Matthias Clasen [Mon, 7 Mar 2016 02:41:46 +0000 (21:41 -0500)]
inspector: Drop the per-object CSS editor
The global editor is good enough for all uses.
Cosimo Cecchi [Sun, 6 Mar 2016 19:08:49 +0000 (11:08 -0800)]
range: remove unneeded gtk_widget_queue_draw()
Cosimo Cecchi [Sun, 6 Mar 2016 18:51:29 +0000 (10:51 -0800)]
range: rename function
The function queues an allocation now, not a draw.
Cosimo Cecchi [Sun, 6 Mar 2016 18:41:34 +0000 (10:41 -0800)]
range: remove duplicated code
This is already called by range_grab_add().
Cosimo Cecchi [Sun, 6 Mar 2016 18:36:43 +0000 (10:36 -0800)]
scale: trivial code cleanup
Cosimo Cecchi [Sun, 6 Mar 2016 08:27:58 +0000 (00:27 -0800)]
scale: deprecate value-spacing style property
Nothing uses it now.
Cosimo Cecchi [Sun, 6 Mar 2016 05:08:47 +0000 (21:08 -0800)]
scale: port scale values to gadgets
And add a default color like it was before.
This also fixes other issues with scale values interacting with scale
mark labels, which were buggy at least since 3.18.
Balázs Úr [Sun, 6 Mar 2016 18:27:26 +0000 (18:27 +0000)]
Updated Hungarian translation
Matthias Clasen [Sun, 6 Mar 2016 15:23:49 +0000 (10:23 -0500)]
Minor documentation improvement
https://bugzilla.gnome.org/show_bug.cgi?id=583780
Matthias Clasen [Sun, 6 Mar 2016 04:41:10 +0000 (23:41 -0500)]
spin button: Limit the entry width to reasonable values
When opening the value editor for any GtkAdjustment properties
in the inspector, the popover stretches out for miles, since
it reserves enough space to draw MAXDOUBLE. This is not useful.
Limit the space we reserve to 8 digits.
Matthias Clasen [Sun, 6 Mar 2016 04:36:54 +0000 (23:36 -0500)]
gtk-demo: Use <accelerator>
Add an example of <accelerator> use in the builder example.
Matthias Clasen [Sun, 6 Mar 2016 04:34:50 +0000 (23:34 -0500)]
builder: Allow 'primary' as a modifier value
Support aliases for flags values in GtkBuilder, and use this
to support the special value 'primary' for GdkModifierType
values.
https://bugzilla.gnome.org/show_bug.cgi?id=699891
Cosimo Cecchi [Sun, 6 Mar 2016 03:08:14 +0000 (19:08 -0800)]
range: simplify calculation
Cosimo Cecchi [Sun, 6 Mar 2016 03:07:55 +0000 (19:07 -0800)]
range: move declarations to inner block
Where they're needed.
Cosimo Cecchi [Sun, 6 Mar 2016 03:04:41 +0000 (19:04 -0800)]
range: use a fixed offset for mark "snap" size
Instead of making it dependent on the slider size.
Cosimo Cecchi [Sun, 6 Mar 2016 02:56:14 +0000 (18:56 -0800)]
Revert "range: use border box for slider area"
Since it causes problems with event coordinates.
This reverts commit
0883ff5eedf73b1197f2a49fb7e55ce227917335.
Anders Jonsson [Sat, 5 Mar 2016 23:03:05 +0000 (23:03 +0000)]
Updated Swedish translation
Matej Urbančič [Sat, 5 Mar 2016 21:37:17 +0000 (22:37 +0100)]
Updated Slovenian translation
Matej Urbančič [Sat, 5 Mar 2016 21:33:50 +0000 (22:33 +0100)]
Updated Slovenian translation
Matthias Clasen [Sat, 5 Mar 2016 20:48:40 +0000 (15:48 -0500)]
Make the settings test fail
Small thinko: We only get to see the test artifacts in
continuous if the test fails. So make it fail, temporarily.
Cosimo Cecchi [Sat, 5 Mar 2016 20:04:52 +0000 (12:04 -0800)]
Adwaita: fix glitch in fine-tune scrollbar theming
The fine-tune mode should not change the overall dimensions of the
scrollbar.
Stas Solovey [Sat, 5 Mar 2016 19:56:37 +0000 (19:56 +0000)]
Updated Russian translation
Matej Urbančič [Sat, 5 Mar 2016 19:56:06 +0000 (20:56 +0100)]
Updated Slovenian translation
Matej Urbančič [Sat, 5 Mar 2016 19:49:05 +0000 (20:49 +0100)]
Updated Slovenian translation
Aurimas Černius [Sat, 5 Mar 2016 19:37:17 +0000 (21:37 +0200)]
Updated Lithuanian translation
Paolo Borelli [Sat, 5 Mar 2016 18:02:15 +0000 (19:02 +0100)]
win32: use a struct for the event source
This removes the event_poll_fd global variable and the (ab)use of
get_default_display. It is also more consistent with other backends.
Also store display
Paolo Borelli [Sat, 5 Mar 2016 18:43:32 +0000 (19:43 +0100)]
win32: fix "unused variable" warning
Rafael Fontenelle [Sat, 5 Mar 2016 19:05:56 +0000 (19:05 +0000)]
Updated Brazilian Portuguese translation
Matthias Clasen [Sat, 5 Mar 2016 18:47:41 +0000 (13:47 -0500)]
testsuite: Remove a test that isn't valid with csd
With csd, the allocation of the window will be larger than the
default size, since it includes borders and decorations. Just
remove this check.
Matthias Clasen [Sat, 5 Mar 2016 18:41:47 +0000 (13:41 -0500)]
Add a test that dumps settings
This is not much of a test, tbh. But it will give us a convenient
dump of all GTK+ settings in continuous, which helps with tracking
down test failures that might depend on settings.
Matthias Clasen [Sat, 5 Mar 2016 17:15:42 +0000 (12:15 -0500)]
status icon: Prevent size dithering
We're seeing loops where the size of some status icons constantly
dithers between 24 and 25. Since I couldn't track down exactly
where the one extra pixel comes from, just stop reacting
to single-pixel size changes.
https://bugzilla.gnome.org/show_bug.cgi?id=758893
Matthias Clasen [Sat, 5 Mar 2016 17:11:05 +0000 (12:11 -0500)]
Minor cleanup
The GtkStatusIcon::size-changed signal expects a boolean return.
Matthias Clasen [Sat, 5 Mar 2016 15:39:07 +0000 (10:39 -0500)]
Update a11y test output
These tests can now assume that the csd header bar will be there.
Matthias Clasen [Sat, 5 Mar 2016 15:38:23 +0000 (10:38 -0500)]
Run accessibility tests always with CSD
This avoids differences in the output that are caused
by the presence or absence of the header bar.
Piotr Drąg [Sat, 5 Mar 2016 13:47:17 +0000 (14:47 +0100)]
Updated Polish translation
Rūdolfs Mazurs [Sat, 5 Mar 2016 10:36:09 +0000 (12:36 +0200)]
Update Latvian translation
Matthias Clasen [Sat, 5 Mar 2016 05:35:08 +0000 (00:35 -0500)]
inspector: Avoid pointless selections
The CSS selector and the class hierarchy are purely
informative, no need to select anything here.
Matthias Clasen [Sat, 5 Mar 2016 05:34:39 +0000 (00:34 -0500)]
inspector: More grid lines
I had forgotten the css property list.
Matthias Clasen [Sat, 5 Mar 2016 05:29:06 +0000 (00:29 -0500)]
inspector: Some ui refinements
Remove pointless expanders from lists, add grid lines and,
in some places, tree lines.
Matthias Clasen [Sat, 5 Mar 2016 04:51:39 +0000 (23:51 -0500)]
Update expected output of a11y tests
The window type attribute shows up here now.
The caret mode changes also affect some tests.
Matthias Clasen [Sat, 5 Mar 2016 04:50:50 +0000 (23:50 -0500)]
a11y: Expose window type hint as attribute
This was an old request, maybe still useful.
https://bugzilla.gnome.org/show_bug.cgi?id=468122
Matthias Clasen [Sat, 5 Mar 2016 03:39:31 +0000 (22:39 -0500)]
messagedialog: Respect the gtk-keynav-use-caret setting
We used to always make the labels in message dialogs selectable,
which is a bit problematic wrt. to keynav - the label can
unexpectedly 'turn blue', which irritates some people.
With the new gtk-keynav-use-caret setting, we can now only
make the labels selectable when it is required for accessibilty
reasons.
Matthias Clasen [Sat, 5 Mar 2016 03:30:51 +0000 (22:30 -0500)]
gdk: Add an X setting for gtk-keynav-use-caret
This will let us toggle the use-caret setting session-wide.
Matthias Clasen [Sat, 5 Mar 2016 03:27:02 +0000 (22:27 -0500)]
text view: Respect gtk-keynav-use-caret
When the gtk-keynav-use-caret setting is TRUE, make the cursor
visible regardless of the cursor-visible property.
Matthias Clasen [Sat, 5 Mar 2016 03:25:37 +0000 (22:25 -0500)]
Add a gtk-keynav-use-caret mode setting
For accessibility reasons, it is useful to have a global setting
that makes carets visible in all text.
https://bugzilla.gnome.org/show_bug.cgi?id=602526
https://bugzilla.gnome.org/show_bug.cgi?id=762799
Cosimo Cecchi [Sat, 5 Mar 2016 02:11:30 +0000 (18:11 -0800)]
range: avoid setting slider coordinates to negative numbers
This can happen if the theme sets a negative margin, but the coordinate
should never be negative.
Cosimo Cecchi [Sat, 5 Mar 2016 01:59:39 +0000 (17:59 -0800)]
range: factor out a function
We're going to modify this in the next commit.
Matthias Clasen [Sat, 5 Mar 2016 01:49:10 +0000 (20:49 -0500)]
gtk-demo: Fix up the inline-toolbar example
The inline toolbar in the style classes demo should use
the icons toolbar-style to ensure that the buttons are square.
Matthias Clasen [Sat, 5 Mar 2016 01:48:13 +0000 (20:48 -0500)]
tool button: Update style classes
When changing tool button contents according to the toolbar-style
property, we need to update the style classes to ensure that the
visual style matches.
https://bugzilla.gnome.org/show_bug.cgi?id=760560
Cosimo Cecchi [Sat, 5 Mar 2016 01:28:04 +0000 (17:28 -0800)]
range: fix warning for gadget slider
The slider is not HFW/WFH - just pass -1 to get rid of the warnings.
Olivier Fourdan [Fri, 4 Mar 2016 20:29:25 +0000 (21:29 +0100)]
gtkwindow: default size with fixed size windows
Allow fixed size windows with a default size to grow or shrink as the
content requires, but not smaller than the given default size.
https://bugzilla.gnome.org/show_bug.cgi?id=762974
Ray Strode [Fri, 4 Mar 2016 16:10:18 +0000 (11:10 -0500)]
gtkapplication-dbus: use application id instead of prgname for SM
gnome-session takes an "app id" for client and inhibitor registration,
This app id is supposed to be a desktop file id (complete with the
.desktop extension), but gtk+ currently uses g_get_prgname ().
This commit changes gtkapplication to use the application id instead,
which is a much more natural fit. gnome-session is going to be updated
to stop using the .desktop extension, too, so everything is consistent.
https://bugzilla.gnome.org/show_bug.cgi?id=763106
Carlos Soriano [Mon, 29 Feb 2016 22:13:51 +0000 (23:13 +0100)]
gtkplacessidebar: add new other locations with flags signal
We weren't using the open flags on the other locations signal, which
makes impossible for applications like nautilus to act in the same
way that for any other location where the user can choose between
opening in the current view, in a new window or in a new tab.
Add a new signal with an open flags parameter and deprecate
the other-location signal.
https://bugzilla.gnome.org/show_bug.cgi?id=754743
Carlos Soriano [Mon, 29 Feb 2016 21:12:03 +0000 (22:12 +0100)]
Revert "gtkplacessidebar: add open flags to other locations"
This reverts commit
63a204e1c5b6c since it broke the API.
We will add a new signal instead in an upcomming patch.
https://bugzilla.gnome.org/show_bug.cgi?id=754743
Lapo Calamandrei [Sat, 5 Mar 2016 01:09:22 +0000 (02:09 +0100)]
Adwaita: treeview column header scrollbar top border
add a top border in vertical scrollbars in treeviews.
Matthias Clasen [Sat, 5 Mar 2016 01:01:57 +0000 (20:01 -0500)]
wayland: Don't crash with offscreen windows
Use gdk_window_get_effective_toplevel when looking for
a suitable transient parent, to skip over offscreen windows
that we might encounter in the window tree. This fixes
a crash in glade.
https://bugzilla.gnome.org/show_bug.cgi?id=763110
Lapo Calamandrei [Sat, 5 Mar 2016 01:01:16 +0000 (02:01 +0100)]
Adwaita: shrink scroolbars
erroneusly pushed 15px scrollbars with the previous commit, back
to 13px now.
Lapo Calamandrei [Sat, 5 Mar 2016 00:50:18 +0000 (01:50 +0100)]
Adwaita: fix a typo
Lapo Calamandrei [Sat, 5 Mar 2016 00:39:13 +0000 (01:39 +0100)]
Adwaita: scrollbar style rewrite
works correctly with the new gadgettified GtkRange Cosimo nicelly
delivered, as a bonus steppers are styled for overlay indicator
usage as well.
Stas Solovey [Fri, 4 Mar 2016 21:41:52 +0000 (21:41 +0000)]
Updated Russian translation
Cosimo Cecchi [Fri, 4 Mar 2016 19:56:08 +0000 (11:56 -0800)]
range: add positional style classes to fill/highlight
Requested by Lapo.
Matthias Clasen [Fri, 4 Mar 2016 19:34:03 +0000 (14:34 -0500)]
x11: Trap errors around an XRandr call
Apparently, we can get X errors here. Ignore them.
https://bugzilla.gnome.org/show_bug.cgi?id=762907
Olivier Fourdan [Fri, 4 Mar 2016 09:34:12 +0000 (10:34 +0100)]
gtkwindow: windows with a fixed size can shrink
One important aspect of non-resizable windows that we need to preserve
is that they shrink when their content requires less size.
Previous changes to allow the default size to be applied to fixed size
windows would have prevented all fixed size windows from shrinking when
their content requires less size.
Allow shrinking for fixed-size windows unless a default size was
specified.
https://bugzilla.gnome.org/show_bug.cgi?id=762974
Balázs Úr [Fri, 4 Mar 2016 19:15:05 +0000 (19:15 +0000)]
Updated Hungarian translation
Lapo Calamandrei [Fri, 4 Mar 2016 19:08:19 +0000 (20:08 +0100)]
Adwaita: shrink back scale trough
like it was before the GtkRange gadgettification.
Matthias Clasen [Fri, 4 Mar 2016 18:54:39 +0000 (13:54 -0500)]
wayland: Clean up key repeat timeout
When returning G_SOURCE_REMOVE from a callback, we need
to reset the source id as well.
Matthias Clasen [Fri, 4 Mar 2016 18:55:33 +0000 (13:55 -0500)]
gadget: Add more details to warning
All the other warnings in this file include the css node name
and the gadget owners type. Be consistent about this.
Matthias Clasen [Fri, 4 Mar 2016 18:43:19 +0000 (13:43 -0500)]
gadget: Warn if for_size is too small
It does not make sense to pass a for_size into the size allocation
machinery that is smaller than the min-size in that direction.
Warn if it happens, so we can track it down when it occurs.
Lapo Calamandrei [Fri, 4 Mar 2016 18:16:48 +0000 (19:16 +0100)]
Adwaita: really fix scale with marks
added a min-height/width compensation too when in fine-tune mode,
so scale with marks and no labels doesn't make the scale node grow
when in fine-tune.
Ray Strode [Fri, 4 Mar 2016 15:02:34 +0000 (10:02 -0500)]
gtk-launch: add hack to close up dbus-daemon race
If gtk-launch exits before the program it starts fully activates,
then the dbus-daemon may avoid doing the activating method call.
This commit works around the problem by pinging the activated
application, and waiting for a reply.
This is similar to the gvfs-open bug here:
https://bugzilla.gnome.org/show_bug.cgi?id=746534
https://bugzilla.gnome.org/show_bug.cgi?id=763103
Lapo Calamandrei [Fri, 4 Mar 2016 17:36:36 +0000 (18:36 +0100)]
Adwaita: fix a thinko wrt fine-tune scales
added a scale node padding compensation for the trough growth in
fine-tune mode, so now the scale node doesn't grow anymore when
mark labels are present.
Matthias Clasen [Fri, 4 Mar 2016 13:50:52 +0000 (08:50 -0500)]
icon helper: Properly update on style change
A GtkCssChange of NULL means 'everything changed!'.
Matthias Clasen [Fri, 4 Mar 2016 13:01:02 +0000 (08:01 -0500)]
entry: Make progress node invisible when appropriate
We were failing to do that, leading to progress not disappearing
anymore after it was initially shown, in the gtk3-widget-factory
entry progress example.
Cosimo Cecchi [Fri, 4 Mar 2016 05:12:21 +0000 (21:12 -0800)]
testscale: change example to use upwards labels
There's no example of that in testscale.
Cosimo Cecchi [Fri, 4 Mar 2016 05:04:07 +0000 (21:04 -0800)]
Adwaita: fix thinko in fine-tune scale selector
Marks always get class top/bottom, regardless of the scale orientation.
Baurzhan Muftakhidinov [Fri, 4 Mar 2016 04:48:34 +0000 (04:48 +0000)]
Updated Kazakh translation
Baurzhan Muftakhidinov [Fri, 4 Mar 2016 04:09:51 +0000 (04:09 +0000)]
Updated Kazakh translation
fiddlerwoaroof [Thu, 3 Mar 2016 15:30:00 +0000 (10:30 -0500)]
Check if XRRGetOutputInfo returned a null pointer.
Fixes bug 763023: in certain circumstances, XRRGetOutputInfo will return
a null pointer. This commit adds a check to detect and handle this
return value.
Matthias Clasen [Fri, 4 Mar 2016 02:28:08 +0000 (21:28 -0500)]
expander: Reinstate rtl allocation behavior
In rtl mode, we must not just reverse the children,
but also allocate from the right. Use the newly introduced
box gadget api to achieve this.
https://bugzilla.gnome.org/show_bug.cgi?id=762945
Matthias Clasen [Fri, 4 Mar 2016 02:26:39 +0000 (21:26 -0500)]
check button: Reinstate rtl allocation behavior
In rtl mode, we must not just reverse the children,
but also allocate from the right. Use the newly introduced
box gadget api to achieve this.
https://bugzilla.gnome.org/show_bug.cgi?id=762945
Matthias Clasen [Fri, 4 Mar 2016 02:24:20 +0000 (21:24 -0500)]
box gadget: Add a way to allocate in reverse
This is needed to preserve expected allocation behavior
in rtl mode.
Piotr Drąg [Fri, 4 Mar 2016 00:52:47 +0000 (01:52 +0100)]
Updated POTFILES.skip
Matt Watson [Wed, 2 Mar 2016 02:26:47 +0000 (18:26 -0800)]
reftests: add a test for animated box shadow updating clip
This was just fixed by
3e0694284785153944255a0501e84a76c491e4b4
Matt Watson [Wed, 2 Mar 2016 07:28:23 +0000 (23:28 -0800)]
reftests: rename util file frame-inhibitor
We'll use the label-text-shadow-changes-modify-clip utils
more generically, lets name the file after its function.
Timm Bäder [Thu, 3 Mar 2016 20:04:39 +0000 (21:04 +0100)]
revealer-size: Fix animation setting
Turn animations off before testing, for all test functions containing
"no_animations".
Evangelos Skarmoutsos [Thu, 3 Mar 2016 19:02:18 +0000 (19:02 +0000)]
Updated Greek translation
Руслан Ижбулатов [Thu, 3 Mar 2016 17:22:02 +0000 (17:22 +0000)]
Remove unused variable
Руслан Ижбулатов [Thu, 3 Mar 2016 17:17:25 +0000 (17:17 +0000)]
GDK W32: Fix redrawing during drag-move with no composition
Timm Bäder [Thu, 3 Mar 2016 12:05:14 +0000 (13:05 +0100)]
revealer: Don't use TRANSITION_NONE for disabled animations
If animations are disabled, the only difference we need to make is that
we don't have to start the transition. Size requests should remain the
same.
https://bugzilla.gnome.org/show_bug.cgi?id=762996
Timm Bäder [Thu, 3 Mar 2016 12:04:58 +0000 (13:04 +0100)]
Add GtkRevealer sizing tests
https://bugzilla.gnome.org/show_bug.cgi?id=762996
Lapo Calamandrei [Thu, 3 Mar 2016 17:01:58 +0000 (18:01 +0100)]
Adwaita: fix a thinko about toolbar items spacing with margins
Olivier Fourdan [Thu, 3 Mar 2016 16:47:35 +0000 (17:47 +0100)]
gtkwindow: Fix regression with fixed size windows
Previous commit to address the default size introduced a regression
with fixed size windows if no default size was given, the resulting
window would end up much smaller than its actual content.
Changwoo Ryu [Thu, 3 Mar 2016 16:32:06 +0000 (16:32 +0000)]
Updated Korean translation
Changwoo Ryu [Thu, 3 Mar 2016 16:12:06 +0000 (16:12 +0000)]
Updated Korean translation
Lapo Calamandrei [Thu, 3 Mar 2016 15:12:20 +0000 (16:12 +0100)]
Adwaita: uniform paddings on image-buttons
fixes https://bugzilla.gnome.org/show_bug.cgi?id=762811
Lapo Calamandrei [Thu, 3 Mar 2016 14:58:13 +0000 (15:58 +0100)]
Adwaita: use margin for spacing widgets on toolbars
kind of a workaround, this approach is pretty unmaintainable, I
hope we'll get a CSS property for spacing directly on the parent
container (a toolbar in this particular case).
See https://bugzilla.gnome.org/show_bug.cgi?id=762397
Olivier Fourdan [Wed, 2 Mar 2016 10:10:29 +0000 (11:10 +0100)]
gtkwindow: Use default size even if not resizable
If a window is not resizable (with gtk_window_set_resizable ()),
the size given with gtk_window_set_default_size() is ignored.
The solution to this would be to use gtk_widget_set_size_request() but
that's a GtkWidget API and therefore does not take into account the
client side decorations when in use with GtkWindow.
Refactor the code so that gtk_window_set_default_size() (which is a
GtkWindow API) gives the expected result on non-resizable windows as
well.
bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=762974
Руслан Ижбулатов [Sun, 27 Dec 2015 03:37:21 +0000 (03:37 +0000)]
W32: Prefer CSD by default
Will make GTK+ more willing to use CSD for all normal windows without
being asked to. Lack of desktop composition will, of course, prevent
it from using CSD (in theory).
GTK_CSD=0 will force CSD to NOT to be used whenever
possible (i.e. in cases where CSD is not specifically requested
by a window, by design).
https://bugzilla.gnome.org/show_bug.cgi?id=759899
Руслан Ижбулатов [Tue, 1 Mar 2016 03:40:36 +0000 (03:40 +0000)]
GDK W32: Set default cursor from our own theme right away
Otherwise WM-dependent default cursor is used, which does not
match our theme. Worse, later GDK will realize that we have
our own left_ptr cursor and will apply it after all, making
the discrepancy even more noticeable.
https://bugzilla.gnome.org/show_bug.cgi?id=762902
Руслан Ижбулатов [Sat, 27 Feb 2016 00:10:12 +0000 (00:10 +0000)]
GDK W32: Force correct mouse cursor for custom resize/move
* Explicitly grab the device, setting appropriate cursor on it.
* Fix gdk_device_virtual_set_window_cursor() to just set the
cursor, without trying to check that mouse is over the given
window. Also prevent it from immediately resetting cursor.
* Alse take into account things that happen in other parts of
GDK - don't look for replacement cursor, GDK already did that,
and don't create a default arrow cursor instead of NULL,
GDK-W32 already did that up the stack as well.
Warn about inappropriate cursor == NULL argument instead.
https://bugzilla.gnome.org/show_bug.cgi?id=762711