0e528ec74e
notebook: No page switching in size request code
...
We have page switching under control now, we don't need to check every
get_preferred_size() call.
2016-01-11 17:05:22 +01:00
1190efebce
notebook: Set the current page when none is set
...
This case happens when you gtk_widget_show_all (notebook).
2016-01-11 17:05:22 +01:00
0045e57e96
notebook: Draw the same things, no matter the show-border value
...
show-border modifies the "frame" style class on the notebook node, but
not what we actually draw.
2016-01-11 17:05:22 +01:00
fd9ffe5d57
notebook: Turn the base notebook into a gadget
2016-01-11 17:05:22 +01:00
92089ee29e
notebook: Draw tab label while drawing tab
...
Instead of having special case code doing it.
2016-01-11 17:05:22 +01:00
3d980da405
notebook: Don't translate before painting notebook
...
This will allow us to draw children directly from within the notebook
drawing code, which will remove a bunch of special casing.
2016-01-11 17:05:22 +01:00
93dc364b02
notebook: Deprecate and stop supporting has-tab-gap style property
2016-01-11 17:05:22 +01:00
71cc212eb5
notebook: Ignore and deprecate style properties
...
Deprecate initial-gap, tab-curvature and tab-overlap properties. All
their features can be achieved using CSS.
This CSS achieves the same effect as a 10px initial gap:
notebook header {
margin-left: 10px;
margin-right: 10px;
}
A tab overlap of 10px can be achieved via:
notebook tabs {
margin-left: 10px;
}
notebook tab {
margin-left: -10px;
}
And tab curvature is essentially the same as padding or border on a tab.
2016-01-11 17:03:07 +01:00
0138af9f96
rendericon: Save and restore cairo matrix
...
... instead of requiring cairo_save()/cairo_restore().
2016-01-11 17:03:07 +01:00
4e627e108e
testnotebookdnd: Add an action widget
...
Just so we have some test that tests action widgets.
2016-01-11 17:03:07 +01:00
fa7816d99c
dnd: When setting a dnd icon widget, unset old one
...
Setting a DND icon twice should actually work and not cause warnings the
2nd time.
2016-01-11 16:56:16 +01:00
fb0a628975
Updated Japanese translation
...
Change B series paper names in accordance with the Japanese standards.
2016-01-11 23:58:16 +09:00
67c6d24bb2
Fix a small mistake in stylecontext docs
...
Fix the "will be also be" to become "will also be".
2016-01-11 12:05:27 +01:00
7a06e0102d
Updated Slovak translation
2016-01-11 09:48:05 +00:00
b035cc2cb6
Update style class docs
...
Instead of pointing to (outdated) widgets here, just refer to
widget docs in general.
2016-01-10 17:08:56 -05:00
cffe2e2c93
Improve GtkStyleContext docs
...
Refer to the foreign drawing example in gtk3-demo, refer to
the widget documentation for style properties, remove
no-longer-used treeview regions.
2016-01-10 17:05:53 -05:00
7ee65da0e3
Add foreigndrawing example to gtk3-demo
...
This makes it easier accessible and nicer to refer to
from the documentation.
2016-01-10 15:32:53 -05:00
6d239cfba7
Add a css style test for adjacent states
...
This test checks that the selector E:state + F works as expected.
https://bugzilla.gnome.org/show_bug.cgi?id=722727
2016-01-10 14:51:36 -05:00
0f120ea510
Cast the return value of g_array_free to void* before casting to char**
...
Let the compiler know that we don't want to use it as char*, so there is
no alignment error.
https://bugzilla.gnome.org/show_bug.cgi?id=760350
https://bugzilla.gnome.org/show_bug.cgi?id=755659
2016-01-11 00:23:46 +08:00
2dc8d1f611
GdkSeat: Fix annotation of gdk_seat_grab()
2016-01-10 16:49:42 +01:00
0ddc1eed19
Fix rendering of unsymmetric borders
...
There was an errant path that threw off our clipping for
the top section of the border.
https://bugzilla.gnome.org/show_bug.cgi?id=722937
2016-01-09 23:44:43 -05:00
caaf537f94
Avoid division by 0 in the border rendering code
...
We skip sides with 0 border width in render_border, but when
we collect sides with the same style, we may pass the 0 width
down to render_frame_stroke anyway. So skip width 0 sides
there as well.
2016-01-09 23:44:43 -05:00
7d0173763e
Updated Finnish translation
2016-01-09 21:39:37 +00:00
12bc8ddb93
Fix dobule border rendering
...
This is the same fix that was applied for ridge and groove
style in 9f27ee7f5a
2016-01-09 14:37:04 -05:00
152171e60c
Adwaita: transparent background for progressbar.osd
...
Hopefully fixes https://bugzilla.gnome.org/show_bug.cgi?id=760331
Added the .osd style class to the progressbar docs.
2016-01-09 20:05:41 +01:00
6da0a0a106
gtkheaderbar: add titlebutton styleclass to the separator
...
...of the window control buttons
2016-01-09 19:27:15 +01:00
99365753bf
Adwaita: shortcut window related fixes
2016-01-09 18:03:26 +01:00
44e0e1359c
Adwaita: avoid colorswatch growing when drop targets
2016-01-09 15:30:20 +01:00
239303b3f8
Adwaita: more scss cosmetics
2016-01-09 14:36:05 +01:00
2050ed04dd
Adwaita: rely on background-image for paned wide separators
...
this workaround fixes the rendering issue coming out using borders
2016-01-09 13:35:48 +01:00
4355cb45d7
Adwaita: make the paned wide separator scss code correct
...
unfortunatelly the results are not, borders behaves in a weird way
here as well, if this don't get fixed I'll rely on background here
too.
2016-01-09 13:04:27 +01:00
2b57ba61d5
Adwaita: some scss cosmetic changes
2016-01-09 12:38:49 +01:00
f09483bb98
Adwaita: fix thin paned separators
2016-01-09 12:31:41 +01:00
b143ed98b3
Adwaita: use stricter selectors on paned separators
...
this fixes the "fat" separator on gedit headerbar, since it was
cought by the paned separator styling.
2016-01-09 12:07:52 +01:00
367543c0e8
Adwaita: remove old headerbar separator styling
...
which was actually selecting nothing now.
2016-01-09 11:55:40 +01:00
4853303dd7
Adwaita: add vertical margins to headebar entries and separators
2016-01-09 11:25:08 +01:00
bf48dd9d7c
Adwaita: add vertical margins to headerbar buttons
...
to avoid them spanning the whole headerbar height with the changes
introduced in my last commit.
2016-01-09 10:19:47 +01:00
3a774abbfe
CSS documenation improvements
...
Work around some problems with the formatting of the online
docs, and split off the properties as a separate chapter.
2016-01-08 23:32:40 -05:00
d8c1e6db21
3.19.6
2016-01-08 18:29:58 -05:00
375619f9f4
Fix make dist
...
I've forgotten another place when renaming HighContrast inverse.
2016-01-08 18:29:58 -05:00
1eaf035a78
reftests: Drop deprecated style properties
2016-01-08 18:11:07 -05:00
fe35836168
Update expect output of a11y tests
2016-01-08 18:11:07 -05:00
e84f1ae3bb
Update the list of required cursors
...
We're now using standard cursor names for DND, so check for these.
2016-01-08 18:11:07 -05:00
771149e19a
wayland: Ensure we have a dnd-ask cursor
...
We use this for DND, so make sure that we fall back to some other
cursor if this one isn't present.
2016-01-08 18:11:07 -05:00
04a9b5b5f8
x11: Ensure we have a dnd-ask cursor
...
We use this for DND, so make sure that we fall back to some other
cursor if this one isn't present.
2016-01-08 18:11:07 -05:00
4b7c8218fb
Use standard cursor names for DND
...
At least as far as we have them. CSS has on equivalent of dnd-ask,
so we keep using that name as-is.
2016-01-08 18:11:07 -05:00
f2685ddf5e
colorswatch: don't apply padding twice
...
Instead of taking the border and manually removing it from the
allocation, render our background over all the border allocation box, as
that's more correct and does not take padding into account twice.
2016-01-08 14:58:25 -08:00
e62f1604f9
paned: Use GtkGestureDrag for anything else than touch
...
The GtkGesturePan behavior of locking onto certain orientations may
come across as confusing, and is not strictly necessary for mice and
other pointing devices.
As GtkGesturePan is also a GtkGestureDrag, we just use the same
callbacks on both gestures.
https://bugzilla.gnome.org/show_bug.cgi?id=759670
2016-01-08 22:03:05 +01:00
ad0fc8ae7e
wayland: Force the grab cursor while a grab is active
...
When a cursor is specified in gdk_seat_grab(), the cursor is reverted as
soon as the pointer enters or leaves another window.
To avoid this issue, store the grab cursor separately, so we force-apply
it in ::set_window_cursor(). Also, unset early the seat info from the
window on gdk_seat_ungrab(), so the next time switch_to_pointer_grab()
happens we end up picking the cursor set for the window underneath the
pointer window.
Based on a patch by Olivier Fourdan <ofourdan@redhat.com >.
https://bugzilla.gnome.org/show_bug.cgi?id=760213
2016-01-08 22:01:25 +01:00
23b629e27c
x11: Initialize GdkWindowAttr struct memory
...
Valgrind complains about jumps based on uninitialized values
otherwise.
2016-01-08 21:58:33 +01:00
