Alexander Larsson [Wed, 26 Dec 2012 21:34:53 +0000 (22:34 +0100)]
broadway: Move broadwayd to bin
Alexander Larsson [Wed, 26 Dec 2012 21:31:52 +0000 (22:31 +0100)]
broadway: During pointer grabs, send pointer events only to one client
Alexander Larsson [Sun, 23 Dec 2012 22:35:03 +0000 (23:35 +0100)]
broadway: Make daemon side server "gdk independent".
Alexander Larsson [Sun, 23 Dec 2012 22:05:05 +0000 (23:05 +0100)]
broadway: Proper handling of http port and display nr
Alexander Larsson [Sun, 23 Dec 2012 21:46:29 +0000 (22:46 +0100)]
broadway: Collect and document some TODOs
Alexander Larsson [Sun, 23 Dec 2012 21:42:41 +0000 (22:42 +0100)]
broadway: Track windows for client and kill on client exit
Alexander Larsson [Sun, 23 Dec 2012 21:27:20 +0000 (22:27 +0100)]
broadway: Rename gdkbroadway-server.c to broadway-server.c
This way the broadway-*.c namespace is for the daemon and gdk* for the
gdk lib.
Alexander Larsson [Sun, 23 Dec 2012 21:24:21 +0000 (22:24 +0100)]
broadway: rename broadway-server to broadwayd
The "server" part is too confusing, there are to many servers around.
Alexander Larsson [Sun, 23 Dec 2012 21:12:08 +0000 (22:12 +0100)]
broadway: Drop libbroadway.la
This is not needed anymore, its just used by the server.
Alexander Larsson [Sun, 23 Dec 2012 21:05:02 +0000 (22:05 +0100)]
broadway: rename broadway.[ch] to broadway-output.[ch]
Alexander Larsson [Sun, 23 Dec 2012 20:58:20 +0000 (21:58 +0100)]
broadway: disconnect in idle when needed
This makes the ownership cleaner
Alexander Larsson [Fri, 21 Dec 2012 23:02:55 +0000 (00:02 +0100)]
broadway: Initial version of separate broadway server
This kinda works but is very rudimentary
Alexander Larsson [Thu, 20 Dec 2012 13:20:20 +0000 (14:20 +0100)]
broadway: Add OPs top protocol
Alexander Larsson [Thu, 20 Dec 2012 12:40:31 +0000 (13:40 +0100)]
broadway: Add enum for even types
Alexander Larsson [Thu, 20 Dec 2012 12:07:29 +0000 (13:07 +0100)]
broadway: Break out broadway protocol in its own header
Matthias Clasen [Thu, 27 Dec 2012 17:38:43 +0000 (12:38 -0500)]
docs: Restart a Deprecated section
We've accumulated enough wholly-deprecated objects again
that it makes sense to move them to their own section.
Matthias Clasen [Thu, 27 Dec 2012 17:15:05 +0000 (12:15 -0500)]
Another trivial sorting fix
Matthias Clasen [Thu, 27 Dec 2012 17:11:12 +0000 (12:11 -0500)]
Rename gtkcolorswatchaccessible.h
Call it gtkcolorswatchaccessibleprivate.h. This matches what we're
doing for non-public widget headers in gtk/.
Matthias Clasen [Thu, 27 Dec 2012 07:10:51 +0000 (02:10 -0500)]
Drop separate scrollbar accessible implementation
It didn't really add anything to the range accessible implementation.
Matthias Clasen [Thu, 27 Dec 2012 17:02:26 +0000 (12:02 -0500)]
Add a test for deriving accessible implementations
This is a trivial example. Just check that we can derive
from GtkButtonAccessible, and have a GtkButton subclass
use the derived accessible implementation.
Matthias Clasen [Tue, 30 Oct 2012 15:59:02 +0000 (11:59 -0400)]
Install gtk-a11y.h
We add a separate gtk-a11y.h single-include header for
them. This header will work much the same as gtkx.h. It
will be installed in /usr/include/gtk-3.0/gtk, but you
have to include it separately.
Matthias Clasen [Thu, 27 Dec 2012 06:46:26 +0000 (01:46 -0500)]
Install a11y headers
We install the headers for the a11y implementations in
/usr/include/gtk-3.0/gtk/a11y/.
Matthias Clasen [Tue, 30 Oct 2012 15:39:38 +0000 (11:39 -0400)]
Add single-include guards to a11y headers
More preparation for installing these headers.
Matthias Clasen [Thu, 27 Dec 2012 06:21:11 +0000 (01:21 -0500)]
Fix inter-header includes
Since we are going to install these headers soon, we need
to make their mutual includes work in the installed location
as well. Also, avoid including individual gtk headers, to
avoid trouble with single-include guards.
Matthias Clasen [Thu, 27 Dec 2012 06:14:15 +0000 (01:14 -0500)]
More gail removal
Rename GailMisc to GtkMiscImpl.
Matthias Clasen [Thu, 27 Dec 2012 06:12:35 +0000 (01:12 -0500)]
Rename and clean up gailutil
Matthias Clasen [Thu, 27 Dec 2012 06:06:30 +0000 (01:06 -0500)]
Some renaming
Move gail.h to gtkaccessibility.h, and libgail to libgtka11y.
Matthias Clasen [Tue, 30 Oct 2012 14:21:44 +0000 (10:21 -0400)]
Cosmetics
Replace GAIL in top comments by 'GTK+ - accessibility implementations'.
Matthias Clasen [Thu, 27 Dec 2012 06:04:46 +0000 (01:04 -0500)]
Make accessible implementations public
This commit exposes the get_type() functions and standard
headers for accessible implementations. This makes it possible
to derive from the GTK accessible implementations without
GType magic tricks. This is necessary, because we require the
a11y type hierarchy to be parallel to the widget type hierarchy.
So, if you derive a widget and need to adjust its a11y implementation,
you have to be able to derive its accessible implementation.
This commit probably exposes more than is absolutely necessary,
it also exposes accessibles of widgets that are unlikely candidates
for deriving from.
Matthias Clasen [Tue, 30 Oct 2012 15:58:47 +0000 (11:58 -0400)]
Trivial sorting fix
Fran Diéguez [Mon, 24 Dec 2012 20:55:07 +0000 (21:55 +0100)]
Updated Galician translations
Paolo Borelli [Sun, 23 Dec 2012 15:34:27 +0000 (16:34 +0100)]
Paint GtkFrame background correctly when there is a label
Kjartan Maraas [Sat, 22 Dec 2012 19:52:49 +0000 (20:52 +0100)]
Updated Norwegian bokmål translation
Benjamin Otte [Thu, 20 Dec 2012 18:51:06 +0000 (19:51 +0100)]
roundedbox: Fix copy-paste error
y coordinates don't care about the right value, they want bottom.
Benjamin Otte [Thu, 20 Dec 2012 18:53:04 +0000 (19:53 +0100)]
testgtk: Remove GtkAlignment usage from progressbar test
We can use gtk_widget_set_[hv]align here.
Cosimo Cecchi [Fri, 21 Dec 2012 16:52:03 +0000 (17:52 +0100)]
aboutdialog: ensure the credits page uses the "view" style class
Since not every theme renders a background for a GtkViewport (and
Adwaita master doesn't), ensure the grid+viewport we use to emulate a
text view here uses the "view" style class.
Kjartan Maraas [Fri, 21 Dec 2012 15:12:10 +0000 (16:12 +0100)]
Updated Norwegian bokmål translation
Paolo Borelli [Fri, 21 Dec 2012 12:53:02 +0000 (13:53 +0100)]
Paint GtkFrame background
It already paints the css border, so let's make it also honor css
background. This is needed to have a box of a different color around
some widgets (e.g. latest gnome-clocks design)
Aurimas Černius [Thu, 20 Dec 2012 21:07:08 +0000 (23:07 +0200)]
Updated Lithuanian translation
Alexander Larsson [Wed, 19 Dec 2012 21:27:08 +0000 (22:27 +0100)]
broadway: Use SOURCE operator when scrolling
Alexander Larsson [Wed, 19 Dec 2012 20:58:25 +0000 (21:58 +0100)]
broadway: Allow reconnecting to a broadway page
Alexander Larsson [Wed, 19 Dec 2012 20:57:58 +0000 (21:57 +0100)]
broadway: Fix possible access-after-free
We need to calculate the buf pointer after the realloc.
Alexander Larsson [Wed, 19 Dec 2012 12:09:38 +0000 (13:09 +0100)]
broadway: Remove separate browser windows hack
This never really worked well, we need to be able to fully handle
sizes from the client.
Alexander Larsson [Wed, 19 Dec 2012 11:37:02 +0000 (12:37 +0100)]
broadway: Separate out the server parts
This (shouldn't) change any behaviour, but it moves the
webserver parts to a separate file, making the broadway display file
smaller and preparing for later separating out the server to its own
process.
Shantha kumar [Wed, 19 Dec 2012 07:51:54 +0000 (13:21 +0530)]
Tamil translation updated
Shantha kumar [Wed, 19 Dec 2012 07:50:51 +0000 (13:20 +0530)]
Tamil translation updated
Benjamin Otte [Tue, 18 Dec 2012 17:23:07 +0000 (18:23 +0100)]
combobox: Don't special-case RTL child positions anymore
If you want to get rounded corners on an hbox, instead of
:first-child {
border-top-left-radius: 5px;
border-bottom-left-radius: 5px;
}
:last-child {
border-top-right-radius: 5px;
border-bottom-right-radius: 5px;
}
you now need to write:
:first-child, :last-child:dir(rtl) {
border-top-left-radius: 5px;
border-bottom-left-radius: 5px;
}
:last-child, :first-child:dir(rtl)
{
border-top-right-radius: 5px;
border-bottom-right-radius: 5px;
}
Benjamin Otte [Tue, 18 Dec 2012 17:18:24 +0000 (18:18 +0100)]
toolbar: Don't special-case RTL toolbar child positions anymore
If you want to get rounded corners on an hbox, instead of
:first-child {
border-top-left-radius: 5px;
border-bottom-left-radius: 5px;
}
:last-child {
border-top-right-radius: 5px;
border-bottom-right-radius: 5px;
}
you now need to write:
:first-child, :last-child:dir(rtl) {
border-top-left-radius: 5px;
border-bottom-left-radius: 5px;
}
:last-child, :first-child:dir(rtl)
{
border-top-right-radius: 5px;
border-bottom-right-radius: 5px;
}
Benjamin Otte [Tue, 18 Dec 2012 17:12:26 +0000 (18:12 +0100)]
box: Don't special-case RTL hbox child positions anymore
If you want to get rounded corners on an hbox, instead of
:first-child {
border-top-left-radius: 5px;
border-bottom-left-radius: 5px;
}
:last-child {
border-top-right-radius: 5px;
border-bottom-right-radius: 5px;
}
you now need to write:
:first-child, :last-child:dir(rtl) {
border-top-left-radius: 5px;
border-bottom-left-radius: 5px;
}
:last-child, :first-child:dir(rtl) {
border-top-right-radius: 5px;
border-bottom-right-radius: 5px;
}
Benjamin Otte [Tue, 18 Dec 2012 17:05:28 +0000 (18:05 +0100)]
stylecontext: Deprecate direction property
... and the functions implementing it. Also deprecate the direction
getter on GtkThemingEngine.
Benjamin Otte [Tue, 18 Dec 2012 16:47:06 +0000 (17:47 +0100)]
stylecontext: Use state flags for direction
... instead of using a custom direction member.
And with that, GtkWidget doesn't need to call
gtk_style_context_set_direction() anymore.
Benjamin Otte [Tue, 18 Dec 2012 16:31:10 +0000 (17:31 +0100)]
widget: Set state flags for direction
Benjamin Otte [Tue, 18 Dec 2012 16:30:58 +0000 (17:30 +0100)]
css: Parse and print the new state flags
Benjamin Otte [Tue, 18 Dec 2012 15:42:04 +0000 (16:42 +0100)]
API: Add state flags for RTL and LTR
See http://www.whatwg.org/specs/web-apps/current-work/#pseudo-classes
for the rationale.
Mattias Põldaru [Tue, 18 Dec 2012 13:56:10 +0000 (15:56 +0200)]
[l10n] Updated Estonian translation
Matej Urbančič [Tue, 18 Dec 2012 07:50:17 +0000 (08:50 +0100)]
Updated Slovenian translation
Matthias Clasen [Tue, 18 Dec 2012 05:43:14 +0000 (00:43 -0500)]
bump version
Matthias Clasen [Tue, 18 Dec 2012 05:37:25 +0000 (00:37 -0500)]
3.7.4
Matthias Clasen [Tue, 18 Dec 2012 05:14:53 +0000 (00:14 -0500)]
Update results for Unicode ellises
Matthias Clasen [Tue, 18 Dec 2012 05:02:09 +0000 (00:02 -0500)]
Update for some changes to selectable/selected state handling
Since this is no longer implemented automatically in atk, some
accessibles, such as menus, loose their selectable state.
Matthias Clasen [Tue, 18 Dec 2012 04:59:23 +0000 (23:59 -0500)]
NotebookPageAccessible: Implement selectable/selected states
The automatic handling for this was removed from atk, so
we need to do it ourselves here.
Matthias Clasen [Tue, 18 Dec 2012 04:58:24 +0000 (23:58 -0500)]
GtkMenuItemAccessible: implement selectable/selected states
The automatic handling for this was removed in atk, so
we need to do it ourselves now.
Matthias Clasen [Tue, 18 Dec 2012 03:45:08 +0000 (22:45 -0500)]
update_type_references: Deal with type_refs_ht being NULL
This can apparently happen, and it was breaking make check
on my system, by causing it to throw a critical out of
g_hash_table_iter_init from here.
Matthias Clasen [Tue, 18 Dec 2012 03:19:59 +0000 (22:19 -0500)]
docs: Add an index for 3.8 api additions
Matthias Clasen [Tue, 18 Dec 2012 03:18:06 +0000 (22:18 -0500)]
Fix up gtk_builder_expose_object addition
It needs an AVAILABLE_IN annotation in the header, and it
needs to be added to the symbols list.
Michael Natterer [Mon, 17 Dec 2012 21:20:40 +0000 (22:20 +0100)]
quartz: make setting_same_owner member of GtkClipboardOwner @public
to fix the build, thanks to parafin for the patch.
(cherry picked from commit
e3e055f8551ac8ee033f361261c849c612554184)
Benjamin Otte [Mon, 17 Dec 2012 15:42:55 +0000 (16:42 +0100)]
stylecontext: Fix gtk_style_context_set_background()
This function is just a sophisitcated optimization.
If we know the GDK window's background will be opaque, we mark it as
opaque. This is so GDK can do all the optimizations it does for opaque
windows and be fast.
This is mainly used when scrolling.
The previous code didn't get this right, in particular it didn't enforce
a transparent background when it knew the background was not opaque.
Benjamin Otte [Mon, 17 Dec 2012 15:42:07 +0000 (16:42 +0100)]
gdk: API: constify argument
gdk_window_set_background_rgba() should take a const RGBA.
Benjamin Otte [Mon, 17 Dec 2012 16:17:54 +0000 (17:17 +0100)]
menushell: Remove unused variable
Introduced in
531d0dc32a38c3bb00a5f9f6fc37e126f62beb07. Welcome to the
club of people not using -Werror Stéphane.
Stéphane Démurget [Sat, 15 Dec 2012 16:21:12 +0000 (17:21 +0100)]
gtkmenushell: Fix an endless loop on focus cycle
This is a simple fallout from sealing gtkmenushell, which only appears
when F10 or Shift-F10 is used in a submenu.
https://bugzilla.gnome.org/show_bug.cgi?id=690266
Nilamdyuti Goswami [Mon, 17 Dec 2012 09:29:55 +0000 (14:59 +0530)]
Assamese translation updated
Christian Kirbach [Mon, 17 Dec 2012 00:45:16 +0000 (01:45 +0100)]
Updated German translation (bug 690302)
Piotr Drąg [Sun, 16 Dec 2012 02:37:09 +0000 (03:37 +0100)]
Updated Polish translation
Michael Natterer [Fri, 14 Dec 2012 14:53:29 +0000 (15:53 +0100)]
gtk: fix scrolling in modal dialogs when event_widget is insensitive
When checking for modal grabs in gtk_main_do_event(), forward
GDK_SCROLL events to event_widget, even if it is insensitive.
Cosimo Cecchi [Fri, 14 Dec 2012 02:54:50 +0000 (21:54 -0500)]
cssstyleproperty: remove unused variable
Cosimo Cecchi [Fri, 14 Dec 2012 02:52:51 +0000 (21:52 -0500)]
a11y: fix uninitialized variables compiler warning
The code path where we update the tooltip text property doesn't set
the state and value variables, and so doesn't need to call
notify_state_change().
Return early, and move the if block at the beginning of the function for
clarity.
Jasper St. Pierre [Fri, 19 Oct 2012 17:33:31 +0000 (13:33 -0400)]
xi2: Abort early if we don't have a proper GDK window
This can happen in mutter or other applications that use GDK filters
but don't actually create GDK windows for everything they get events
for.
https://bugzilla.gnome.org/show_bug.cgi?id=689401
Alexander Larsson [Thu, 13 Dec 2012 09:08:37 +0000 (10:08 +0100)]
IconHelper: Avoid warnings for non-existing stock ids
There are some registred stock ids like gtk-discards that have no icons,
and you could also pass a non-registred stock id. Both of these means
gtk_style_context_lookup_icon_set returns NULL, which causes
a critical in gtk_icon_set_render_icon_pixbuf.
We avoid this by just making these render as EMPTY.
Chun-wei Fan [Mon, 26 Nov 2012 04:57:05 +0000 (12:57 +0800)]
Bug 668239 - texts disappear when notebook switch page at zh_CN locate
In gtkimcontextime.c, use gdk_win32_window_get_impl_hwnd() to get to
the impl's existing native window instead of GDK_WINDOW_HWND() which
implicitly ensures a native window for the widget itself. This seems
to work around whatever GDK problem with native subwindows and fixes
the bug.
This is based on Michael Natterer's fix for gtk-2-24.
Alexander Larsson [Wed, 12 Dec 2012 13:59:06 +0000 (14:59 +0100)]
Use GTK_RESIZE_PARENT resize_mode for GtkViewport
We used to use GTK_RESIZE_QUEUE, but that is problematic for e.g
a GtkScrolledWindow with NEVER scroll policies, as size changes
in ancestors will never get propagated to the scrolled window, causing
it to not have the correct size.
This is a slight performance hit, but in practice its not bound to be
problematic. In typical UIs there is only a single "large" GtkScrolledWindow
visible at a time, so a size requeust propagating out of such a window
will only hit the smaller amount of widgetry outside the scrolled window,
and additionally all such widgets will have their size request caches
still valid.
https://bugzilla.gnome.org/show_bug.cgi?id=690099
Yaron Shahrabani [Wed, 12 Dec 2012 09:24:50 +0000 (11:24 +0200)]
Updated Hebrew translation.
Matthias Clasen [Wed, 12 Dec 2012 02:30:22 +0000 (21:30 -0500)]
Trivial whitespace fix
Alexander Larsson [Tue, 11 Dec 2012 14:41:50 +0000 (15:41 +0100)]
overlay: Fix child window position/size on realize
We don't get an automatic queue resize on realize anymore, which
was papering over this bug where we did not set the child window
size/position at realize time.
John Ralls [Tue, 11 Dec 2012 00:51:31 +0000 (16:51 -0800)]
Fix compilation on MacOSX Tiger
NSInteger isn't defined until MacOSX Leopard, so in the 3 files that
use it, include ../gdk/quartz/gdkquartz.h which provides a typedef when
needed.
Juan Pablo Ugarte [Mon, 10 Dec 2012 17:53:53 +0000 (14:53 -0300)]
Added new function gtk_builder_expose_object() based on the original work by
Marco Diego Aurélio Mesquita on bug #447972
Alexander Larsson [Mon, 10 Dec 2012 13:49:51 +0000 (14:49 +0100)]
css: Avoid looking up the GtkCssStyleProperty class a lot
Instead of constantly looking up the class we just stash it away in
class_init.
Alexander Larsson [Mon, 10 Dec 2012 11:57:10 +0000 (12:57 +0100)]
css: Speed up name matching
We use the new g_type_get_type_registration_serial() so that we can
cache and properly invalidate the result of g_type_from_name().
This bumps the glib requirement to 2.35.3 to get the new function.
https://bugzilla.gnome.org/show_bug.cgi?id=689847
Alexander Larsson [Mon, 10 Dec 2012 11:28:44 +0000 (12:28 +0100)]
css: Clean up tree_match implementations
A bunch of repeated code is broken out into the helper
gtk_css_selector_tree_match_previous().
Alexander Larsson [Mon, 10 Dec 2012 10:27:43 +0000 (11:27 +0100)]
css: Do get_change directly on the tree without matching first
Rather than first collecting matches and then getting the change
for them we do the change collection directly on the tree. This
is about twice as fast.
Alexander Larsson [Fri, 7 Dec 2012 17:55:49 +0000 (18:55 +0100)]
css: Factor out some of the position matching code
Mario Sanchez Prada [Sun, 9 Dec 2012 10:49:35 +0000 (11:49 +0100)]
Add missing NULL-check in GtkEntryAccessible
Only call to atk_object_set_name if gtk_entry_get_icon_name() is not NULL.
https://bugzilla.gnome.org/show_bug.cgi?id=689923
Daniel Mustieles [Sat, 8 Dec 2012 18:50:40 +0000 (19:50 +0100)]
Updated Spanish translation
Colin Walters [Fri, 7 Dec 2012 19:05:58 +0000 (14:05 -0500)]
Fix build by un-translating org.gtk.Settings.FileChooser
At the moment, gtk+ doesn't depend on intltool, which is the program
that knows how to translate schemas. Attempting to translate them
causes a build failure, so for now, let's leave them in en_US.
https://bugzilla.gnome.org/show_bug.cgi?id=689584
Timothy Arceri [Mon, 3 Dec 2012 20:59:49 +0000 (07:59 +1100)]
Add summary and description to "show-hidden" key
https://bugzilla.gnome.org/show_bug.cgi?id=689584
Cosimo Cecchi [Fri, 7 Dec 2012 16:41:03 +0000 (11:41 -0500)]
scale: render scale mark lines with separator style class
So that the theme can distinguish them from the mark text.
Cosimo Cecchi [Fri, 7 Dec 2012 16:34:49 +0000 (11:34 -0500)]
build: fix some GCC warnings
Michael Natterer [Fri, 7 Dec 2012 11:19:52 +0000 (12:19 +0100)]
quartz: fix crash in the recent clipboard "fix", and really fix it
We must not release the GtkClipboardOwner in pasteboardChangedOwner
becaue we don't own a reference to ourselves (NSPasteboard does).
Instead, release the owner right after setting it, transferring
ownership to NSPasteboard
Also, fix repeated setting of the same owner by keeping the
owner around in GtkCLipboard, and re-use it if "user_data"
doesn't change. To avoid clipboard_unset()ting our own contents
in the process, add an ugly "setting_same_owner" boolean to
GtkClipboardOwner, set it during re-setting the same owner,
and avoid calling clipboard_unset() from pasteboardChangedOwner
if it's TRUE.
(cherry picked from commit
4a8df7a33c298d22bf78b947d0e861fc03ec70e1)
Jasper St. Pierre [Thu, 6 Dec 2012 22:48:11 +0000 (17:48 -0500)]
gtkcssselector: Fix build
destroy vs. free strikes again
Jasper St. Pierre [Thu, 6 Dec 2012 22:38:06 +0000 (17:38 -0500)]
gtkcssselector: Don't leak the hash table
We should probably free the memory we allocate. Sounds like a winning
strategy.
Alexander Larsson [Thu, 6 Dec 2012 21:28:11 +0000 (22:28 +0100)]
treeview: Don't invalidate whole tree unless needed
We currently invalidate the whole tree every time the style state
changes in the tree view. The primary reason for this is to catch
default font changes as that may affect text cell renderers. But
cell renderers could *potentially* also read other style properties
(although that seems weird and unlikely).
We handle this by invalidating only when some state that affects sizes
is changed. This includes all the font properties.