34be1fd497cce84962798f7578e78a32c9
Benjamin Otte [Fri, 22 Apr 2011 02:56:03 +0000 (04:56 +0200)]
tests: Add a test for small ellipsized labels
Tests the fix from
bf413928611e004ae3b5a9f73ff9c777cb7e8e70
Benjamin Otte [Wed, 20 Apr 2011 15:07:49 +0000 (17:07 +0200)]
reftests: Add grid-homogeneous test
Checks that the homogeneous properties of GtkGrid actually do what
they're supposed to.
Tests fixes from
8331b45a5fcbfb919a3bb02ccb93ac6d66c515b9
Benjamin Otte [Tue, 19 Apr 2011 20:15:24 +0000 (22:15 +0200)]
reftests: Add a test for correct entry text clipping in progress drawing
Tests for fixes in
da1f618b096973704bd0b7db071baa1444ef5c7b
Benjamin Otte [Sun, 17 Apr 2011 17:56:52 +0000 (19:56 +0200)]
reftests: Add label-sizing test
Checks that the size requests for labels are as they should be for
required and natural size given various combinations of wrap, ellipsize,
width-chars and max-width-chars.
See
http://mail.gnome.org/archives/gtk-devel-list/2011-April/msg00036.html
for the discussion spawning this test.
Benjamin Otte [Tue, 3 May 2011 07:44:45 +0000 (09:44 +0200)]
reftests: Add
Add a new test runner supposed to do a lot of generic tests. Run it like
this:
./gtk-reftest [OPTIONS] TESTFILE [TESTFILES...]
where FILE is a GtkBuilder ui file to run.
For a general test named "test", you want to have the following files:
1) test.ui
2) test.ref.ui
3) test.css (optional)
The test will then check that test.ui and test.ref.ui are rendered
identically with the provided css.
In detail, for every provided TESTFILE the test runner will:
1) Add the css to the default screen
2) Load the test.ui file and the test.ref.ui file
3) Grab the first GtkWindow subclass widget
4) gtk_widget_show() it and take a snapshot image of its contents into
a cairo surface.
5) Compare the two images to be bitwise identical. If they are not, a
diff image will be created hilighting the differences.
6) Save the images as png files to the output directory named:
- test.out.png (rendering of test.ui)
- test.ref.png (rendering of test.ref.ui)
- test.diff.png (optional, differences from step 5)
7) Fail the test if the two images are not bitwise identical
Credit for the idea of reftests goes to Mozilla and in particular David
Baron. For a larger introduction of why reftests are useful, see
http://weblogs.mozillazine.org/roc/archives/2008/12/reftests.html
Benjamin Otte [Sun, 24 Apr 2011 23:48:59 +0000 (01:48 +0200)]
label: Allow using more underline modes
Martin Pitt [Tue, 3 May 2011 07:30:49 +0000 (09:30 +0200)]
GdkWindow: Add missing (allow-none) annotations
Benjamin Otte [Mon, 2 May 2011 18:07:49 +0000 (20:07 +0200)]
accellabel: Remove leftover debug code
The disabling of the check was added in
63a0ab5e73a618e4fe907f69a1bd6a74cb399dc2
Ooops.
Benjamin Otte [Mon, 2 May 2011 16:36:14 +0000 (18:36 +0200)]
cellarea: Fix warning messages to display the right type
The warning displayed the type of the area instead of the cell
renderer's type before.
Javier Jardón [Mon, 2 May 2011 17:19:09 +0000 (18:19 +0100)]
tests/testgtk.c: Fix crash in cursor test
This was introduced in
commit
ac4c12ff60374e77d88f8e65fc9cb3b5603a193e
Benjamin Otte [Mon, 2 May 2011 14:49:03 +0000 (16:49 +0200)]
menuitem: Deprecate gtk_menu_item_set_right_justified()
And the getter, too: gtk_menu_item_get_right_justified() and the
corresponding property. Also make the only caller use the private
structure (as it did before the recent patches).
Benjamin Otte [Mon, 2 May 2011 14:47:06 +0000 (16:47 +0200)]
testgtk: right-justify menuitem by using expand/align flags
gtk_menu_item_set_right_justify() is about to be deprecated and this
thing should work. (Note: should, because I didn't implemented it yet.)
Benjamin Otte [Mon, 2 May 2011 14:46:06 +0000 (16:46 +0200)]
gtk-demo: Don't right-justify menu items
We don't want to demo things that are a bad idea (quoted from the docs).
Javier Jardón [Mon, 2 May 2011 14:42:54 +0000 (15:42 +0100)]
gtkbuilder.c: gtk_builder_value_from_string() also support GdkRGBA values
Javier Jardón [Mon, 2 May 2011 14:38:36 +0000 (15:38 +0100)]
docs: Remove documentation about deprecated and removed GtkText
Arnel Borja [Sun, 1 May 2011 05:13:45 +0000 (08:13 +0300)]
Bypass libtool for linking with libuuid on win32
Fixes a regression introduced in
07d49ee5 .
Libtool by default refuses to link static libraries into shared
libraries. In Windows, libuuid is however a static library and needed
for shared libgtk; as a work around, use "-Wl,-luuid" to pass the option
directly to the linker.
https://bugzilla.gnome.org/show_bug.cgi?id=642214
Javier Jardón [Mon, 2 May 2011 13:59:53 +0000 (14:59 +0100)]
gtkcalendar: Remove unused variable
Javier Jardón [Tue, 26 Apr 2011 15:14:52 +0000 (16:14 +0100)]
docs/reference/gtk/text_widget.sgml: Use GdkRGBA instead GdkColor
Javier Jardón [Tue, 26 Apr 2011 15:04:52 +0000 (16:04 +0100)]
docs/tools/widgets.c: Use GdkRGBA instead GdkColor
Abduxukur Abdurixit [Mon, 2 May 2011 06:44:33 +0000 (08:44 +0200)]
Added UG translation
Abduxukur Abdurixit [Mon, 2 May 2011 06:34:06 +0000 (08:34 +0200)]
Added UG translation
Kristian Rietveld [Sun, 1 May 2011 14:38:56 +0000 (16:38 +0200)]
tooltips: treat scroll events like key/button press instead of motion
This avoids tooltips from staying visible when scrolling using scroll
events. Suggested by Paul Davis.
Matthias Clasen [Sun, 1 May 2011 14:02:58 +0000 (10:02 -0400)]
GtkWidget: avoid an unused variable warning
Benjamin Otte [Sun, 1 May 2011 06:47:58 +0000 (08:47 +0200)]
window: Make non-resizable windows use default size for geometry hints
Previously, we were trying to size them by the default size, but then
setting the minimum size as the geometry hints' minimum and maximum
size.
Benjamin Otte [Sun, 1 May 2011 06:00:50 +0000 (08:00 +0200)]
accellabel: Only daw accelerator after natural size has been reached
Benjamin Otte [Sun, 1 May 2011 05:19:55 +0000 (07:19 +0200)]
window: Remove overeagerness with new size guessing
Don't use the guessed size when we are interested in the minimum size.
So now we can really shrink menubars.
This reverts parts of
08b2ac1d90b4f3dfa76d6a76fa04ca28c6b7ba12
Benjamin Otte [Sun, 1 May 2011 05:06:06 +0000 (07:06 +0200)]
menubar: Rewrite size allocation to take natural sizes into account
Yay for properly ellipsizing menubars!
Benjamin Otte [Sun, 1 May 2011 03:38:47 +0000 (05:38 +0200)]
menubar: Only query ipadding when we actually need it
Benjamin Otte [Sun, 1 May 2011 03:30:42 +0000 (05:30 +0200)]
menubar: Respect ipadding also in y direction
Benjamin Otte [Sun, 1 May 2011 02:18:53 +0000 (04:18 +0200)]
tests: Make testmenubars make all menubar items ellipsized
This reproduces a bug in Nautilus. And it allows to shrink things,
woohoo!
Benjamin Otte [Sun, 1 May 2011 02:15:42 +0000 (04:15 +0200)]
settings: Recompute styles when font name changes
Benjamin Otte [Sat, 30 Apr 2011 04:41:56 +0000 (06:41 +0200)]
menubar: Implement height-for-width / width-for-height
Pretty much just required hooking up the 2 vfuncs, the previous code
reorg did the rest of the work.
Benjamin Otte [Sat, 30 Apr 2011 04:34:58 +0000 (06:34 +0200)]
menubar: Actually query minimum and natural size of children
Previously, we only queried the minimum size and reported that as
natural size, too.
Benjamin Otte [Sat, 30 Apr 2011 04:31:01 +0000 (06:31 +0200)]
menubar: Query only the necessary size from the child
The get_preferred_size() function was compied form elsewhere. We should
really make that a common API.
Benjamin Otte [Sat, 30 Apr 2011 04:24:51 +0000 (06:24 +0200)]
menubar: Only compute the necessary size
There's no need to compute height when we're only interested in width
Benjamin Otte [Sat, 30 Apr 2011 04:16:18 +0000 (06:16 +0200)]
menubar: compute in advance wether to add toggle size
Benjamin Otte [Sat, 30 Apr 2011 03:56:25 +0000 (05:56 +0200)]
menubar: Change function prototype for size request function
This is in preparation for redoing that function to actually care about
natural sizes.
Benjamin Otte [Sat, 30 Apr 2011 03:50:17 +0000 (05:50 +0200)]
menubar: Remove unnecessary return_if_fail() in vfuncs
Benjamin Otte [Sat, 30 Apr 2011 03:49:20 +0000 (05:49 +0200)]
menubar: Remove useless function declaration
Benjamin Otte [Sat, 30 Apr 2011 03:42:00 +0000 (05:42 +0200)]
menubar: Remove unused variable
Benjamin Otte [Sat, 30 Apr 2011 03:31:05 +0000 (05:31 +0200)]
imagemenuitem: No need to query image size by default
Also makes the code look nicer, so woohoo
Benjamin Otte [Sat, 30 Apr 2011 03:10:55 +0000 (05:10 +0200)]
menubar: Compute proper size even when invisible
Fran Diéguez [Sat, 30 Apr 2011 21:52:35 +0000 (23:52 +0200)]
Updated galician translations
Matthias Clasen [Fri, 29 Apr 2011 20:50:17 +0000 (16:50 -0400)]
Fix the build
G_GNUC_FUNCTION is deprecated.
Colin Walters [Fri, 29 Apr 2011 14:32:56 +0000 (10:32 -0400)]
Drop the last remaining translated g_warning/g_error calls
http://mail.gnome.org/archives/desktop-devel-list/2011-March/msg00069.html
https://bugzilla.gnome.org/show_bug.cgi?id=648943
Colin Walters [Fri, 29 Apr 2011 14:24:21 +0000 (10:24 -0400)]
updateiconcache: Don't translate g_warning/g_error
Kalev Lember [Fri, 29 Apr 2011 11:56:01 +0000 (14:56 +0300)]
Disable gtk-update-icon-cache.exe.manifest with --enable-gtk2-dependency
When gtk-update-icon-cache building is suppressed, its manifest file
should also not be installed.
https://bugzilla.gnome.org/show_bug.cgi?id=648938
Kalev Lember [Fri, 29 Apr 2011 08:42:20 +0000 (11:42 +0300)]
perf: Use different marshaller prefix than in gtk
Avoids multiple symbol definitions in gtk/gtkmarshalers.c
and perf/marshalers.c when both objects are linked in.
https://bugzilla.gnome.org/show_bug.cgi?id=648926
Benjamin Otte [Fri, 29 Apr 2011 09:04:01 +0000 (11:04 +0200)]
notebook: Match horizontal and vertical code paths
I have no idea what this code does, but it looks better now. :o
Benjamin Otte [Fri, 29 Apr 2011 07:38:56 +0000 (09:38 +0200)]
notebook: Respect left border when allocating tab widgets
This mirrors the code for the other tab locations.
Benjamin Otte [Fri, 29 Apr 2011 07:38:23 +0000 (09:38 +0200)]
notebook: Respect padding when computing tabs for left/right
This mirrors the top/bottom code
Benjamin Otte [Fri, 29 Apr 2011 06:59:55 +0000 (08:59 +0200)]
Don't adjust for values that we have previously adjusted for
The padding had been taken into account by the computation of
priv->allocation, so don't compute it twice.
Benjamin Otte [Fri, 29 Apr 2011 06:48:36 +0000 (08:48 +0200)]
notebook: Use the right padding for inactive tabs
Benjamin Otte [Fri, 29 Apr 2011 01:27:02 +0000 (03:27 +0200)]
window: Invert check that was wrong
Benjamin Otte [Thu, 28 Apr 2011 20:11:55 +0000 (22:11 +0200)]
arrow: Make minimum size 5px, as opposed to the 15 it was before
Number chosen by staring out the window and counting visible stars.
This is necessary so that path bars don't underallocate the arrow widget
they use.
Benjamin Otte [Thu, 28 Apr 2011 20:04:31 +0000 (22:04 +0200)]
widget: Allow underallocation for scrollables
Also document this fact.
Benjamin Otte [Thu, 28 Apr 2011 20:04:02 +0000 (22:04 +0200)]
docs: Fix liststore docs
gtk-doc was very unhappy. It's not anymore
Benjamin Otte [Thu, 28 Apr 2011 19:25:29 +0000 (21:25 +0200)]
widget: Update underallocation warning
Also print the parent widget. As the parent's size_allocate
implementation is usually the culprit for this warning happening, it
makes sense to print it.
Benjamin Otte [Thu, 28 Apr 2011 18:27:40 +0000 (20:27 +0200)]
paned: Move showing of the handle from realize() to map()
Also only show the paned if it's actually supposed to be visible.
Benjamin Otte [Thu, 28 Apr 2011 12:53:45 +0000 (14:53 +0200)]
paned: Remove needless check
Kizito Birabwa [Thu, 28 Apr 2011 19:29:03 +0000 (20:29 +0100)]
Updated Luganda translation
Benjamin Otte [Thu, 28 Apr 2011 03:12:00 +0000 (05:12 +0200)]
widget: Warn when allocation for gtk_widget_size_allocate() is too small
Widgets may not be underallocated, it is the responsibility of the
caller to ensure this never happens.
Benjamin Otte [Wed, 27 Apr 2011 22:51:37 +0000 (00:51 +0200)]
notebook: Warn if builder file defines multiple tab widgets per page
Glade causes such files when a notebook page has no content. And it's
especially bad because the last tab widget wins, so the label displayed
in Glade will not be what you see when you run the app.
Giorgos Stephanakis [Wed, 27 Apr 2011 20:20:24 +0000 (23:20 +0300)]
l10n: Updated Greek translation for gtk+
Benjamin Otte [Wed, 27 Apr 2011 17:59:02 +0000 (19:59 +0200)]
recentchooser: Fix default size setting
Set the min content size on the scrolled window instead of using
gtk_widget_set_size_request() on a widget where it's not recognized.
Benjamin Otte [Tue, 26 Apr 2011 22:49:00 +0000 (00:49 +0200)]
window: Make comparison be the right direction
Messed something up there...
Benjamin Otte [Sun, 24 Apr 2011 23:34:27 +0000 (01:34 +0200)]
label: Don't crash when a builder file contains unknown attributes
Benjamin Otte [Wed, 20 Apr 2011 14:56:03 +0000 (16:56 +0200)]
grid: Fix homogeneuous properties being flipped
Benjamin Otte [Tue, 12 Apr 2011 11:35:10 +0000 (13:35 +0200)]
cssprovider: #include gio
We reference GFile...
Benjamin Otte [Tue, 26 Apr 2011 22:25:48 +0000 (00:25 +0200)]
tooltip: Use the actual size of the window we're going to show
It's easy by realizing the window first.
Don't try to guess the alorithm used by gtkwindow.c for sizing the
window, it might not be what we expect.
Benjamin Otte [Tue, 26 Apr 2011 22:05:01 +0000 (00:05 +0200)]
window: Be smarter about computing the default size
See the code comments for the reasoning behind this. After we don't
force a "guessed" minimum size for labels anymore, a lot of issues
started to surface that this patch attempts to fix. In particular:
1) Tooltips where wrapped as much as possible.
2) The recentchooser submenu displayed only ellipsize dots.
Benjamin Otte [Tue, 26 Apr 2011 16:42:54 +0000 (18:42 +0200)]
window: Split out size guessing function
This doesn't do anything special yet, but I'm about to change that.
Benjamin Otte [Tue, 26 Apr 2011 20:40:37 +0000 (22:40 +0200)]
window: Use int instead of uint consistently
Chun-wei Fan [Tue, 26 Apr 2011 17:02:07 +0000 (01:02 +0800)]
VS2010 support-update property sheet
Update the Gdk-Pixbuf 2.0 include path to be consistent with
autotools-based builds
Chun-wei Fan [Tue, 26 Apr 2011 16:53:16 +0000 (00:53 +0800)]
VS2008/2010 support: Update README.txt
-Make the GDK-Pixbuf requirements more clear
-Tell people about the usage of ATK-2.x
Chun-wei Fan [Tue, 26 Apr 2011 16:46:52 +0000 (00:46 +0800)]
Bug 643270: VS2010 Project Files (autotools)
-Update to distribute the VS2010 files.
-Added rules in Makefile.am's of GDK and GTK to fill in the
project/filter files templates with up-to-date source file
listings to simplify maintenace.
Any comments on the usage of the VS2010 files are welcome!
Chun-wei Fan [Tue, 26 Apr 2011 16:44:30 +0000 (00:44 +0800)]
Bug 643270: VS 2010 Project Files
These are the VS2010 Project files to compile GDK and GTK+, using
the Win32 backend, along with a brief README.txt explaining the
process.
The GDK and GTK project/filter files are templates that are filled
in during "make dist" with up-to-date source file listings as far
as possible, to simplify maintenance.
Comments on their usage are most welcome.