gtk4.git
11 years agomir: reconstruct surface if window is moved
William Hua [Mon, 26 Jan 2015 23:29:07 +0000 (18:29 -0500)]
mir: reconstruct surface if window is moved

11 years agomir: stop using deprecated surface api
William Hua [Mon, 19 Jan 2015 17:08:43 +0000 (12:08 -0500)]
mir: stop using deprecated surface api

11 years agomir: fix issues with mouse over parent menu
William Hua [Thu, 8 Jan 2015 05:24:42 +0000 (00:24 -0500)]
mir: fix issues with mouse over parent menu

Mouse over a parent menu[bar] didn't work while the menu was open.
The fix was to correct the behaviour of pointer crossing events so that
the pointer appears to be only inside one window at a time.

See: http://tronche.com/gui/x/xlib/events/window-entry-exit/normal.html

11 years agomir: add back window hierarchy functions
William Hua [Thu, 8 Jan 2015 05:23:02 +0000 (00:23 -0500)]
mir: add back window hierarchy functions

11 years agomir: stop forwarding motion events to subwindows
William Hua [Wed, 7 Jan 2015 17:18:03 +0000 (12:18 -0500)]
mir: stop forwarding motion events to subwindows

11 years agomir: generate leave events for transient children
William Hua [Wed, 17 Dec 2014 17:37:28 +0000 (12:37 -0500)]
mir: generate leave events for transient children

11 years agomir: use monitor plug names more similar to the ones used in Xmir
Marco Trevisan (Treviño) [Fri, 12 Dec 2014 17:42:05 +0000 (18:42 +0100)]
mir: use monitor plug names more similar to the ones used in Xmir

11 years agomir: only paint background if window has one
Marco Trevisan (Treviño) [Fri, 12 Dec 2014 02:01:38 +0000 (03:01 +0100)]
mir: only paint background if window has one

11 years agomir: use old scrolling system if we get simple integer values
Marco Trevisan (Treviño) [Fri, 12 Dec 2014 00:40:55 +0000 (01:40 +0100)]
mir: use old scrolling system if we get simple integer values

11 years agomir: end the current grab on ungrab
William Hua [Tue, 9 Dec 2014 13:57:21 +0000 (08:57 -0500)]
mir: end the current grab on ungrab

11 years agomir: only render non-top-level transients.
William Hua [Mon, 8 Dec 2014 21:24:17 +0000 (16:24 -0500)]
mir: only render non-top-level transients.

11 years agomir: handle new MirCloseSurfaceEvent
Marco Trevisan (Treviño) [Fri, 5 Dec 2014 14:53:03 +0000 (15:53 +0100)]
mir: handle new MirCloseSurfaceEvent

11 years agomir: reset cursor_inside on hide/withdraw
William Hua [Thu, 4 Dec 2014 15:03:40 +0000 (10:03 -0500)]
mir: reset cursor_inside on hide/withdraw

Without this, we generate incorrect crossing events for transient
windows.

11 years agomir: disable some debug output
William Hua [Wed, 3 Dec 2014 21:48:58 +0000 (16:48 -0500)]
mir: disable some debug output

11 years agomir: skip windows that aren't visible
William Hua [Wed, 3 Dec 2014 16:30:34 +0000 (11:30 -0500)]
mir: skip windows that aren't visible

11 years agomir: fix IsModifierKey definition
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 19:04:54 +0000 (20:04 +0100)]
mir: fix IsModifierKey definition

11 years agomir: check the log events env variable just once
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 17:43:32 +0000 (18:43 +0100)]
mir: check the log events env variable just once

11 years agomir: set event source device as well
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 17:19:39 +0000 (18:19 +0100)]
mir: set event source device as well

11 years agomir: fix time conversion macro
William Hua [Tue, 2 Dec 2014 16:12:11 +0000 (11:12 -0500)]
mir: fix time conversion macro

11 years agomir: implement stub _gdk_mir_keymap_key_is_modifier
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 16:08:00 +0000 (17:08 +0100)]
mir: implement stub _gdk_mir_keymap_key_is_modifier

Using IsModifierKey for now

11 years agomir: check (using simple keycode checking) if we're handling a modifier key
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 15:52:03 +0000 (16:52 +0100)]
mir: check (using simple keycode checking) if we're handling a modifier key

11 years agomir: compute the modifiers using the saved button_state on key events
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 15:27:28 +0000 (16:27 +0100)]
mir: compute the modifiers using the saved button_state on key events

11 years agomir: add a NANO_TO_MILLI conversion macro
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 15:25:18 +0000 (16:25 +0100)]
mir: add a NANO_TO_MILLI conversion macro

11 years agomir: properly converts mir nanoseconds timestamps to milliseconds
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 15:06:59 +0000 (16:06 +0100)]
mir: properly converts mir nanoseconds timestamps to milliseconds

11 years agomir: use actual events time
Marco Trevisan (Treviño) [Tue, 2 Dec 2014 14:18:50 +0000 (15:18 +0100)]
mir: use actual events time

11 years agomir: add timestamps to generated GdkEvents
William Hua [Mon, 1 Dec 2014 21:27:29 +0000 (16:27 -0500)]
mir: add timestamps to generated GdkEvents

We need this because it fixes menu activation. The menu activation code
looks at the time between events to determine if mouse clicks happen too
quickly.

11 years agomir: pass motion events down to transient children
William Hua [Mon, 1 Dec 2014 13:54:13 +0000 (08:54 -0500)]
mir: pass motion events down to transient children

11 years agomir: use parent_window's impl for GDK_INPUT_ONLY windows
William Hua [Fri, 28 Nov 2014 13:33:03 +0000 (08:33 -0500)]
mir: use parent_window's impl for GDK_INPUT_ONLY windows

Fixes a crash that occurs when the user selects text.

11 years agoscrolledwindow: Don't hover the scrollbars while a button is being pressed
Carlos Garnacho [Thu, 5 Feb 2015 16:16:21 +0000 (17:16 +0100)]
scrolledwindow: Don't hover the scrollbars while a button is being pressed

This avoids showing the scrollbars if we are positive the child widget
is being manipulated, regardless of the pointer being close to any of these.
On the next motion event after finishing the operation, the scrollbar will
be shown if necessary.

https://bugzilla.gnome.org/show_bug.cgi?id=743894

11 years agopopover: Unset the grab when the popover loses its parent
Carlos Garnacho [Tue, 3 Feb 2015 11:03:22 +0000 (12:03 +0100)]
popover: Unset the grab when the popover loses its parent

This ensures the grab and signal handlers are removed if the parent goes
NULL.

11 years agowindow: Unparent popover before unmapping/unrealizing
Carlos Garnacho [Mon, 2 Feb 2015 15:13:30 +0000 (16:13 +0100)]
window: Unparent popover before unmapping/unrealizing

Otherwise spurious invariants warnings may happen before the popover
is removed from the window.

11 years agowindow: call gtk_widget_unmap() on popover widgets unconditionally
Carlos Garnacho [Mon, 2 Feb 2015 14:34:17 +0000 (15:34 +0100)]
window: call gtk_widget_unmap() on popover widgets unconditionally

is_visible() wasn't the right check to perform here before unmapping,
and gtk_widget_unmap() already avoids being doubly called by checking
gtk_widget_get_mapped() anyway.

11 years agopopover: Keep a ref on the parent scrollable
Carlos Garnacho [Thu, 5 Feb 2015 13:17:51 +0000 (14:17 +0100)]
popover: Keep a ref on the parent scrollable

This ensures the scrollable is still alive on dispose(), and properly unset
after disconnecting all signal handlers.

11 years agoGtkStatusIcon: Expand deprecation docs
Matthias Clasen [Thu, 5 Feb 2015 09:42:57 +0000 (10:42 +0100)]
GtkStatusIcon: Expand deprecation docs

https://bugzilla.gnome.org/show_bug.cgi?id=743975

11 years agostylecontext: Don't crah when invalidate() is called on saved context
Benjamin Otte [Wed, 4 Feb 2015 20:08:54 +0000 (21:08 +0100)]
stylecontext: Don't crah when invalidate() is called on saved context

We need to invalidate the root node, not the current one.
Fixes crashes on startup of eclipse.

Testcase included.

https://bugzilla.redhat.com/show_bug.cgi?id=1185828

11 years agostylecontext: Do not have a default style
Benjamin Otte [Wed, 4 Feb 2015 17:26:32 +0000 (18:26 +0100)]
stylecontext: Do not have a default style

A bunch of applications expect the style to be as close to the actual
style as possible even during creation of the widget. Try to accomodate
these apps.

This reverses the effects of 27285f1f0fb00bb4507a8bf3e426bef940ae41af

https://bugzilla.gnome.org/show_bug.cgi?id=743395
https://bugzilla.redhat.com/show_bug.cgi?id=1185999

11 years agocss: Having a static style per screen isn't useful
Benjamin Otte [Wed, 4 Feb 2015 17:20:33 +0000 (18:20 +0100)]
css: Having a static style per screen isn't useful

Just have one static style for everything.

11 years agoAdd support for loading gtk3-only modules
Lars Uebernickel [Tue, 3 Feb 2015 10:01:29 +0000 (11:01 +0100)]
Add support for loading gtk3-only modules

Introduce GTK3_MODULES environment variable for modules that don't work
in gtk3. The list of modules is now $GTK3_MODULES:$GTK_MODULES.

https://bugzilla.gnome.org/show_bug.cgi?id=743917

11 years agoMSVC Builds: Rename "install" Projects
Chun-wei Fan [Wed, 4 Feb 2015 05:04:39 +0000 (13:04 +0800)]
MSVC Builds: Rename "install" Projects

Rename the "install" projects as "gtk-install" as we are planning to have a
grand solution file that incorporates all project files of the GTK+ stack
with their dependencies, to make it easier for people to build GTK+ from
scratch from a stock installation of Visual Studio 2008 and later.

11 years agoMSVC Builds: "Install" autocleanup headers
Chun-wei Fan [Wed, 4 Feb 2015 04:42:24 +0000 (12:42 +0800)]
MSVC Builds: "Install" autocleanup headers

These are public headers that are needed in the package, although the
macros are essentially no-ops for Visual Studio builds currently.

11 years agoAdd reference to HowDoI and getting-started
Bastian Ilsø [Mon, 2 Feb 2015 07:13:28 +0000 (08:13 +0100)]
Add reference to HowDoI and getting-started

11 years agoBump version to 3.15.5
Emmanuele Bassi [Mon, 2 Feb 2015 16:51:14 +0000 (16:51 +0000)]
Bump version to 3.15.5

We need the bump in order to allow projects to depend on newly added
API.

11 years agox11: Detect libinput touchpads
Carlos Garnacho [Sun, 1 Feb 2015 11:08:04 +0000 (12:08 +0100)]
x11: Detect libinput touchpads

These aren't reported as XIDependentTouch devices, so make it poke a
property that's specific to touchpads managed by the libinput driver.

11 years agogtkx: add missing header
Víctor Manuel Jáquez Leal [Mon, 2 Feb 2015 11:54:37 +0000 (12:54 +0100)]
gtkx: add missing header

The header file gdk/x11/gdkx-autocleanups.h is not installed, but
it is required by gdkx.h

https://bugzilla.gnome.org/show_bug.cgi?id=743640

11 years agoscrolledwindow: Don't hide indicators if leaving towards a child
Carlos Garnacho [Mon, 2 Feb 2015 11:58:56 +0000 (12:58 +0100)]
scrolledwindow: Don't hide indicators if leaving towards a child

In that case the scrolledwindow still gets events through the captured
handler, and can thus still control visibility, so hiding the indicators
in this case can only lead to flashing.

11 years agoUpdated Spanish translation
Daniel Mustieles [Mon, 2 Feb 2015 10:58:00 +0000 (11:58 +0100)]
Updated Spanish translation

11 years agogtkx: Add support for g_autoptr()
Alexander Larsson [Mon, 2 Feb 2015 08:34:43 +0000 (09:34 +0100)]
gtkx: Add support for g_autoptr()

https://bugzilla.gnome.org/show_bug.cgi?id=743640

11 years agogtkunixprint: Add support for g_autoptr()
Alexander Larsson [Mon, 2 Feb 2015 08:32:45 +0000 (09:32 +0100)]
gtkunixprint: Add support for g_autoptr()

https://bugzilla.gnome.org/show_bug.cgi?id=743640

11 years agogtk: Add support for g_autoptr()
Alexander Larsson [Mon, 2 Feb 2015 08:29:54 +0000 (09:29 +0100)]
gtk: Add support for g_autoptr()

https://bugzilla.gnome.org/show_bug.cgi?id=743640

11 years agogdk: Add support for g_autoptr()
Alexander Larsson [Mon, 2 Feb 2015 08:28:22 +0000 (09:28 +0100)]
gdk:  Add support for g_autoptr()

https://bugzilla.gnome.org/show_bug.cgi?id=743640

11 years agogtkstylecontext.c: Include fallback-c89.c
Chun-wei Fan [Mon, 2 Feb 2015 07:00:45 +0000 (15:00 +0800)]
gtkstylecontext.c: Include fallback-c89.c

round() is now being used in the function, so that compilers without
round() in their math.h is supported.

11 years agoBe safe against stray booleans
Matthias Clasen [Sun, 1 Feb 2015 22:30:10 +0000 (17:30 -0500)]
Be safe against stray booleans

Since gboolean is a typedef for int, the compiler won't complain
about gdk_window_set_event_compression (w, 2). So, make it work.

https://bugzilla.gnome.org/show_bug.cgi?id=742566

11 years agoPlug another memory leak in plugman
Matthias Clasen [Sun, 1 Feb 2015 16:07:12 +0000 (11:07 -0500)]
Plug another memory leak in plugman

11 years agoplugman: Fix leaking attribute string
Felix Riemann [Sat, 31 Jan 2015 15:17:24 +0000 (16:17 +0100)]
plugman: Fix leaking attribute string

https://bugzilla.gnome.org/show_bug.cgi?id=743770

11 years agoscrolledwindow: add a new edge-reached signal
Cosimo Cecchi [Sun, 25 Jan 2015 11:53:49 +0000 (11:53 +0000)]
scrolledwindow: add a new edge-reached signal

This will be used to just detect when an edge of the scrollable area is
reached - as opposed to the edge-overshot signal that is emitted when
the user scrolls past the edge.

https://bugzilla.gnome.org/show_bug.cgi?id=742848

11 years agoReduce symbolic icon warnings to a debug message
Matthias Clasen [Sat, 31 Jan 2015 18:33:42 +0000 (13:33 -0500)]
Reduce symbolic icon warnings to a debug message

These icons are out there in the wild, and the warning causes
distcheck to fail. So, reduce it to a debug message.

11 years agoUpdated Slovak translation
Dušan Kazik [Sat, 31 Jan 2015 11:07:59 +0000 (11:07 +0000)]
Updated Slovak translation

11 years agoimage: Optimize non-resize changes
Matthias Clasen [Sat, 31 Jan 2015 04:32:50 +0000 (23:32 -0500)]
image: Optimize non-resize changes

Suggested by Owen in

https://bugzilla.gnome.org/show_bug.cgi?id=613833

11 years agostylecontext: Use old rounding for get_padding/border/margin()
Benjamin Otte [Sat, 31 Jan 2015 01:25:18 +0000 (02:25 +0100)]
stylecontext: Use old rounding for get_padding/border/margin()

610f7e8f70ce165e04a4cdc3fd20b560f500b925 changed the way we query those
values for performance reasons but didn't copy the rounding used.

Fixes border-half-pixel reftest.

11 years agoglarea: Remove useless function
Benjamin Otte [Sat, 31 Jan 2015 00:55:04 +0000 (01:55 +0100)]
glarea: Remove useless function

The function should have been gone in
27cf0fa34c3a51ddb50391f21aaf2e3984d6dddb

11 years agoroundedbox: Treat path segments differently
Benjamin Otte [Thu, 22 Jan 2015 14:47:36 +0000 (15:47 +0100)]
roundedbox: Treat path segments differently

If a side of the box is 0px wide, make the corners owned by the adjacent
sides. This avoids spilling over of unwanted colors from the 0-width
side into the corner.

This is the fix that should actually have been part of
7e5ac3a58d0362080c4eb7c9b68ead99674c7b87 but my amazing amending skills
got rid of it when amending the test.

11 years agoUpdates for 3.15.5
Matthias Clasen [Fri, 30 Jan 2015 23:58:41 +0000 (18:58 -0500)]
Updates for 3.15.5

11 years agoFix an oversight
Matthias Clasen [Fri, 30 Jan 2015 23:42:06 +0000 (18:42 -0500)]
Fix an oversight

Commit ff256956b2072cd422 introduced a frame_clock_events_paused
flag, but only ever set it to TRUE, instead of unsetting it when
events are resumed. This was leading to assertion failures in
_gdk_display_unpause_events().

11 years agoGtkSettings: Consider default settings for font hinting and antialias
Matias De lellis [Wed, 3 Dec 2014 00:09:38 +0000 (21:09 -0300)]
GtkSettings: Consider default settings for font hinting and antialias

Arrange things so that gtk-xdg-hinting==-1 and gtk-xfg-antialias==-1
end up as CAIRO_HINT_STYLE_DEFAULT and CAIRO_ANTIALIAS_DEFAULT in the
cairo font options.

This will not change anything on Linux desktops where xsettings will
always provide values different from -1. But on other platforms, we
can benefit from getting the platform-specific defaults in cairo.

Based on the first patch in:
https://bugzilla.gnome.org/show_bug.cgi?id=735316

11 years agoGetting started: BIG update of basics section
Bastian Ilsø [Wed, 28 Jan 2015 18:43:45 +0000 (18:43 +0000)]
Getting started: BIG update of basics section

https://bugzilla.gnome.org/show_bug.cgi?id=743680

11 years agoResume events when disconnecting from a frame clock
Tom Hughes [Mon, 26 Jan 2015 21:52:00 +0000 (21:52 +0000)]
Resume events when disconnecting from a frame clock

If we are disconnecting from a frame clock that has paused event
processing and hasn't issued a resume yet make sure we resume the
events or they will stay blocked forever.

https://bugzilla.gnome.org/show_bug.cgi?id=742636

11 years agoUpdated Hungarian translation
Balázs Úr [Fri, 30 Jan 2015 21:29:35 +0000 (21:29 +0000)]
Updated Hungarian translation

11 years agoUpdated Hungarian translation
Balázs Úr [Fri, 30 Jan 2015 21:26:29 +0000 (21:26 +0000)]
Updated Hungarian translation

11 years agoRobustness fixes for drag cancel animation
Matthias Clasen [Fri, 30 Jan 2015 11:07:34 +0000 (06:07 -0500)]
Robustness fixes for drag cancel animation

gtk_widget_get_frame_clock() can return NULL. Deal with that
by silently stopping the drag cancel animation in this case.

11 years agotestnotebookdnd: Defer deletion to an idle
Matthias Clasen [Fri, 30 Jan 2015 14:20:04 +0000 (09:20 -0500)]
testnotebookdnd: Defer deletion to an idle

The code here was always a bit buggy: We removed the tab from
the notebook in a ::drag-data-received handler. But with
GTK_DEST_DEFAULT_DROP, that signal is emitted before we inform
the source side that the drag is finished. With its improved drag
handling, GtkNotebook now interprets this as a 'spontaneous'
removal of the tab being dragged, and cancels the drag, leading
to an unwanted cancel animation.

The easiest fix is to just defer the tab removal to an idle.

11 years agonotebook: Protect removal-from-within
Matthias Clasen [Fri, 30 Jan 2015 14:26:04 +0000 (09:26 -0500)]
notebook: Protect removal-from-within

When a tab is dropped, we have to remove it from the notebook to
insert it elsewhere. This is expected part of the tab dnd operation,
and we need to differentiate it from 'spontaneous' removals which
cause us to cancel the drag operation.

11 years agonotebook: Make dnd safe against tab disappearance
Matthias Clasen [Fri, 30 Jan 2015 04:51:26 +0000 (23:51 -0500)]
notebook: Make dnd safe against tab disappearance

when the dragged tab is being removed during the drag operation,
we need to cancel the drag. To do so, we have to keep around
a reference to the drag context.

This should fix
https://bugzilla.gnome.org/show_bug.cgi?id=732051

11 years agonotebook: Make reordering safe for tab disappearance
Matthias Clasen [Fri, 30 Jan 2015 04:48:06 +0000 (23:48 -0500)]
notebook: Make reordering safe for tab disappearance

If the dragged tab is removed during reordering, we need to
call gtk_notebook_stop_reorder(), to clean up and bring the
remaining tabs back in order.

11 years agodnd: Allow programmatic drag cancellation
Matthias Clasen [Fri, 30 Jan 2015 04:49:48 +0000 (23:49 -0500)]
dnd: Allow programmatic drag cancellation

Add a gtk_drag_cancel() function that can be used on the
source side to cancel an ongoing drag operation. This can
be useful, e.g. if the data that is being dragged becomes
unavailable.

11 years agodnd: Make dest side safe against destroy
Matthias Clasen [Fri, 30 Jan 2015 04:45:28 +0000 (23:45 -0500)]
dnd: Make dest side safe against destroy

Use a weak pointer to clean up when the current drag destination
gets destroyed.

11 years agofile chooser button: Make dnd work as expected
Matthias Clasen [Thu, 29 Jan 2015 03:36:47 +0000 (22:36 -0500)]
file chooser button: Make dnd work as expected

Update the button contents when accepting a drop.

https://bugzilla.gnome.org/show_bug.cgi?id=743451

11 years agoUpdated Icelandic translation
Sveinn í Felli [Thu, 29 Jan 2015 15:37:35 +0000 (15:37 +0000)]
Updated Icelandic translation

11 years agoAdwaita: add comments
Lapo Calamandrei [Thu, 29 Jan 2015 10:34:01 +0000 (11:34 +0100)]
Adwaita: add comments

11 years agoAdwaita: undershoot simplification and tweaks
Lapo Calamandrei [Thu, 29 Jan 2015 09:49:59 +0000 (10:49 +0100)]
Adwaita: undershoot simplification and tweaks

11 years agofile chooser: Unset sidebar during search
Matthias Clasen [Wed, 28 Jan 2015 23:26:42 +0000 (18:26 -0500)]
file chooser: Unset sidebar during search

It is a bit misleading to have one of the places in the sidebar
selected while in search mode, since the search is not relative
to any location. So, unset the sidebar selection when we enter
search mode.

11 years agoplaces sidebar: Make a function NULL-safe
Matthias Clasen [Wed, 28 Jan 2015 23:25:32 +0000 (18:25 -0500)]
places sidebar: Make a function NULL-safe

gtk_places_sidebar_set_location's location argument is documented
as allow-none, but if one actually passes NULL, a warning results.
Fix that.

11 years agofile chooser: Unify search mode leaving
Matthias Clasen [Wed, 28 Jan 2015 23:21:18 +0000 (18:21 -0500)]
file chooser: Unify search mode leaving

Make all ways to get out of search mode go through the same
code path, so we always populate the list.

11 years agofile chooser: Fix key bindings
Matthias Clasen [Wed, 28 Jan 2015 23:17:51 +0000 (18:17 -0500)]
file chooser: Fix key bindings

Key press handlers must chain up for key bindings to work.

11 years agofile chooser: Load a directory when leaving search
Matthias Clasen [Wed, 28 Jan 2015 23:15:19 +0000 (18:15 -0500)]
file chooser: Load a directory when leaving search

Ensure that some folder contents are shown in
the list when we leave search mode.

https://bugzilla.gnome.org/show_bug.cgi?id=743659

11 years agofile chooser: Don't switch too eagerly to search mode
Matthias Clasen [Wed, 28 Jan 2015 22:01:30 +0000 (17:01 -0500)]
file chooser: Don't switch too eagerly to search mode

If the search entry doesn't consume the key event, don't
change to search mode. Otherwise, we interfere e.g. with
Ctrl and Shift keys for selection.

https://bugzilla.gnome.org/show_bug.cgi?id=743660

11 years agoFix VA marshaller for GtkWidget's "event-after" signal
Garrett Regier [Tue, 27 Jan 2015 17:22:26 +0000 (09:22 -0800)]
Fix VA marshaller for GtkWidget's "event-after" signal

Just a copy and paste typo from the "event" signal.

https://bugzilla.gnome.org/show_bug.cgi?id=743593
Signed-off-by: Garrett Regier <garrettregier@gmail.com>
11 years agoGet started: port example-0.c to GtkApplication
Bastian Ilsø [Wed, 28 Jan 2015 11:39:01 +0000 (11:39 +0000)]
Get started: port example-0.c to GtkApplication

https://bugzilla.gnome.org/show_bug.cgi?id=743638

11 years agoGtkAboutDialog: Fix crash on link click
Christian Persch [Tue, 27 Jan 2015 12:37:15 +0000 (13:37 +0100)]
GtkAboutDialog: Fix crash on link click

https://bugzilla.gnome.org/show_bug.cgi?id=743592

11 years agofile chooser: Make Escape go out of search mode
Matthias Clasen [Wed, 28 Jan 2015 21:06:02 +0000 (16:06 -0500)]
file chooser: Make Escape go out of search mode

This works in most other places where we have search,
and is the expected behavior.

11 years agofile chooser: Simplify search entry callback
Matthias Clasen [Wed, 28 Jan 2015 21:02:03 +0000 (16:02 -0500)]
file chooser: Simplify search entry callback

11 years agofile chooser: Drop unused callbacks
Matthias Clasen [Wed, 28 Jan 2015 20:57:12 +0000 (15:57 -0500)]
file chooser: Drop unused callbacks

11 years agofile chooser: Remove the search/browse buttons
Matthias Clasen [Wed, 28 Jan 2015 20:37:53 +0000 (15:37 -0500)]
file chooser: Remove the search/browse buttons

Enabling search is now handled in the file chooser dialog,
so we don't need these buttons anymore.

11 years agofile chooser dialog: Add a search button to the header
Matthias Clasen [Wed, 28 Jan 2015 20:36:57 +0000 (15:36 -0500)]
file chooser dialog: Add a search button to the header

This is where we put similar search buttons in applications
and in other dialogs, such as the GtkAppChooserDialog.

11 years agofile chooser: Add a search-mode property
Matthias Clasen [Wed, 28 Jan 2015 20:00:08 +0000 (15:00 -0500)]
file chooser: Add a search-mode property

This will be used to control search from the dialog.

11 years agofile chooser: Clean up includes
Matthias Clasen [Wed, 28 Jan 2015 15:35:49 +0000 (10:35 -0500)]
file chooser: Clean up includes

11 years agoAdwaita: undershoot tweaks
Lapo Calamandrei [Wed, 28 Jan 2015 19:22:05 +0000 (20:22 +0100)]
Adwaita: undershoot tweaks

11 years agoAdwaita: undershoot teeth removal
Lapo Calamandrei [Wed, 28 Jan 2015 19:05:03 +0000 (20:05 +0100)]
Adwaita: undershoot teeth removal

11 years agoAdwaita: shade content-view
Jakub Steiner [Wed, 28 Jan 2015 17:22:59 +0000 (18:22 +0100)]
Adwaita: shade content-view

https://bugzilla.gnome.org/show_bug.cgi?id=740336

11 years agoAdwaita: set the right background color for message dialogs
Lapo Calamandrei [Wed, 28 Jan 2015 17:19:20 +0000 (18:19 +0100)]
Adwaita: set the right background color for message dialogs

11 years agowindow: Only claim the sequence if keeping the grab
Carlos Garnacho [Sun, 25 Jan 2015 17:55:01 +0000 (17:55 +0000)]
window: Only claim the sequence if keeping the grab

If the grab belongs elsewhere, the window won't claim the sequence right
away. The sequence may still be claimed afterwards when window dragging
starts, but simple clicks won't be consumed this way.

This makes it possible to close popovers when clicking on the title region,
while still permitting touch/button 1 interaction for every other purpose.

https://bugzilla.gnome.org/show_bug.cgi?id=743257