Carlos Garnacho [Mon, 14 Dec 2015 14:43:54 +0000 (15:43 +0100)]
broadway: Use GdkSeatDefault to Implement GdkSeat
Carlos Garnacho [Wed, 9 Dec 2015 11:56:26 +0000 (12:56 +0100)]
wayland: Improve creation of windowing surface roles
We no longer need a grabbed seat, instead we'll just use the default
seat if this happens, not without first warning and recommending
gdk_seat_grab() for the operation.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Wed, 9 Dec 2015 11:55:39 +0000 (12:55 +0100)]
wayland: Replace deprecated functions
We can just gdk_seat_ungrab() here.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Wed, 9 Dec 2015 11:29:52 +0000 (12:29 +0100)]
gdk: Deprecate GdkDeviceManager and gdk_device_grab/ungrab()
GdkSeat is now the preferred way to deal with input devices and grabs.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Wed, 9 Dec 2015 11:49:43 +0000 (12:49 +0100)]
wayland: Emit cancelled on touchpoint used on window dragging/moving
This allows GDK to unset the grab itself. Also, make sure we unset
the "pointer emulating" touch on the device if this is the
pointer emulating sequence.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Wed, 9 Dec 2015 11:46:08 +0000 (12:46 +0100)]
wayland: Unset "pointer emulating" touch on wl_touch.cancel
And emit the corresnponding leave event on its master pointer.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Wed, 9 Dec 2015 11:39:43 +0000 (12:39 +0100)]
gdk: Manage GDK_TOUCH_CANCEL events on gdk_windowing_got_event()
These events must get active/implicit grabs undone, and can be done
on client-side code.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Wed, 9 Dec 2015 11:42:33 +0000 (12:42 +0100)]
wayland: Make gdk_wayland_device_get_focus() work on touch
So we can figure out the focus for the master device.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Wed, 9 Dec 2015 11:25:10 +0000 (12:25 +0100)]
GdkDisplay: Add ::seat-added/removed signals
These will be emitted as seats come and go.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:40:56 +0000 (19:40 +0100)]
wayland: Add GdkSeat implementation
GdkWaylandDeviceData conceptually gathers the data that belongs to
a seat, so it's been renamed (although the old typedef stays, plenty
of refactoring is due here...).
The methods in GdkSeatClass have also been implemented, the most
remarkable is ::grab, which ensures the grab is performed on all
the relevant "master" devices.
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:56:41 +0000 (19:56 +0100)]
GtkTreeView: Use gdk_seat_grab()
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:56:25 +0000 (19:56 +0100)]
GtkNotebook: Use gdk_seat_grab()
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:55:45 +0000 (19:55 +0100)]
GtkEntryCompletion: Use gdk_seat_grab()
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:55:15 +0000 (19:55 +0100)]
GtkComboBox: Use gdk_seat_grab()
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:54:54 +0000 (19:54 +0100)]
GtkMenu: Use gdk_seat_grab()
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:54:31 +0000 (19:54 +0100)]
GtkCellRendererAccel: Use gdk_seat_grab()
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:54:01 +0000 (19:54 +0100)]
GtkButton: Use gdk_seat_grab()
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:53:36 +0000 (19:53 +0100)]
x11: Use GdkSeatDefault to implement GdkSeat
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Mon, 14 Dec 2015 19:24:40 +0000 (20:24 +0100)]
x11: Use gdk_event_set_device() consistently
Move the variable definitions above the function, and use those
throughout all branches of the event handling switch.
Carlos Garnacho [Thu, 26 Nov 2015 18:53:09 +0000 (19:53 +0100)]
GdkEvent: Add GdkSeat getter and internal setter
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:52:23 +0000 (19:52 +0100)]
GdkDisplay: Add GdkSeat getters
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:50:57 +0000 (19:50 +0100)]
GdkDevice: Add GdkSeat property and getter
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:49:10 +0000 (19:49 +0100)]
gdk: Add GdkSeatDefault
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Thu, 26 Nov 2015 18:47:38 +0000 (19:47 +0100)]
gdk: Add GdkSeat
https://bugzilla.gnome.org/show_bug.cgi?id=759309
Carlos Garnacho [Fri, 11 Dec 2015 12:16:13 +0000 (13:16 +0100)]
x11: Use client pointer as default core pointer for gtk+2 API
There is no need to fiddle with the master devices list.
https://bugzilla.gnome.org/show_bug.cgi?id=759323
Lucas Baudin [Sun, 13 Dec 2015 18:24:50 +0000 (19:24 +0100)]
wayland: recursively search for the settings schema.
On some systems, the gtk settings are not used properly for wayland.
Indeed, g_settings_schema_source_get_default is used, and as the docs says it,
"all lookups performed against the default source should probably be done
recursively.".
https://bugzilla.gnome.org/show_bug.cgi?id=759409
Dušan Kazik [Mon, 14 Dec 2015 10:53:55 +0000 (10:53 +0000)]
Updated Slovak translation
Piotr Drąg [Mon, 14 Dec 2015 00:45:11 +0000 (01:45 +0100)]
Updated POTFILES.in
Matthias Clasen [Sun, 13 Dec 2015 23:24:41 +0000 (18:24 -0500)]
dnd: Cleanups
Remove some now unused includes and dead code, and rename
gtk_drag_set_icon_window to gtk_drag_set_icon_widget_internal,
since it is no longer restricted to toplevel windows.
Matthias Clasen [Sun, 13 Dec 2015 21:01:05 +0000 (16:01 -0500)]
testdnd2: Add some verbosity
This makes it easier to see that we're missing some signals
under Wayland.
Matthias Clasen [Mon, 7 Dec 2015 20:08:27 +0000 (15:08 -0500)]
dnd: Stop doing cancel animation in GTK+
Under Wayland, the compositor does it, so there is no need
for us to move the window ourselves. For X11, we are now
doing the animation from the X11 backend. Trigger that by
calling gdk_drag_drop_done().
What changes here is that we have to keep the icon_window
alive for as long as the drag context exists. Use a weak
reference to do so.
Matthias Clasen [Mon, 7 Dec 2015 23:28:06 +0000 (18:28 -0500)]
x11: Implement drag cancel animation
Showing the drag cancel animation can be done in the X11
drag context implementation now that we hold the drag
window there, and have the start coordinates.
Since we can't control if and when the application destroys
the drag widget, we take a snapshot of the window contents
and display that during the animation. This should be good
enough for all practical purposes.
Matthias Clasen [Mon, 7 Dec 2015 20:07:13 +0000 (15:07 -0500)]
gdk: Add gdk_drag_drop_done
This will allow us to move the drag cancel animation to GDK.
For now, it does nothing.
Matthias Clasen [Mon, 7 Dec 2015 20:07:54 +0000 (15:07 -0500)]
x11: Store drag start coordinates
These will be used in later commits.
Matthias Clasen [Mon, 7 Dec 2015 19:21:20 +0000 (14:21 -0500)]
dnd: Pass start coordinates when creating the drag context
This will be used in subsequent commits.
Matthias Clasen [Mon, 7 Dec 2015 18:47:45 +0000 (13:47 -0500)]
gdk: Allow passing the start coordinates in drag_begin
Add a variant of gdk_drag_begin that takes the start position
in addition to the device. All backend implementation have been
updated to accept (and ignore) the new arguments.
Subsequent commits will make use of the data in some backends.
Benjamin Otte [Sun, 13 Dec 2015 15:02:34 +0000 (16:02 +0100)]
cssnode: Set new style if it's animated
In commit
2c613166771ea9118ebda91c311f11b3462330fb we avoided emitting
the style-changed signal if no CSS property changed.
Unfortunately, this also caused CSS styles to not be updated when
animations started if those animations did not change any CSS value
immediately. In those cases the animation would just never start.
The obvious example was the spinner.
Daniel Mustieles [Sun, 13 Dec 2015 12:31:25 +0000 (13:31 +0100)]
Updated Spanish translation
Daniel Mustieles [Sun, 13 Dec 2015 12:31:15 +0000 (13:31 +0100)]
Updated Spanish translation
Pedro Albuquerque [Sun, 13 Dec 2015 08:16:32 +0000 (08:16 +0000)]
Updated Portuguese translation
Benjamin Otte [Sun, 13 Dec 2015 03:08:44 +0000 (04:08 +0100)]
widget: Pass a GtkCssStyleChange instead of a bitmask
Timm Bäder [Fri, 11 Dec 2015 09:05:37 +0000 (10:05 +0100)]
Gestures: Add some nullable annotations
Timm Bäder [Thu, 10 Dec 2015 15:54:59 +0000 (16:54 +0100)]
GtkFlowBox: Add missing nullable annotation
Timm Bäder [Wed, 9 Dec 2015 16:15:04 +0000 (17:15 +0100)]
GtkFlowBox: Fix copy/paste mistake in docs
Benjamin Otte [Sat, 12 Dec 2015 02:24:32 +0000 (03:24 +0100)]
cellrendererpixbuf: Don't set icon size for pixbufs
The code before the refactorings didn't set it, so we now don't set it
either.
Fixes images being way too smal inside iconviews.
Benjamin Otte [Sat, 12 Dec 2015 01:09:01 +0000 (02:09 +0100)]
cssnode: Catch case where a style didn't change
Catch the case where a CSS style did not change and don't emit the
style-changed signal in that case.
This saves not only the emission of the signal, but also doesn't cause
invalidation in child nodes, which would previously get a PARENT_STYLE
Benjamin Otte [Sat, 12 Dec 2015 01:02:04 +0000 (02:02 +0100)]
cssnode: Change style-changed signal
Instead of having old and new style, now have a GtkCssStyleChange opaque
object that will compute the changes you are interested in for you.
This simplifies change signal handlers quite a bit and avoids lots of
repeated computation in every signal handler.
Matthias Clasen [Fri, 11 Dec 2015 22:11:07 +0000 (17:11 -0500)]
inspector: Hide request mode for non-widgets
Only widgets have a request mode.
Lapo Calamandrei [Fri, 11 Dec 2015 20:08:53 +0000 (21:08 +0100)]
Adwaita: menu color change on the dark variant only
Matthias Clasen [Fri, 11 Dec 2015 17:25:31 +0000 (12:25 -0500)]
file chooser: Also apply the settings
I forgot that the settings object we're dealing with here
is in delay mode, so settings won't apply immediately.
Matthias Clasen [Fri, 11 Dec 2015 17:15:03 +0000 (12:15 -0500)]
file chooser: Store size more frequently
We were only storing the dialog size on unmap, but resetting to the
stored default value more often, e.g. on focus-out. This was causing
the dialog to 'jump back' to its remembered size after the user
manually resized it, leading to frustration and bug reports.
Instead, save the dialog size on every ::size-allocate of the toplevel.
To avoid needlessly spamming dconf, only write the new value if it
changed.
Matthias Clasen [Fri, 11 Dec 2015 15:50:25 +0000 (10:50 -0500)]
Add a drag-end callback
This will be helpful in figuring out life-cycle issues
with icon widgets.
Matthias Clasen [Fri, 11 Dec 2015 15:44:46 +0000 (10:44 -0500)]
Add a dnd testcase
This has tests for drag images, widgets, hotspots.
Jonas Ådahl [Tue, 8 Dec 2015 10:19:33 +0000 (18:19 +0800)]
wayland: Implement DND icon hotspot API
In Wayland, the hotspot of a DND icon is set using the buffer offset in
wl_buffer.attach. To implement this, add a private API to cause the
next wl_surface.attach to offset the new buffer with a given offset.
Setting a DND icon hotspot sets this offset while also queuing a redraw
of the window to trigger the wl_surface.attach.
https://bugzilla.gnome.org/show_bug.cgi?id=759168
Georges Basile Stavracas Neto [Fri, 11 Dec 2015 04:28:39 +0000 (02:28 -0200)]
placesviewrow: remove unneeded functions
These function declarations are not used nor implemented
anywhere, so remove them.
Matt Watson [Thu, 10 Dec 2015 23:50:25 +0000 (15:50 -0800)]
gtkstack: remove needless queue_resize
In gtk_stack_set_transition_position we should only need to
queue_resize if we are non-homogeneous, which is already done
earlier in the function.
Georges Basile Stavracas Neto [Tue, 3 Nov 2015 02:50:06 +0000 (00:50 -0200)]
placesview: implement available space
GtkPlacesView is a widget to display locations
in the computer, such as root ("/") and volumes,
separating the persistent devices from removable
ones.
From the latest mockups[1], GtkPlacesView would
display the available space of local drives like
partitions. This, however, is not implemented in
the current codebase.
Fix that by implementing the measurement of disk
space, and adding a new property GtkPlacesView::show-disk-usage
which controls the visibility of measured disks.
[1] https://raw.githubusercontent.com/gnome-design-team/gnome-mockups/master/nautilus/nautilus-next/other-locations.png
https://bugzilla.gnome.org/show_bug.cgi?id=759225
Cosimo Cecchi [Wed, 9 Dec 2015 18:45:26 +0000 (10:45 -0800)]
statusicon: plug memory leak
Lapo Calamandrei [Wed, 9 Dec 2015 17:56:38 +0000 (18:56 +0100)]
Adwaita: make menu bg color the same as popovers
fixes https://bugzilla.gnome.org/show_bug.cgi?id=759260
Lapo Calamandrei [Wed, 9 Dec 2015 16:43:40 +0000 (17:43 +0100)]
Adwaita: another try at color tweakings...
Lapo Calamandrei [Wed, 9 Dec 2015 16:38:38 +0000 (17:38 +0100)]
Adwaita: some more color tweaks and sass simplification
Lapo Calamandrei [Wed, 9 Dec 2015 13:52:03 +0000 (14:52 +0100)]
Adwaita: fix popover styling
...in the process simplified the touch-selection styling, check
and radios not fixed there since I'm going to add proper osd assets
for those (istead of forcing the dark variant assets there as before).
Benjamin Otte [Wed, 9 Dec 2015 12:29:59 +0000 (13:29 +0100)]
iconhelper_ Don't use deprecated functions
Use gtk_widget_get_direction() instead of
gtk_style_context_get_direction().
Benjamin Otte [Tue, 8 Dec 2015 19:57:24 +0000 (20:57 +0100)]
iconhelper: Remove the context argument
Instead, use the widget's context.
Benjamin Otte [Tue, 8 Dec 2015 14:52:35 +0000 (15:52 +0100)]
iconhelper: Get rid of _gtk_icon_helper_set_window()
Instead use the window of the owner widget.
Benjamin Otte [Tue, 8 Dec 2015 14:39:10 +0000 (15:39 +0100)]
iconhelper: Require a widget as construction argument
Note that the caller needs to ensure the widget lives as long as the
iconhelper as the iconhelper will not ref the widget.
Benjamin Otte [Tue, 8 Dec 2015 14:18:55 +0000 (15:18 +0100)]
statusicon: Port from iconhelper to imagedefinition
Benjamin Otte [Tue, 8 Dec 2015 10:27:37 +0000 (11:27 +0100)]
cellrendererpixbuf: Create iconhelper on-demand
Instead of storing one in the priv structure, only store an image
definition there.
This will allow future refactorings of the icon helper.
Benjamin Otte [Mon, 7 Dec 2015 18:49:51 +0000 (19:49 +0100)]
iconhelper: Pass only the CssStyle when loading pixbufs
Benjamin Otte [Mon, 7 Dec 2015 18:47:49 +0000 (19:47 +0100)]
iconhelper: Remove unused GtkStyleContext parameters
Benjamin Otte [Mon, 7 Dec 2015 18:45:55 +0000 (19:45 +0100)]
iconhelper: Pass only the CssStyle when loading iconsets
Benjamin Otte [Mon, 7 Dec 2015 18:22:48 +0000 (19:22 +0100)]
iconfactory: Move private functions into private header
Benjamin Otte [Mon, 7 Dec 2015 03:44:39 +0000 (04:44 +0100)]
iconhelper: Pass only the CssStyle when loading icons
This is the first step in replacing StyleContext usage with CssNode
usage.
Matthias Clasen [Wed, 9 Dec 2015 01:47:21 +0000 (20:47 -0500)]
toolitem: Set a css name
Sébastien Wilmet [Sun, 6 Dec 2015 16:49:01 +0000 (17:49 +0100)]
Don't use gtk_text_iter_begins_tag() (deprecated)
Use gtk_text_iter_starts_tag() instead.
https://bugzilla.gnome.org/show_bug.cgi?id=759092
Sébastien Wilmet [Sun, 6 Dec 2015 16:39:27 +0000 (17:39 +0100)]
textiter: add starts_tag() and deprecate begins_tag()
The name gtk_text_*_begins_* was used only for begins_tag(). All other
similar functions use "starts": starts_line(), starts_word(), etc.
So for consistency, add gtk_text_iter_starts_tag() and deprecate
gtk_text_iter_begins_tag().
Also change (allow-none) to (nullable), to use the new annotation.
https://bugzilla.gnome.org/show_bug.cgi?id=759092
Matthias Clasen [Tue, 8 Dec 2015 18:02:37 +0000 (13:02 -0500)]
Revert "wayland: Implement DND icon hotspot API"
This reverts commit
3ab9d96623bc30678cbffc3bf9ec5133e714e87c.
Pushed prematurely.
Kalev Lember [Tue, 8 Dec 2015 17:51:05 +0000 (18:51 +0100)]
Remove GtkShortcutsGesture from docs
Commit
c83ab24 that dropped GtkShortcutsGesture left in one last reference in
docs; this drops it as well.
Matthias Clasen [Tue, 8 Dec 2015 17:32:00 +0000 (12:32 -0500)]
gtk-demo: Add an example for direction-dependent shortcuts
Jonas Ådahl [Tue, 8 Dec 2015 10:19:33 +0000 (18:19 +0800)]
wayland: Implement DND icon hotspot API
In Wayland, the hotspot of a DND icon is set using the buffer offset in
wl_buffer.attach. To implement this, add a private API to cause the
next wl_surface.attach to offset the new buffer with a given offset.
Setting a DND icon hotspot sets this offset while also queuing a redraw
of the window to trigger the wl_surface.attach.
https://bugzilla.gnome.org/show_bug.cgi?id=759168
Khaled Hosny [Tue, 8 Dec 2015 06:37:51 +0000 (10:37 +0400)]
Update Arabic translation
Kalev Lember [Mon, 7 Dec 2015 16:21:51 +0000 (17:21 +0100)]
headerbar: Don't leak internal widgets
Make sure to clean up internal widgets in destroy, as these aren't
cleaned up when GtkContainer walks the childrens list in
gtk_container_destroy().
This also fixes a gedit crash as reported in
https://bugzilla.redhat.com/show_bug.cgi?id=
1288669
https://bugzilla.gnome.org/show_bug.cgi?id=759132
Matthias Clasen [Tue, 8 Dec 2015 02:57:45 +0000 (21:57 -0500)]
Trivial code restructuring
Matthias Clasen [Tue, 8 Dec 2015 02:54:55 +0000 (21:54 -0500)]
dnd: pass hotspot to GDK
This uses the new api introduced in the previous commits.
Matthias Clasen [Tue, 8 Dec 2015 02:53:38 +0000 (21:53 -0500)]
wayland: Add a dummy implementation of gdk_drag_context_set_hotspot
This just records the hotspot coordinates. Still to do:
apply the hotspot when updating the drag window.
Matthias Clasen [Tue, 8 Dec 2015 02:52:51 +0000 (21:52 -0500)]
x11: Implement gdk_drag_context_set_hotspot
This makes the drag behavior under X11 match what we had previously.
Matthias Clasen [Tue, 8 Dec 2015 02:52:03 +0000 (21:52 -0500)]
gdk: Add api to set drag window hotspot
With GdkDragContext now being in charge of placing the window,
it needs to know about the hotspot to place under the cursor.
Matthias Clasen [Mon, 7 Dec 2015 23:32:57 +0000 (18:32 -0500)]
dnd: Remove an unused struct field
We only have one screen nowadays, so the fallback_icon is
never used.
Carlos Garnacho [Mon, 7 Dec 2015 15:32:15 +0000 (16:32 +0100)]
GtkNotebook: Use gdouble in coordinate calculations
Otherwise rounding errors fool the "tab under coordinates" checks on
crossing events, which will be triggered close enough to the window
rectangle if the pointer moves slowly enough.
With this, the tab_prelight() function correctly figures out we've
moved the pointer outside the tab area when called in
gtk_notebook_leave_notify().
https://bugzilla.gnome.org/show_bug.cgi?id=759091
Adam Reichold [Sun, 6 Dec 2015 10:30:58 +0000 (11:30 +0100)]
Fix method to update menu sensitivity of combo box
After removal of the selectable header and separator from the combo box,
the method to update the menu sensitivity must be changed as it assumes
at least two items within each sub menu and contains special handling
for the separator. Removing this fixes bug #759079.
Matthias Clasen [Mon, 7 Dec 2015 15:11:06 +0000 (10:11 -0500)]
Avoid ugly seams on half-tiled terminals
Since we're no longer doing geometry widgets, don't send
base size and increments to the window manager anymore either.
This avoids an ugly 2 pixel gap to the right and bottom of half-tiled
terminals under gnome-shell.
Matthias Clasen [Mon, 7 Dec 2015 12:04:43 +0000 (07:04 -0500)]
accel cellrenderer: Fix rendering
We were getting the "New Accelerator" text mixed with the
content of the underlying cell, since plain labels don't
have a background. Go back to putting the label in selected
state, and fix the theme to render that white-on-blue. This
was lost when we switched to using a selection sub-node.
Erika [Sun, 6 Dec 2015 14:31:00 +0000 (14:31 +0000)]
Correct annotations for gtk_cell_layout_get_area
The function can return NULL.
https://bugzilla.gnome.org/show_bug.cgi?id=759081
Matthias Clasen [Mon, 7 Dec 2015 04:06:33 +0000 (23:06 -0500)]
inspector: Redo the CSS node page
Showing two lists in a paned was a bit awkward, and space was
getting too limited. Go back to showing just the node list at
first, and make the CSS properties available via a stack. At
the same time, add a right-click context menu to the node list
to make the name and class editing more easily available.
Matthias Clasen [Sun, 6 Dec 2015 22:25:31 +0000 (17:25 -0500)]
inspector: Avoid a crash
gkt_cell_layout_get_area may return NULL, handle it.
Matthias Clasen [Sun, 6 Dec 2015 20:40:59 +0000 (15:40 -0500)]
css node: Some more docs
Matthias Clasen [Sun, 6 Dec 2015 00:18:27 +0000 (19:18 -0500)]
places view: Use names for cursors
Use the standard name for the busy+interactive cursor.
Matthias Clasen [Sun, 6 Dec 2015 00:17:52 +0000 (19:17 -0500)]
entry: Always use cursor names
The names are the preferred API for cursors, so use it.
Matthias Clasen [Sun, 6 Dec 2015 00:09:17 +0000 (19:09 -0500)]
Add a test for required cursors
We use these cursors in GTK+, so check for their existence.
Matthias Clasen [Sat, 5 Dec 2015 23:55:05 +0000 (18:55 -0500)]
Be forgiving if cursors are missing
No need to crash here. Missing cursors are ugly, but we
shouldn't crash.