Chris Cummins [Tue, 26 Mar 2013 14:44:24 +0000 (14:44 +0000)]
wayland: Do not free wayland cursor buffers
The cursor buffer is only non-null when a cursor is created from pixbuf,
so it is not necessary to keep track of whether to free this buffer on
finalize.
Chris Cummins [Tue, 26 Mar 2013 14:19:37 +0000 (14:19 +0000)]
wayland: Differentiate wayland and pixbuf cursors
By keeping a pointer to the wl_cursor struct in GdkWaylandCursor, it is
no longer necessary to duplicate cursor data (width, height, hotspots,
etc.) between wl_cursor and GdkWaylandCursor.
Andika Triwidada [Thu, 28 Mar 2013 14:05:01 +0000 (21:05 +0700)]
Updated Indonesian translation of po-properties
Alexander Larsson [Thu, 28 Mar 2013 12:24:04 +0000 (13:24 +0100)]
GtkOverlay: Respect overlay child min size
Never allocate an overlayed child less than its minimum request.
If the minimum doesn't fit we will gracefully clip via the
widget window.
https://bugzilla.gnome.org/show_bug.cgi?id=696623
Alexander Larsson [Thu, 28 Mar 2013 11:10:09 +0000 (12:10 +0100)]
GtkOverlay: Don't allocate child except in size_allocate
We were calling gtk_overlay_child_allocate() both in realize
and in add as we wanted to create and position the child windows
for the widgets. However, this call also actually called
gtk_widget_size_allocate() on the child, which it shouldn't. In some
cases the overlay is realized before being allocated, and thus it
will allocate the child at 0x0 which is an invalid size for it to be in.
In particular, if the child has margins set this will result in negative
allocations and warnings.
This fix splits out the allocation computation so that
gtk_overlay_create_child_window can use it without callers
having to call gtk_overlay_child_allocate() to move the windows.
https://bugzilla.gnome.org/show_bug.cgi?id=696623
Kristian Rietveld [Sun, 30 Dec 2012 16:09:22 +0000 (17:09 +0100)]
quartz: move atom/pasteboard type conversions functions to GDK
(cherry picked from commit
a269c2f8d26a588c4b61ad605ef496e89759cc03)
Jan Arne Petersen [Wed, 20 Mar 2013 15:40:26 +0000 (16:40 +0100)]
docs: Add wayland documentation to Gdk reference
Kristian Høgsberg [Wed, 27 Mar 2013 17:38:15 +0000 (13:38 -0400)]
wayland: Use the wl_display_sync() request to track pending init roundtrips
Instead of maintaining the init refcount in regular event handlers that can
fire in case of hotplug or mode changes, use a dedicated sync callback
to wait for roundtrips.
Kristian Høgsberg [Wed, 27 Mar 2013 17:21:54 +0000 (13:21 -0400)]
wayland: Pass most recent button press serial to set_popup_surface
Alexander Larsson [Wed, 27 Mar 2013 10:47:37 +0000 (11:47 +0100)]
gdkwindow: Apply layered areas to the shape for native windows
If a window is overlapped by a layered (i.e. partially transparent)
window then that region will not disappear from the native window clip
region. This lets us handle compositing multiple layers of windows.
For native subwindows this doesn't really work. For them we apply the
clip region as a shape to the native window which lets us have client
side windows overlapping the native window. However, with the addition
of the layered stuff the "overlapped-by-alpha-csw" part got broken, as
this area is not removed from the clip region of the native window.
We fix this by also removing the layered area when applying the shape.
This means alpha and alpha backgrounds don't work over native windows,
but there is not much to do about that.
https://bugzilla.gnome.org/show_bug.cgi?id=696370
Alexander Larsson [Wed, 27 Mar 2013 10:42:44 +0000 (11:42 +0100)]
gdkwindow: Add alpha_bg to window debug print
Matthias Clasen [Wed, 27 Mar 2013 04:23:19 +0000 (00:23 -0400)]
Fix GtkApplicationWindow menubar placement
Since the menubar is part of the content, we need to give
it the content_window as parent window, to make things
work again.
https://bugzilla.gnome.org/show_bug.cgi?id=696561
Matthias Clasen [Wed, 27 Mar 2013 03:30:10 +0000 (23:30 -0400)]
Don't return FALSE from a non-boolean function
https://bugzilla.gnome.org/show_bug.cgi?id=696608
Matthias Clasen [Wed, 27 Mar 2013 02:57:46 +0000 (22:57 -0400)]
wayland: Handle seat removal
Also, emit ::device-added and ::device-removed signals
as devices appear and disappear.
Matthias Clasen [Wed, 27 Mar 2013 02:16:15 +0000 (22:16 -0400)]
wayland: Make monitor removal work
The global_removal argument is the _name_ of the object.
We were comparing it to the _object id_ of the object.
To fix this, store the name at the time the object is bound.
Matthias Clasen [Wed, 27 Mar 2013 01:53:41 +0000 (21:53 -0400)]
wayland: Also emit monitors-changed when monitors are unplugged
Matthias Clasen [Wed, 27 Mar 2013 01:47:49 +0000 (21:47 -0400)]
wayland: Also emit size-changed when appropriate
We need to be a bit more careful when updating the screen
size - the code that was there would not do the right thing
if e.g. the width of one monitor was reduced.
Matthias Clasen [Wed, 27 Mar 2013 01:37:21 +0000 (21:37 -0400)]
wayland: Emit monitors-changed when monitors change
Matthias Clasen [Wed, 27 Mar 2013 01:27:21 +0000 (21:27 -0400)]
Don't decrement the init_ref_count towards -infinity
Once we are at zero, initialization is done.
Matthias Clasen [Wed, 27 Mar 2013 00:46:13 +0000 (20:46 -0400)]
Pimp up testheaderbar some more
Marek Černocký [Tue, 26 Mar 2013 22:06:55 +0000 (23:06 +0100)]
Updated Czech translatio
Kristian Høgsberg [Tue, 26 Mar 2013 21:23:31 +0000 (17:23 -0400)]
wayland: Set screen size to bounding box of all outputs
Don't hardcode 8192x8192. We don't get this info from wayland, so we
compute it as we receive information about the available wl_outputs.
https://bugzilla.gnome.org/show_bug.cgi?id=692871
Kristian Høgsberg [Tue, 26 Mar 2013 18:42:14 +0000 (14:42 -0400)]
wayland: Roundtrip until we've receive initial input and output configuration
We use a ref-count mechanism to track whether parts of the init sequence
still needs round trips to receive remaining initial state. Typically
we need a couple of roundtrips total to get the global list, then the
input and output configurations, but with the ref-count we avoid making
global assumptions like that.
https://bugzilla.gnome.org/show_bug.cgi?id=696340
Chun-wei Fan [Mon, 18 Mar 2013 01:16:13 +0000 (09:16 +0800)]
Bug 695895: Fix gtk3-demo on Windows
In commit
4e41577b, we are using g_content_type_is_a() to determine how
to display the demo resources in the right pane of the gtk3-demo program.
Use g_content_type_get_mime_type(), so that we can obtain the mime
type of the demo resources on all platforms, as g_content_type_guess()
returns a platform-specific string, as
https://developer.gnome.org/gio/2.35/gio-GContentType.html states.
As .ui files and .css files are normally registered with a different mime
type string on Windows, check for those strings as well.
This will ensure the demo resources can be properly displayed on Windows
as well.
Chun-wei Fan [Tue, 26 Mar 2013 03:56:32 +0000 (11:56 +0800)]
Fix Visual Studio property sheets
The creation of the folder for the "installation" of a11y headers is wrong,
so fix that.
Daniel Korostil [Mon, 25 Mar 2013 20:39:47 +0000 (22:39 +0200)]
Added uk translation
Mario Blättermann [Mon, 25 Mar 2013 18:29:37 +0000 (19:29 +0100)]
[l10n] Updated German translation
Carlos Garcia Campos [Mon, 25 Mar 2013 12:01:23 +0000 (13:01 +0100)]
gtkprintbackendcups: Fix a crash in avahi_create_browsers()
In avahi_request_printer_list() a new connection to the DBus system bus
is started asynchronously, but it's not cancellable and it's not taking
any reference of the GtkPrintBackendCups. This means that when the
callback is called, the object might have been destroyed already. We can
just pass the cancellable created and check for the cancelled error in
the callback before trying to use the GtkPrintBackendCups. The code to
cancel avahi operations and to unsibscribe from the DBus signals has
been moved from finalize to dispose to make sure it happens as soon as
possible.
https://bugzilla.gnome.org/show_bug.cgi?id=696553
Carlos Garcia Campos [Mon, 25 Mar 2013 11:09:52 +0000 (12:09 +0100)]
gtkprintbackendcups: Fix a crash in gtk_print_backend_cups_finalize()
If GtkPrintBackendCups is finalized and cups_get_printer_list hasn't
been called, g_object_unref is called for the GDBusConnection pointer
that is NULL. Use g_clear_object() instead.
https://bugzilla.gnome.org/show_bug.cgi?id=696546
Alexandre Franke [Mon, 25 Mar 2013 11:07:06 +0000 (12:07 +0100)]
Update French translation
Shantha kumar [Mon, 25 Mar 2013 09:48:45 +0000 (15:18 +0530)]
Tamil Translations Updated
Shantha kumar [Mon, 25 Mar 2013 09:48:30 +0000 (15:18 +0530)]
Tamil Translations Updated
Ani Peter [Mon, 25 Mar 2013 08:33:52 +0000 (14:03 +0530)]
Completed for Malayalam
Matthias Clasen [Sun, 24 Mar 2013 18:55:52 +0000 (14:55 -0400)]
wayland: Add toolbar settings
Luca Ferretti [Sun, 24 Mar 2013 19:57:22 +0000 (20:57 +0100)]
[l10n] updated Italian translation
Jan Arne Petersen [Wed, 20 Mar 2013 15:38:36 +0000 (16:38 +0100)]
wayland: Add support for custom surfaces
Allow to set a GdkWindow to use a custom surface instead of a
wl_shell_surface. It allows to register the surface as a custom type
with some Wayland interface.
https://bugzilla.gnome.org/show_bug.cgi?id=695861
Matthias Clasen [Sun, 24 Mar 2013 18:21:41 +0000 (14:21 -0400)]
broadway: Don't use deprecated gdk_cursor_ref
Sébastien Wilmet [Tue, 19 Mar 2013 13:31:43 +0000 (14:31 +0100)]
GtkWidget: add missing deprecation marks
https://bugzilla.gnome.org/show_bug.cgi?id=696138
Petr Kovar [Sun, 24 Mar 2013 18:19:07 +0000 (19:19 +0100)]
Update Czech translation
Sébastien Wilmet [Sat, 23 Mar 2013 15:44:49 +0000 (16:44 +0100)]
GtkLabel: fix underlining of mnemonics
The underscore was shown in front of the mnemonic instead of underlining
it (e.g., _Save).
Thanks to Lars Uebernickel.
https://bugzilla.gnome.org/show_bug.cgi?id=674759
Ask H. Larsen [Sun, 24 Mar 2013 13:50:40 +0000 (14:50 +0100)]
Updated Danish translation
Mario Blättermann [Sun, 24 Mar 2013 11:53:30 +0000 (12:53 +0100)]
[l10n] Updated German translation
Matthias Clasen [Sun, 24 Mar 2013 01:06:14 +0000 (21:06 -0400)]
csd: Reparent the content
This lets us avoid problems with cursor changes.
Matthias Clasen [Sat, 23 Mar 2013 23:41:39 +0000 (19:41 -0400)]
Make gtk_init_check work as expected
When no GDK backend can be initialized (either because
GDK_BACKEND has been set to the wrong value, or the backends
are simply not included), the expectation is that gtk_init_check
should return FALSE, not error out.
This commit makes it so, by using gdk_display_manager_peek
instead of gdk_display_manager_get in code paths that are used
during initialization.
Matthias Clasen [Sat, 23 Mar 2013 23:40:16 +0000 (19:40 -0400)]
Make gdk_display_open_default_libgtk_only more robust
Make this function return NULL if GDK could not be
initialized. We check this by seeing if there is a
display manager singleton.
Matthias Clasen [Sat, 23 Mar 2013 23:39:58 +0000 (19:39 -0400)]
Add gdk_display_manager_peek
Instead of failing with an error if no GDK backend is found like
gdk_display_manager_get(), the new peek function silently returns
NULL.
Matthias Clasen [Sat, 23 Mar 2013 22:31:20 +0000 (18:31 -0400)]
wayland: Remove a few unused variables
Matthias Clasen [Sat, 23 Mar 2013 22:02:28 +0000 (18:02 -0400)]
wayland: Don't recreated keymaps on layout change
The GDK model for keymaps expects the keymap object to stay
around and emit a ::keys-changed signal. So, do that. This
should make layout changes work, but it remains untested since
weston does not support layout changes at runtime.
At the same time, plug a memory leak where GdkWaylandKeymap
forgot to free its xkb objects in finalize.
https://bugzilla.gnome.org/show_bug.cgi?id=696339
Tarnyko [Sat, 23 Mar 2013 21:42:55 +0000 (17:42 -0400)]
win32: Bring back spinner animation
Define the spinner animation in css.
https://bugzilla.gnome.org/show_bug.cgi?id=696202
Matthias Clasen [Sat, 23 Mar 2013 21:10:31 +0000 (17:10 -0400)]
wayland: remove an unused variable
Matthias Clasen [Sat, 23 Mar 2013 21:09:38 +0000 (17:09 -0400)]
wayland: Respect keymap repeat information
xkb has a function to tell us whether a key should be repeated
or not. Lets use that information.
Jorge Pérez Pérez [Sat, 23 Mar 2013 21:19:11 +0000 (22:19 +0100)]
Updated Aragonese translation
Colin Walters [Sat, 23 Mar 2013 18:09:20 +0000 (14:09 -0400)]
gtktoolpalette: Only call gdk_atom_intern when we're constructed
Otherwise we end up trying to instantiate a display manager in
class_init which breaks introspection scanning.
https://bugzilla.gnome.org/show_bug.cgi?id=696457
Colin Walters [Sat, 23 Mar 2013 17:51:24 +0000 (13:51 -0400)]
gdkkeys: Avoid instantiating a display manager if none exists
For introspection scanning which ends up calling class_init() which in
turn calls into the keybindings code, we can just use the fallback
keyval conversion code.
https://bugzilla.gnome.org/show_bug.cgi?id=696457
Marc-André Lureau [Wed, 20 Mar 2013 22:12:56 +0000 (23:12 +0100)]
win32: do not crash on invalid utf8 conversion
g_utf8_to_utf16() is not guaranteed to succeed. Check the error
and return if it failed.
https://bugzilla.gnome.org/show_bug.cgi?id=696232
Tarnyko [Sat, 23 Mar 2013 19:47:39 +0000 (15:47 -0400)]
win32: Improve GtkAssistant theming
The highlighted color chosen for GtkAssistant header font
(deep blue) makes it hardly readable is most cases. Switch
to light gray.
https://bugzilla.gnome.org/show_bug.cgi?id=696171
Matthias Clasen [Sat, 23 Mar 2013 19:43:20 +0000 (15:43 -0400)]
Expand testgrid to cover row/column removal
Matthias Clasen [Sat, 23 Mar 2013 19:40:44 +0000 (15:40 -0400)]
Add gtk_grid_remove_{row,column}
It is sometimes convenient to deal with entire rows or
columns at a time.
https://bugzilla.gnome.org/show_bug.cgi?id=695994
John Lindgren [Sat, 23 Mar 2013 18:41:09 +0000 (19:41 +0100)]
Use natural size to set adjustment ranges
This is an (unintentional) side effect of my changes to GtkTreeView's
get_preferred_size() implementation. It seems odd to me that
GtkTreeView directly determines its own size when inside a
GtkScrolledWindow, but since it does, it should be using its natural
size, not its minimum size.
Vadim Godunko [Sat, 23 Mar 2013 18:01:35 +0000 (14:01 -0400)]
Fix drawing of grid lines in RTL
The vertical grid lines were not properly positions in
RTL locales.
https://bugzilla.gnome.org/show_bug.cgi?id=696051
Matthias Clasen [Sat, 23 Mar 2013 16:56:54 +0000 (12:56 -0400)]
wayland: Pick up more settings
With this commit, we pick up xft settings from GSettings
as well. Among other things, this makes the Large Text
setting work. Still to do: pick up fontconfig changes without
having all clients use up inotify watches for all font
directories.
Yuri Myasoedov [Sat, 23 Mar 2013 16:03:18 +0000 (20:03 +0400)]
Updated Russia translation
Tristan Van Berkom [Sat, 23 Mar 2013 11:44:12 +0000 (20:44 +0900)]
GtkBuilder: Fixed documentation
The documentation was referring to an <external-object> tag
which we never implemented
Matthias Clasen [Sat, 23 Mar 2013 04:16:24 +0000 (00:16 -0400)]
Add gdk_set_allowed_backends
This new function allows programmatic control over the
GDK backends that will be used at runtime.
Matthias Clasen [Sat, 23 Mar 2013 04:47:22 +0000 (00:47 -0400)]
Implement GInitable in GdkBroadwayDisplayManager
Add GInitable implementation and fail the initialisation if it is not
possible to set up a broadway server.
Matthias Clasen [Sat, 23 Mar 2013 04:37:53 +0000 (00:37 -0400)]
Implement GInitable in GdkX11DisplayManager
Add GInitable implementation and fail the initialisation if it is not
possible to connect to the display server.
Thomas Wood [Fri, 22 Feb 2013 16:44:13 +0000 (16:44 +0000)]
wayland: implement GInitable and check the connection to the display server
Add GInitable implementation and fail the initialisation if it is not
possible to connect to the display server.
https://bugzilla.gnome.org/show_bug.cgi?id=694465
Thomas Wood [Fri, 22 Feb 2013 16:41:12 +0000 (16:41 +0000)]
gdk: Implement GInitable on GdkDisplayManager
Add GInitable interface with a default implementation that always
succeeds. This allows backends to override the GInitable implementation
and add their own checks to determine if the backend can be loaded. If
a backend cannot be loaded, GDK can attempt to load the next available
backend.
Since backends may need to read any relevant options (such as the
display flag) to determine if they can be created successfully, this
patch also removes calls that attempt to create the display manager
before the options have been parsed.
https://bugzilla.gnome.org/show_bug.cgi?id=694465
Matthias Clasen [Sat, 23 Mar 2013 00:31:50 +0000 (20:31 -0400)]
Update cursor theme on wayland
Call the new new api from gtksettings.c when it changes, just
as we do for X11.
https://bugzilla.gnome.org/show_bug.cgi?id=696428
Matthias Clasen [Sat, 23 Mar 2013 00:30:51 +0000 (20:30 -0400)]
wayland: Implement cursor theme changing
Add gdk_wayland_display_set_cursor_theme and implement it.
Matthias Clasen [Fri, 22 Mar 2013 23:32:39 +0000 (19:32 -0400)]
wayland: Respect cursor size setting
Respect the cursor size setting when loading the cursor theme.
Matthias Clasen [Fri, 22 Mar 2013 23:27:36 +0000 (19:27 -0400)]
wayland: Add cursor-size setting
This will be picked up by cursor themes soon.
Matthias Clasen [Fri, 22 Mar 2013 23:09:20 +0000 (19:09 -0400)]
wayland: Avoid segfault when cursors are freed
The check for GDK_CURSOR_IS_PIXMAP was ineffective, since _all_
cursors have this type, from the looks of it. Instead, store
buffer ownership information separately.
Matthias Clasen [Fri, 22 Mar 2013 22:34:44 +0000 (18:34 -0400)]
wayland: Emit GdkKeymap::state-changed when appropriate
This makes the caps lock warning in password entries work.
Matthias Clasen [Fri, 22 Mar 2013 22:26:33 +0000 (18:26 -0400)]
wayland: Make key repeat configurable
These might be candidates for a future settings interface; until
then, we use GSettings directly. Note again that we are careful
to avoid a dependency on GNOME schemas.
Matthias Clasen [Fri, 22 Mar 2013 22:09:23 +0000 (18:09 -0400)]
wayland: Factor out key repeat parameters
This is in preparation for picking these values up from
existing configuration.
Matthias Clasen [Fri, 22 Mar 2013 21:53:28 +0000 (17:53 -0400)]
wayland: Drop an unused variable
The keymap member in GdkWaylandDisplay is not used at all,
so drop it.
Matthias Clasen [Fri, 22 Mar 2013 21:43:50 +0000 (17:43 -0400)]
wayland: Fix key repeat
Key repeat under X is not affected by modifiers. And on some systems
(e.g my Thinkpad), NumLock is permanently on, rendering key repeat
nonfunctional. This commit changes the Wayland backend to do
key repeat regardless of modifiers.
http://bugzilla.gnome.org/show_bug.cgi?id=695497
Cosimo Cecchi [Fri, 22 Mar 2013 16:44:27 +0000 (12:44 -0400)]
appchooserbutton: filter out applications not supposed to be visible
Ensure NoDisplay=true applications don't end up in the list populated by
GtkAppChooserButton.
https://bugzilla.gnome.org/show_bug.cgi?id=696405
Rajesh Ranjan [Fri, 22 Mar 2013 09:31:16 +0000 (15:01 +0530)]
hindi translation
Matthias Clasen [Thu, 21 Mar 2013 18:01:33 +0000 (14:01 -0400)]
Don't react to clicks in content
This breaks way too many things, and we explicitly trigger
drags from menubars and toolbars, where we want it.
Matthias Clasen [Thu, 21 Mar 2013 17:39:06 +0000 (13:39 -0400)]
csd: Make link activation work reliably
The label code was just letting some button press events bubble
up, which caused misbehaviour with the window dragging code.
https://bugzilla.gnome.org/show_bug.cgi?id=695506
Matthias Clasen [Thu, 21 Mar 2013 17:07:29 +0000 (13:07 -0400)]
csd: Allow moving maximized windows
This make the nice 'snap off' feature of gnome-shell work
with client-side decorations. weston moves the maximized window
around, which is less ideal...
Thomas Hindoe Paaboel Andersen [Thu, 21 Mar 2013 20:05:32 +0000 (21:05 +0100)]
wayland: complete cursor_for_pixbuf
Finishes the implementation for loading cursors from pixbufs.
Gnome bug #696223
Kalev Lember [Thu, 21 Mar 2013 21:26:02 +0000 (22:26 +0100)]
win32: link with -lwinmm
... to resolve the timeBeginPeriod() / timeEndPeriod() symbols after
commit
5dbf814f0c0c94733ef96d6dc3c8c5ddb7a63f3a.
Cosimo Cecchi [Thu, 21 Mar 2013 12:47:06 +0000 (08:47 -0400)]
window: don't add the background style class when drawing CSD elements
We already set it in init, so this is just redundant.
The additional window-content style class here is needed so that we can
distinguish between the full X window background and the background for
the actual window contents.
Cosimo Cecchi [Thu, 21 Mar 2013 12:46:43 +0000 (08:46 -0400)]
window: add a "client-decorated" style class when the window is so
The theme might want to apply some specific theming to a client
decorated toplevel window, so add a style class for that.
Fran Diéguez [Thu, 21 Mar 2013 19:29:59 +0000 (20:29 +0100)]
Updated galician translations
Baurzhan Muftakhidinov [Thu, 21 Mar 2013 18:22:37 +0000 (00:22 +0600)]
Updated Kazakh translation.
Kristian Høgsberg [Thu, 21 Mar 2013 12:51:35 +0000 (08:51 -0400)]
wayland: Clear current cusor on leave so we always set it on enter
Commit
0d9d808217a04ee965ce5f60f9aff00b2ec4938e fixed the hotspot issue,
but commit
f2cc52fddd74b3064eec85e17161ebbc4ee14f49 then optimized away
cursor changes a little too aggressively. We always need to set the
cursor on enter. Make sure we clear the current cursor on leave so we
don't think it's already set on the next enter.
https://bugzilla.gnome.org/show_bug.cgi?id=695512
Matthias Clasen [Thu, 21 Mar 2013 10:52:42 +0000 (06:52 -0400)]
Wayland: Read settings out of GSettings directly
Until we figure out where we want to go with settings under
Wayland, this makes GTK+ applications a lot easier to deal
with under Wayland.
Note that we are careful to deal with the absence of schemas,
so this does not introduce a dependency on GNOME settings.
Mario Blättermann [Thu, 21 Mar 2013 08:20:34 +0000 (09:20 +0100)]
[l10n] Updated German translation
Matthias Clasen [Thu, 21 Mar 2013 03:54:49 +0000 (23:54 -0400)]
Fix DND
The removal of the Motif DND code accidentally changed the
value of the TARGET_DELETE enumeration value, breaking DND
pretty badly.
Matthias Clasen [Thu, 21 Mar 2013 02:38:36 +0000 (22:38 -0400)]
GtkColorChooserDialog: make alternative button order work
The alternative button order that we set did not actually
work, since the ok button did not have a response id.
Partial fix for
https://bugzilla.gnome.org/show_bug.cgi?id=695493
Kristian Høgsberg [Thu, 21 Mar 2013 02:42:44 +0000 (22:42 -0400)]
wayland: Only set cursor when it changes
Kristian Høgsberg [Thu, 21 Mar 2013 02:34:03 +0000 (22:34 -0400)]
wayland: Add a cursor cache
We avoid creating a new GdkCursor object every time
_gdk_wayland_display_get_cursor_for_name/type() is called.
Kristian Høgsberg [Thu, 21 Mar 2013 02:02:42 +0000 (22:02 -0400)]
wayland: Use the pointer enter serial when setting cursor surface
wl_pointer.set_cursor is rejected if the serial number doesn't match
the enter serial number for the wl_pointer. We passed the right serial
number when setting the cursor surface in response to the enter event.
Later set_cursor requests fail, but we can still attach new buffers to
our cursor surface, which is why the cursor changed, but the hotspot
didn't update. Clicking in the decoration results in a leave/enter pair
which triggers wl_pointer.set_cursor with the right serial. That's why
clicking the decoration sets the right cursor.
https://bugzilla.gnome.org/show_bug.cgi?id=695512
Matthias Clasen [Thu, 21 Mar 2013 02:17:14 +0000 (22:17 -0400)]
GtkColorChooser: Fix selection of colors
Prevent the new window dragging code from interfering with
selection of colors on button release, by handling button
presses we care about instead of letting them bubble up to
the window.
Partial fix for
https://bugzilla.gnome.org/show_bug.cgi?id=695493
Matthias Clasen [Thu, 21 Mar 2013 01:47:12 +0000 (21:47 -0400)]
wayland: Keep hotspot when changing cursors
We need to pass the delta between the old and new hotspot
when attaching the new cursor surface, to keep the hotspot
at the same position. We can't deal with this in the compositor,
since the set_cursor call already overwrites the old hotspot,
so the information is lost by the time the attach happens.
Unfortunately, we can't query the initial hotspot from
the compositor, so the first cursor change will make the
hotspot jump.
https://bugzilla.gnome.org/show_bug.cgi?id=695512
Gabor Kelemen [Wed, 20 Mar 2013 23:31:21 +0000 (00:31 +0100)]
Updated Hungarian translation