Jasper St. Pierre [Thu, 17 Jul 2014 15:46:21 +0000 (11:46 -0400)]
gtkwindow: Clean up gtk_window_move_resize a bit more
Don't shadow existing variables. Instead of sharing the allocation and
then overwriting the width/height when convenient, declare it in the
block we use it in, as, really, the three different paths are all
extremely different, and there's no sense in sharing the variable.
Jasper St. Pierre [Thu, 17 Jul 2014 15:03:27 +0000 (11:03 -0400)]
gtkwindow: Remove a bogus comment
gtk_window_configure_event certainly doesn't fill the new allocation
in. We're filling it in right here!
Jasper St. Pierre [Thu, 17 Jul 2014 17:09:48 +0000 (13:09 -0400)]
icon-browser: Improve the look a tiny bit
Add some padding and a border around the scrolled window.
Matthias Clasen [Thu, 17 Jul 2014 16:53:54 +0000 (12:53 -0400)]
GtkDialog: Be more careful in handling the headerbar
Just because an action widget has a parent thats a headerbar,
it is not guaranteed that it is 'our' headerbar. Add some
assertions to enforce this.
Matthias Clasen [Thu, 17 Jul 2014 16:49:57 +0000 (12:49 -0400)]
GtkDialog: rejigger headerbar handling
We can't avoid calling set_titlebar in buildable_add_child, since
GtkBuilder assumes that it can e.g. call grab_default at the end,
which only works if the widget is inside a toplevel at that point.
Instead, drop the titlebar and clean up priv->headerbar in constructed
if we find that we don't want a headerbar.
Matthias Clasen [Thu, 17 Jul 2014 16:11:35 +0000 (12:11 -0400)]
Add another dialog to test
This one is taken from polari, and currently doesn't work
as expected.
Sébastien Wilmet [Tue, 15 Jul 2014 17:02:54 +0000 (19:02 +0200)]
textiter: fix bug in find_visible_by_log_attrs()
find_by_log_attrs() can return true only in this case:
return moved && !gtk_text_iter_is_end (arg_iter);
So if the iter moved (i.e. something has been found), but is the end
iter, find_by_log_attrs() returns false.
Now the same checks are made in find_visible_by_log_attrs(). The public
functions using find_visible_by_log_attrs() say in their documentation
that false is returned for the end iter, hence the check with
gtk_text_iter_is_end().
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Sébastien Wilmet [Tue, 15 Jul 2014 15:07:14 +0000 (17:07 +0200)]
textiter: fix bug in FindLogAttrFunc functions
attrs[len] is the last PangoLogAttr available, at the iter position after the
last character of the line.
For a line in the middle or the start of the buffer, the '\n' is taken
into account by 'len'. For example the is_word_end is generally reached
before the '\n', not after. But for the last line in the buffer, where
there is no trailing '\n', it is important to test until attrs[len].
The bug didn't occur before because find_by_log_attrs() worked directly
on the iter passed as the function argument. But now it is no longer the
case.
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Sébastien Wilmet [Tue, 15 Jul 2014 13:24:36 +0000 (15:24 +0200)]
textiter: make the FindLogAttrFunc functions clearer
- Return true (found) and false (not found) explicitly.
- Set found_offset only when something has been found.
find_backward_cursor_pos_func() was a bit different, the while loop had
the condition "offset > 0" but the return was "offset >= 0". Probably a
micro-optimization, since offset == 0 is always a cursor position.
Anyway now the code is the same as the other functions.
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Sébastien Wilmet [Tue, 15 Jul 2014 12:51:02 +0000 (14:51 +0200)]
textiter: simplify FindLogAttrFunc
The min_offset parameter was always 0. Since there are some bugs in this
code, it'll be clearer if there are fewer parameters.
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Sébastien Wilmet [Sun, 13 Jul 2014 20:20:25 +0000 (22:20 +0200)]
textiter: fix bug in find_by_log_attrs()
Do not work with the iter passed as the function argument. Work with
another iter, and set it back to the function argument only if something
has been found.
This fixes a few unit tests. But there are regressions for a few others.
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Sébastien Wilmet [Sun, 13 Jul 2014 20:04:04 +0000 (22:04 +0200)]
tests textiter: unit tests for sentence boundaries
As the FIXME comment shows, there is a bug. It is the same bug as for
the word boundaries.
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Sébastien Wilmet [Sun, 13 Jul 2014 17:33:29 +0000 (19:33 +0200)]
tests textiter: unit tests for visible word boundaries
As the FIXME comments show, there are several problems.
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Matthias Clasen [Thu, 17 Jul 2014 10:48:36 +0000 (06:48 -0400)]
GtkPathBar: Add a style class
This will allow nautilus to reuse the style on its own
pathbar widget.
Matthias Clasen [Thu, 17 Jul 2014 04:59:34 +0000 (00:59 -0400)]
GtkIconTheme: Make spinners scale in hi-dpi
The Adwaita icon theme ships spinners in a scalable directory
with MaxSize=32 and Scale=1. One way to make them scale up in
hi-dpi would be to add an @2 directory with MaxSize=32 and Scale=2,
but that directory would also be consulted in non hi-dpi situations
and give us an effective spinner max size of 64.
Instead, treat svg icons implicitly as hi-dpi, and scale them
up to MaxSize * 2 when in hi-dpi.
Volker Sobek [Tue, 15 Jul 2014 23:09:05 +0000 (01:09 +0200)]
GtkMenu: Don't subtract shadow from requisition
This was causing menus to show up in the wrong position in case the menu
popped up towards the top and/or left.
The change to the requisition was in error; it is the allocated size
of the menu, not the toplevel, and doesn't include the shadow.
https://bugzilla.gnome.org/show_bug.cgi?id=591258
Matthias Clasen [Thu, 17 Jul 2014 01:08:27 +0000 (21:08 -0400)]
GtkStackSwitcher: Simplify setting of .needs-attention
Always add the .needs-attention style class, even if the button
is active. Themes can already avoid showing anything in this case.
Lapo Calamandrei [Wed, 16 Jul 2014 17:47:54 +0000 (19:47 +0200)]
Adwaita: increase padding on header-bar and action-bar text buttons
Lapo Calamandrei [Wed, 16 Jul 2014 17:42:17 +0000 (19:42 +0200)]
Adwaita: stack switcher needs-attention class style
Lapo Calamandrei [Wed, 16 Jul 2014 17:35:29 +0000 (19:35 +0200)]
GtkStackSwitcher: center the button child
Set button child halign to GTK_ALIGN_CENTER since I need it to style
the needs-attention class, see:
https://bugzilla.gnome.org/show_bug.cgi?id=707153
Volker Sobek [Wed, 16 Jul 2014 11:25:22 +0000 (13:25 +0200)]
Adwaita: Add back a margin of 4px to .menu
This had originally been added to Adwaita in the gnome-themes-standard
commit
7bf01517bc2 but was lost during 3.13 theme changes, causing bug
591258 to reappear.
Chun-wei Fan [Wed, 16 Jul 2014 10:04:38 +0000 (18:04 +0800)]
MSVC Introspection Builds: Filter Out gtkx.h
Like the autotools builds of the introspection files, don't include gtkx.h
in the introspection prepropcessing as it should not be included by other
GTK+ headers.
Chun-wei Fan [Wed, 16 Jul 2014 09:58:20 +0000 (17:58 +0800)]
gtk-inspector: size-groups.c: Fix build on pre-C99
Declare variables at the top of the block.
Matthias Clasen [Wed, 16 Jul 2014 03:45:27 +0000 (23:45 -0400)]
GtkListBox: Mention activatable and selectable in the docs
Matthias Clasen [Wed, 16 Jul 2014 03:53:37 +0000 (23:53 -0400)]
widget-factory: Mark selectable listbox rows non-activatable
Having them selectable is sufficient to get the hover effect.
Matthias Clasen [Wed, 16 Jul 2014 03:14:28 +0000 (23:14 -0400)]
GtkListBox: Implement non-selectable rows
This commit changes things so that we never mark non-selectable
rows as selected.
Matthias Clasen [Wed, 16 Jul 2014 03:13:43 +0000 (23:13 -0400)]
GtkListBox: Style selectable rows differently
We now apply the .button style class to rows that are either
activatable or selectable. Selectable rows only get the .button
if the listbox allows selection. This implies that we need to
update row styles when the selection mode changes, or when the
row gets added to a listbox.
Matthias Clasen [Wed, 16 Jul 2014 03:10:51 +0000 (23:10 -0400)]
GtkListBox: Add a GtkListBoxRow:selectable property
Having an explicit property for this will make it easier
to have a hover style only for rows which are activatable
or selectable.
Rows are selectable by default, to preserve compatibility.
Matthias Clasen [Wed, 16 Jul 2014 03:18:27 +0000 (23:18 -0400)]
Fix documentation typos
Matthias Clasen [Wed, 16 Jul 2014 02:42:29 +0000 (22:42 -0400)]
GtkListBox: Refactor ::activatable implementation
Store the value in the row struct, and split the style updating
out into a separate function.
Matthias Clasen [Wed, 16 Jul 2014 01:29:59 +0000 (21:29 -0400)]
Try harder to fix dialog titles
https://bugzilla.gnome.org/show_bug.cgi?id=733099
Matthias Clasen [Wed, 16 Jul 2014 01:29:20 +0000 (21:29 -0400)]
Update examples for new open-menu icon
Руслан Ижбулатов [Tue, 15 Jul 2014 16:42:39 +0000 (16:42 +0000)]
GktIconTheme: Be even more careful with paths
This is a followup for
5a252f13a863c00ccd9901afa110e85836a6af4c
https://bugzilla.gnome.org/show_bug.cgi?id=733189
Jasper St. Pierre [Tue, 15 Jul 2014 18:42:36 +0000 (14:42 -0400)]
gtkwindow: Clean up the code for an optimization
It's hard to figure out what the "expected_reply" means except under
close examination -- it's actually talking about whether this was a
reply to a ConfigureRequest or not. The inversion in the check doesn't
help either.
Make the code cleaner by moving it above the freeze/thaw case, and
making the check more explicit and without a confusing variable. If we
haven't sent any ConfigureRequests out, then it must be a gratuitous
ConfigureNotify.
Jasper St. Pierre [Tue, 15 Jul 2014 18:41:04 +0000 (14:41 -0400)]
gtkwindow: Fix the comment for an optimization
The optimization was incorrect, and so it was changed in commit
2bf6824, but the comment to go along with it was never updated.
Lapo Calamandrei [Tue, 15 Jul 2014 17:09:58 +0000 (19:09 +0200)]
Adwaita: white search bar for the file chooser
Lapo Calamandrei [Tue, 15 Jul 2014 15:58:13 +0000 (17:58 +0200)]
Adwaita: correct background color for backdrop treeview column headers
Lapo Calamandrei [Tue, 15 Jul 2014 15:52:11 +0000 (17:52 +0200)]
Adwaita: brighter depressed buttons
Benjamin Otte [Tue, 15 Jul 2014 12:35:32 +0000 (14:35 +0200)]
pathbar: Implement clipping
.. so the path bar's buttons can have shadows.
Lapo Calamandrei [Tue, 15 Jul 2014 11:28:40 +0000 (13:28 +0200)]
Adwaita: reorderable tabs inhert in backdrop
Lapo Calamandrei [Tue, 15 Jul 2014 11:01:48 +0000 (13:01 +0200)]
Adwaita: fix backdrop insensitie flat button
Lapo Calamandrei [Tue, 15 Jul 2014 08:24:08 +0000 (10:24 +0200)]
Adwaita: selected list rows inhert in backdrop
Lapo Calamandrei [Tue, 15 Jul 2014 07:11:56 +0000 (09:11 +0200)]
Adwaita: darker hover state again for spinbuttons button in the dark variant.
And for hovered list rows, since it works better and it's less
distracting.
Jasper St. Pierre [Mon, 14 Jul 2014 22:58:27 +0000 (18:58 -0400)]
window: Use CAIRO_OPERATOR_SOURCE to clear to background
Some windows, like GtkWindow and some other apps, set a transparent
background. The guarantee for begin_paint_region is that there should
be a full clear to the specified background color, not a composite
against what was there before.
This fixes repaint artifacts in Wayland and Weston in a better way than
76922c169f0699d9b85ef3eff2170f2cede1782c.
Jasper St. Pierre [Mon, 14 Jul 2014 23:12:19 +0000 (19:12 -0400)]
Revert "wayland: Prevent stale paints and weird artifacts when using Weston"
This reverts commit
76922c169f0699d9b85ef3eff2170f2cede1782c.
This is too local a fix, and is broken if the user paints to a small
region, as the entire buffer will be wiped.
Matthias Clasen [Mon, 14 Jul 2014 22:44:00 +0000 (18:44 -0400)]
GktIconTheme: Be more careful with paths
Slapping file:// in front of a path does not guarantee a working
uri (e.g. if you are on windows and the path looks like F:\\...).
Therefore, go back to using g_file_new_for_path if we don't have
to deal with a resource.
Matthias Clasen [Mon, 14 Jul 2014 22:26:33 +0000 (18:26 -0400)]
Bump version
Matthias Clasen [Mon, 14 Jul 2014 18:11:25 +0000 (14:11 -0400)]
3.13.4
Matthias Clasen [Mon, 14 Jul 2014 22:21:50 +0000 (18:21 -0400)]
Force the simple im module for reftests
I've seen reftests fail with IBus warnings. We should avoid that.
Matthias Clasen [Mon, 14 Jul 2014 18:36:41 +0000 (14:36 -0400)]
Update expected output for a11y test
Lapo Calamandrei [Mon, 14 Jul 2014 16:42:37 +0000 (18:42 +0200)]
Adwaita: brighter spinbutton button active state for the dark variant.
Lapo Calamandrei [Mon, 14 Jul 2014 16:37:31 +0000 (18:37 +0200)]
Adwaita: bright spinbutton button hover effect in the dark variant.
In line with what we do for hovered list rows.
Lapo Calamandrei [Mon, 14 Jul 2014 16:29:08 +0000 (18:29 +0200)]
Adwaita: list-row style
Matthias Clasen [Mon, 14 Jul 2014 13:17:47 +0000 (09:17 -0400)]
widget-factory: Fix multi-selection example
It turns out that activate-on-single-click interferes with the
expected multi-selection behavior. Turn it off, since there's nothing
to activate in this list, anyway.
Matthias Clasen [Mon, 14 Jul 2014 12:49:54 +0000 (08:49 -0400)]
widget-factory: Use multi-selection for the listbox example
Matthias Clasen [Mon, 14 Jul 2014 12:48:33 +0000 (08:48 -0400)]
listbox: Update docs
Multiple selection is supported now.
Matthias Clasen [Mon, 14 Jul 2014 03:33:26 +0000 (23:33 -0400)]
Don't use guint32 with XChangeProperty
As pointed out in https://bugzilla.gnome.org/show_bug.cgi?id=710449,
XChangeProperty expects data to be a long array for format 32.
Matthias Clasen [Mon, 14 Jul 2014 02:00:59 +0000 (22:00 -0400)]
GtkTreeView: Avoid interference with animation
The animated scrolling interferes with incremental validation.
As short-term solution, disable scrolling animation during
incremental validation. This is not a proper solution, but
it avoids broken behavior like scrollbars that are not reacting
to clicks. The problem was visible, e.g. in the list view
example in gtk3-demo.
Matthias Clasen [Mon, 14 Jul 2014 03:04:08 +0000 (23:04 -0400)]
Fix a typo
A getter shouldn't return void.
Matthias Clasen [Mon, 14 Jul 2014 02:54:19 +0000 (22:54 -0400)]
Revert "GtkTreeView: Avoid interference with animation"
This reverts commit
faa676ffa73060538a4a456531ad1810f3ff7b95.
It fixed the problematic list view example in gtk3-demo, but
broke all others :-(
Matthias Clasen [Mon, 14 Jul 2014 02:00:59 +0000 (22:00 -0400)]
GtkTreeView: Avoid interference with animation
The animated scrolling interferes with incremental validation.
As short-term solution, disable scrolling animation during
incremental validation. This is not a proper solution, but
it avoids broken behavior like scrollbars that are not reacting
to clicks. The problem was visible, e.g. in the list view
example in gtk3-demo.
Matthias Clasen [Mon, 14 Jul 2014 02:00:30 +0000 (22:00 -0400)]
Add a getter for the adjustment animation duration
Matthias Clasen [Sun, 13 Jul 2014 20:08:44 +0000 (16:08 -0400)]
inspector: Make values in general tab selectable
Matthias Clasen [Sun, 13 Jul 2014 20:03:32 +0000 (16:03 -0400)]
inspector: Make listbox rows non-activatable
Matthias Clasen [Sun, 13 Jul 2014 19:51:03 +0000 (15:51 -0400)]
listbox: Avoid a warning
Don't try to set an adjustment value if we don't have an adjustment.
Owen W. Taylor [Fri, 11 Jul 2014 20:42:38 +0000 (16:42 -0400)]
Handle resolution changes in the GDK backend code
gdk_x11_display_set_window_scale() affects the interpretation of the
Xft/DPI XSETTING - it is substituted inside GDK with the value of
Gdk/UnscaledDPI xsetting. However, this change is not propagated to
GTK+ and from GTK+ back to gdk_screen_set_resolution() until the
main loop is run.
Fix this by handling the screen resolution directly in gdk/x11.
This requires duplication of code between GDK and GTK+ since we still
have to handle DPI in GTK+ in the case that GdkSettings:gtk-xft-dpi
is set by the application.
https://bugzilla.gnome.org/show_bug.cgi?id=733076
Stefano Facchini [Sat, 12 Jul 2014 12:59:59 +0000 (14:59 +0200)]
inspector: Do not accidentally change tab in the object notebook
The "hide-and-show" dance makes the notebook select the next tab.
https://bugzilla.gnome.org/show_bug.cgi?id=733092
Matthias Clasen [Sun, 13 Jul 2014 18:38:43 +0000 (14:38 -0400)]
Adwaita: Work towards working touch selection popups
This fixes some of the confusion around toolbars inside .osd.
The look is not perfect, but less broken than before.
Matthias Clasen [Sun, 13 Jul 2014 17:57:23 +0000 (13:57 -0400)]
GtkTextHandle: Better draw() implementation
Conceptually, text handles are boxes, whose content is a 'handle',
so draw background, frame and handle. With this, and the previous
commit, the cursor-handle theming in Adwaita now works as intended.
Matthias Clasen [Sun, 13 Jul 2014 17:56:14 +0000 (13:56 -0400)]
Allow to use -gtk-icon-source for handles
This will let us use icons for touch selection handles, once
they use gtk_render_handle.
Matthias Clasen [Sun, 13 Jul 2014 17:34:59 +0000 (13:34 -0400)]
Adwaita: Try to fix up selection touch handles
The text handles reuse the style context from their parent widget,
and just add .cursor-handle. That means that the more specific
entry selectors override the .cursor-handle theming unless we
select for .entry.cursor-handle. We also need to work harder to
keep the .entry styling from drawing decorations around the outlines.
Matthias Clasen [Sun, 13 Jul 2014 16:31:28 +0000 (12:31 -0400)]
listbox: Scroll headers on screen
The listbox code relies on the container focus adjustment handling
to scroll the cursor row on screen. But GtkContainer has no idea
about row headers, so ensure that we scroll the header on screen too.
Matthias Clasen [Sun, 13 Jul 2014 16:31:14 +0000 (12:31 -0400)]
widget-factory: Fix a typo
Sébastien Wilmet [Tue, 15 Apr 2014 20:11:29 +0000 (22:11 +0200)]
tests textiter: unit tests for cursor positions
See the FIXME comment.
https://bugzilla.gnome.org/show_bug.cgi?id=618852
Sébastien Wilmet [Fri, 11 Apr 2014 11:41:50 +0000 (13:41 +0200)]
textiter: small optimization for find_by_log_attrs()
Use gtk_text_iter_set_line_offset (&tmp_iter, 0) instead of
gtk_text_iter_get_line(). The difference should not be big. In the first
case the line doesn't need to be traversed thanks to the offset 0. For
get_line(), the btree must be traversed.
A temporary iter is needed to not break the behavior. But the behavior
is quite strange, the function works directly on the iter passed as an
argument to the function, even if the function returns FALSE (not
found). So maybe a later commit will fix this strange behavior.
https://bugzilla.gnome.org/show_bug.cgi?id=629129
Sébastien Wilmet [Thu, 10 Apr 2014 21:53:41 +0000 (23:53 +0200)]
textiter: remove recursivity of find_by_log_attrs()
find_by_log_attrs() was a recursive function. It is replaced by an
iteration.
The already_moved_initially parameter was TRUE only for the recursive
call, so the paramater is removed.
There is also a small cleanup of the find_visible_by_log_attrs()
(remove trailing spaces, fix indentation).
There is still a part to optimize for a later commit.
https://bugzilla.gnome.org/show_bug.cgi?id=629129
Matthias Clasen [Sun, 13 Jul 2014 13:56:06 +0000 (09:56 -0400)]
GtkDialog: Set window titles in header bars
This is fallout from the recent dialog changes. We don't set the
headerbar as titlebar early enough anymore, so when the window title
gets set, it does not get passed on to the headerbar. So, re-set it
manually when the titlebar is put in place.
https://bugzilla.gnome.org/show_bug.cgi?id=733099
Paolo Borelli [Sun, 13 Jul 2014 13:45:21 +0000 (15:45 +0200)]
list-box: fix typo in natural size computation
Matthias Clasen [Sun, 13 Jul 2014 13:01:11 +0000 (09:01 -0400)]
Adwaita: Some theming for activatable list rows
Matthias Clasen [Sun, 13 Jul 2014 01:04:54 +0000 (21:04 -0400)]
widget-factory: Use the new ::activatable property
Matthias Clasen [Sun, 13 Jul 2014 00:37:41 +0000 (20:37 -0400)]
GtkListBox: Add an activatable property for rows
This will let us theme activatable rows differently.
We also avoid emitting the ::row-activated signal for
rows that are not activatable. For compatibility reasons,
rows are activatable by default.
https://bugzilla.gnome.org/show_bug.cgi?id=733112
Matthias Clasen [Sat, 12 Jul 2014 23:48:06 +0000 (19:48 -0400)]
widget-factory: Improve listbox example spacing
Matthias Clasen [Sat, 12 Jul 2014 21:30:25 +0000 (17:30 -0400)]
widget-factory: Fix the row activation example
Turns out ::activate on listbox rows is only for key bindings,
so use ::row-activated.
Matthias Clasen [Sat, 12 Jul 2014 18:46:00 +0000 (14:46 -0400)]
Trivial typo fix
Matthias Clasen [Sat, 12 Jul 2014 18:45:16 +0000 (14:45 -0400)]
widget-factory: Demonstrate listbox headers
Add a less trivial examples of headers in the colors listbox.
Matthias Clasen [Sat, 12 Jul 2014 16:54:21 +0000 (12:54 -0400)]
widget-factory: Add some row activation to the listbox example
This shows some fallout from the gesture conversion.
Piotr Drąg [Sat, 12 Jul 2014 13:15:44 +0000 (15:15 +0200)]
Updated POTFILES.skip
Stefano Facchini [Sat, 12 Jul 2014 09:35:42 +0000 (11:35 +0200)]
widget-factory: Do no crash on variadic arguments
Matthias Clasen [Sat, 12 Jul 2014 07:31:29 +0000 (03:31 -0400)]
widget-factory: Add another listbox example
This one is long enough to scroll, and has selection.
Matthias Clasen [Sat, 12 Jul 2014 05:56:55 +0000 (01:56 -0400)]
Fix the build
gtkx.h can no longer be included internally, so skip it when
generating the gir. We include the individual headers anyway.
Matthias Clasen [Sat, 12 Jul 2014 05:43:03 +0000 (01:43 -0400)]
inspector: Reshuffle Visual and General tabs
Move the touchscreen switch to the other debug switches, and
move the hidpi spin to the other graphical controls. Since the
Visual tab is getting large, make it scroll. The General tab
is purely informational again.
Benjamin Otte [Sat, 12 Jul 2014 05:16:33 +0000 (07:16 +0200)]
GtkStack: Render a background
https://bugzilla.gnome.org/show_bug.cgi?id=733027
Benjamin Otte [Sat, 12 Jul 2014 05:12:49 +0000 (07:12 +0200)]
gtk-demo: This is not a unique app
When running gtk-demo a 2nd time, it's usually to compare its
output with a different version of gtk-demo.
Benjamin Otte [Sat, 12 Jul 2014 05:11:30 +0000 (07:11 +0200)]
widget-factory: This is not a unique app
When running widget-factory a 2nd time, it's usually to compare its
output with a different version of widget-factory.
Benjamin Otte [Sat, 12 Jul 2014 03:33:33 +0000 (05:33 +0200)]
reftests: Add test for latest fix
Ensure we draw the full backgorund on labels, not just where test
appears.
Benjamin Otte [Sat, 12 Jul 2014 03:32:12 +0000 (05:32 +0200)]
label: Include all of the allocation in the clip
Otherwise, the CSS background we draw would be clipped if the ink rect
was smaller than the allocation (a very common thing).
Broken since
37030a771059f43ad701700380719d19cbc7c966 where we clipped
to the ink rect.
Benjamin Otte [Sat, 12 Jul 2014 03:29:54 +0000 (05:29 +0200)]
gtkx: Warn if we include gtkx.h during GTK compilation
Benjamin Otte [Sat, 12 Jul 2014 03:27:49 +0000 (05:27 +0200)]
a11y: Do not include gtkx.h directly
Benjamin Otte [Sat, 12 Jul 2014 02:46:20 +0000 (04:46 +0200)]
socket: Don't include gtk.h
And fix remaining headers to not emit warnings
Matthias Clasen [Sat, 12 Jul 2014 04:33:19 +0000 (00:33 -0400)]
Revert "GtkStack: Render background and chain up gtk_widget_draw ()."
This reverts commit
f6695b962ff6462cfc8250f949505a17d0d9df6d.