summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Avoid a crash in the pixbuf engineMatthias Clasen2013-12-011-1/+1
| | | | | This is fallout from e4c83bbfdb60fdfe0bae207b1ddae295dc267a23, which removed a necessary check.
* Remove an unused struct memberMatthias Clasen2013-12-011-4/+0
| | | | AsyncFuncData.folder was entirely unused. Drop it.
* pixbuf-engine: Clean up error conditions and destructorsEmmanuele Bassi2013-11-221-36/+23
| | | | | | | Simplify the error checks and move all common behaviour into a utility function. https://bugzilla.gnome.org/show_bug.cgi?id=712536
* pixbuf-engine: Improve ThemePixbuf clean up functionsEmmanuele Bassi2013-11-222-2/+20
| | | | | | | Make theme_pixbuf_destroy() NULL-safe like g_free(), and add a clear function in the spirit of the g_clear_* family of functions. https://bugzilla.gnome.org/show_bug.cgi?id=712536
* Themes with specified gap images but no actual file crash.Jehan2013-11-221-0/+21
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=712536
* Prevent a possible warningMatthias Clasen2013-11-091-4/+4
| | | | | | | | | | | | | | When doing updates, gtk_progress_bar_real_update can calculate compute a negative value for activity_pos. If the allocation width is smaller than (xthickness + size), pbar->activity_pos will be set to a negative value and this will cause the drawing code to emit a warning about coordinates being out of bounds. Prevent this by keeping the value at least 0. https://bugzilla.gnome.org/show_bug.cgi?id=701751
* Win32: Honor the geometry when GDK_HINT_MAX_SIZE is set.Cody Russell2013-10-311-2/+6
| | | | | | | | The MINMAXINFO struct was being populated based upon geometry hints when GDK_HINT_MAX_SIZE flag was enabled, then promptly having its values blown away with default values. https://bugzilla.gnome.org/show_bug.cgi?id=711110
* bgo#710936 - Fix typo in Japanese translationFederico Mena Quintero2013-10-281-1/+1
|
* Bump versionMatthias Clasen2013-10-111-2/+2
|
* 2.24.222.24.22Matthias Clasen2013-10-111-0/+11
|
* Fix buildMatthias Clasen2013-10-061-2/+2
|
* Support automake 1.14Dominique Leuenberger2013-10-051-0/+4
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=709293
* Actually look in /usr/share/gtk-2.0 for default rc filesMatthias Clasen2013-09-301-1/+1
| | | | I managed to forget the 'share' component of the path.
* Updated Hungarian translationBalázs Úr2013-09-221-172/+175
|
* Post-release version bumpMatthias Clasen2013-09-171-2/+2
|
* 2.24.212.24.21Matthias Clasen2013-09-172-2/+32
|
* Disable filechooser testMatthias Clasen2013-09-171-3/+3
| | | | It does't work in make distcheck.
* Make filechooser test passMatthias Clasen2013-09-171-1/+1
|
* Updated Thai translation.Theppitak Karoonboonyanan2013-09-111-1826/+1609
|
* [gdkcairo] Fix color premultiplicationBehdad Esfahbod2013-09-091-1/+1
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=513812
* Fix config dir <> data dir confusionMatthias Clasen2013-09-061-0/+4
| | | | | | We added code to look for gtkrc in system config dirs, and then proceeded to move it to /usr/share/gtk-2.0 :-(. So, look in that location as well.
* Fix missing return from quartz_filter_keypressEk Kato2013-08-291-2/+2
| | | | (cherry picked from commit 0ae728e2424406f1dab3878a31942d9dd8862c1a)
* combobox: Add missing Since tags for entry constructorsDavid King2013-08-221-0/+4
|
* Bug 705750 Quartz input method doesn't work correctly for Chinese charactersEk Kato2013-08-165-41/+114
|
* Fix a crasher that was introduced by 9ca802161ec6387419d160f21fe477008a567477Cody Russell2013-08-151-1/+1
| | | | due to preedit_str being initialized as an empty string rather than NULL.
* Bug 705182 Reset Cocoa IME state when immodule is resetMichael Hutchison2013-08-151-0/+34
| | | | | | | Commits the pre-edit string on receipt of focus_out and reset commands. Patch refinements by Cody Russell <bratsche@gnome.org> and Ek Kato <ek.kato@gmail.com>
* win32: avoid assertion when creating a GdkPixmap failsAleksander Morgado2013-08-141-6/+11
| | | | | | | | | | | | | | | | | If gdk_pixmap_new() fails (e.g. CreateDIBSection() failure) we end up g_object_unref()-ing the temporary GdkPixmap, and we do this before having set the drawable_impl->hdc and drawable_impl->hdc_count. Now, this could have just been a couple of warnings being dumped, but instead it really crashes the application because in win32-specific GdkDrawable's finalize() we assert if drawable_impl->hdc_count is not zero: Gdk:ERROR:gdkdrawable-win32.c:2047:_gdk_win32_drawable_finish: assertion failed: (impl->hdc_count == 0) In order to avoid this crash, we make sure we don't decrement the hdc_count in the GdkPixmap's finalize() unless we have it set first (i.e. we avoid hdc_count going down to -1). https://bugzilla.gnome.org/show_bug.cgi?id=699236
* Bug 705181 Annoying beep on arrow keysEk Kato2013-08-131-5/+2
| | | | Original patch was a bit excessive, just needed to not forward the command.
* Bug 701571 NSApp doesn't notice NSWindow destructionJohn Ralls2013-08-111-16/+26
| | | | | | | Part 1 of the fix; part 2 awaits Glib developer approval (see https://bugzilla.gnome.org/show_bug.cgi?id=704374) and is more correctly associated with https://bugzilla.gnome.org/show_bug.cgi?id=674108
* Reset cursor when mouse leaves a toplevel window.Daniel Sabo2013-08-081-1/+4
| | | | Fixes https://bugzilla.gnome.org/show_bug.cgi?id=692548
* Bug 705181: Fix annoying beeping introduced by Mac IMEMichael Hutchinson2013-07-301-1/+5
| | | | NSTextInputClient should not chain unhandled commands to super
* win32: always recreate the cairo surface if requested to do soAleksander Morgado2013-07-301-11/+33
| | | | | | | | | | | | When _gdk_windowing_create_cairo_surface() gets called, we should always create a fully new cairo surface, instead of just referencing the available one, which may already be finished (i.e. in CAIRO_STATUS_SURFACE_FINISHED state). A new user_data key is added to the surface to explicitly release the acquired DC when the surface is destroyed, independent to the user_data key added to clear the impl->user_data pointer. https://bugzilla.gnome.org/show_bug.cgi?id=695636
* Limit includes in gtksignal.hMatthias Clasen2013-07-291-1/+3
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=536767
* X11: Fix the cut-off for too large windowsMatthias Clasen2013-07-291-7/+7
| | | | | | | We can only handle 2^15-1 pixels, not 2^16-1. Pointed out by Morten Welinder. http://bugzilla.gnome.org/show_bug.cgi?id=698758
* Bug Bug 703789 - gtk.immodules still referencedJohn Ralls2013-07-2812-27/+22
| | | | After switch to immodules.cache
* Bug 701332 - Patch for minor glitch in NSTextInputEk Kato2013-07-271-2/+8
|
* imquartz: Fallback to slave IM context if no NSEvent exists.Takuro Ashie2013-07-271-4/+7
| | | | | | | | | | When an application translates a key event and drop its native event before passing to imquartz, it can't recognize the NSEvent. On this case imquartz doesn't emit any signals such as "commit" signal so that the application doesn't insert any text. To avoid no response, at least imquartz should fallback to slave GtkIMContextSimple. https://bugzilla.gnome.org/show_bug.cgi?id=694273#c27
* build: fix --disable-manMarc-Antoine Perennou2013-07-261-1/+1
| | | | | | | | The third argument to AC_ARG_ENABLE is the action done when the argument is specified, not when it is enabled. This resulted in --disable-man enabling those. Signed-off-by: Marc-Antoine Perennou <Marc-Antoine@Perennou.com>
* quartz: fix crashes that sometimes happened on the first key pressMichael Natterer2013-07-251-1/+13
| | | | | | | because nobody has called gdk_keymap_get_for_display() before. Now we simply make sure the default display's keymap exists if somebody calls any gdk_keymap_*() function with a NULL keymap (which is an allowed value).
* gtkdialog - more informative warning when ordering unexisting response_idJehan2013-07-211-2/+10
| | | | (cherry picked from commit 0e4d9b7793c16fd98b396ffe6b8db48b8e09f1d2)
* Define INITGUID only for mingw.org compilerHib Eris2013-07-131-0/+4
| | | | | | | | Defining INITGUID causes a build failure with mingw-w64 > r5589. https://bugzilla.gnome.org/show_bug.cgi?id=699673 (cherry picked from commit 6a71820ff399b32c4a95e7b17979513783c4e052)
* 2.24.202.24.20Matthias Clasen2013-07-042-2/+23
|
* Let query utilities update the cache file directlyMatthias Clasen2013-07-042-38/+75
| | | | | This is much more convenient for packagers than having to redirect the output into the cache file, and much less error-prone.
* Move the module cache files below libdirMatthias Clasen2013-07-044-65/+70
| | | | | | | | These files contain architecture-dependent paths, and thus placing them into sysconfdir causes unnecessary hassle. Now the immodule cache file is looked for in libdir/gtk-2.0/2.10.0/immodules.cache. Belated backport of a change that was done in the run-up to 3.0.
* Look for rc files in system config dirs as wellMatthias Clasen2013-07-041-1/+11
| | | | | This will allow us to ship package defaults in /usr instead of /etc.
* Updated Polish translationPiotr Drąg2013-06-261-219/+224
|
* printing: List Avahi printersMarek Kasik2013-06-256-152/+1205
| | | | | | | | | | | | | | | | Show printers advertised by avahi on local network. CUPS backend now looks for _ipps._tcp and _ipp._tcp services offered by avahi. If it finds such a service (printer) it requests its attributes through IPP_GET_PRINTER_ATTRIBUTES ipp request and adds it to the list of printers. Such printer behaves like a remote printer then. If an avahi printer is a default printer then it is considered default by the backend only if there is no local or remote default printer. This functionality is enabled when building Gtk+ with CUPS 1.6 or later because it replaces browsing protocol removed in CUPS 1.6. https://bugzilla.gnome.org/show_bug.cgi?id=702455
* GtkFileChooser: Missing hour on today's files for it&hu locales on Win32Valerio Messina2013-06-202-4/+4
| | | | | | | | | | | | | The file chooser does not show the hours for files modified today when using it & hu locales on Win32. This is due to the %k format conversion for strftime() that is in Olson's Timezone Package, which msvcrt.dll doesn't have. Cygwin has it, though. So for now, we just use a %H format. While %k would render six-in-the-morning as " 6" (6 preceded by a space), %H will render it as "06". https://bugzilla.gnome.org/show_bug.cgi?id=560337
* quartz: remove check for keymap changes from update_keymap()Michael Natterer2013-06-201-218/+206
| | | | | The function is now only called when the keymap has actually changed. bug #698183.
* quartz: update the keymap only if the input method changedMichael Natterer2013-06-201-8/+21
| | | | | | and not on each keystroke, which for some IMs apparently caused a full update on each keystroke, not just a check for changes. Patch from Takuro Ashie, bug #698183.
View Lorry Controller Status