Commit Graph

27015 Commits

Author SHA1 Message Date
Christian Hergert
7ecd337877 entry: handle NULL cursor when releasing reference
If we happen to have gotten NULL back for a cursor, then we
need to check before unreffing it.
2016-09-27 14:09:02 -07:00
Christian Hergert
3c775eb2b8 treeviewcolumn: use g_clear_object when freeing cursor
There is the chance that the cursor returned NULL on systems
with missing cursors. This defensively handles that by checking
for a NULL cursor.
2016-09-27 12:41:55 -07:00
Matthias Clasen
8a7d0ab481 Make gtk_init_with_args behave as documented
This docs say it should return FALSE if the display can't be opened,
but it was failing.

https://bugzilla.gnome.org/show_bug.cgi?id=771959
2016-09-27 15:12:47 -04:00
Matthias Clasen
6181ff2552 paned: quietly handle missing cursors
The row-/col-resize cursors should really be present, but if they
aren't we don't need to crash.

https://bugzilla.gnome.org/show_bug.cgi?id=771963
2016-09-27 14:55:13 -04:00
Gautier Pelloux-Prayer
fea851b3eb Fix wrong URL link in gtkshortcutswindow documentation
https://bugzilla.gnome.org/show_bug.cgi?id=772057
2016-09-27 14:48:38 -04:00
Daniel Drake
fa23641992 _gtk_icon_helper_draw: get style earlier
After checking for rendered_surface, the call to gtk_css_node_get_style
can invalidate the style and result in rendered_surface being set to
NULL. This was result in some icon views appearing blank on
Endless OS on armv7hl, and this error:

Gtk-CRITICAL **: gtk_css_style_render_icon_surface: assertion 'surface != NULL' failed

Call gtk_css_node_get_style earlier to ensure we always pass a valid
surface to gtk_css_style_render_icon_surface.

https://bugzilla.gnome.org/show_bug.cgi?id=765649
https://phabricator.endlessm.com/T13524
2016-09-26 14:23:51 -07:00
Lapo Calamandrei
ebc852249a Adwaita: always use images for button backgrounds
transitions gets messed up when some background-image and
background-color are in the mix.
Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=769500
2016-09-24 16:11:39 +02:00
Philip Withnall
506426a30e gtktreeviewcolumn: Fix typo in typedef documentation 2016-09-23 21:10:13 -07:00
Carlos Soriano
cc814fdeac gtkplacesview: mark string as translatable
We missed it.

https://bugzilla.gnome.org/show_bug.cgi?id=771666
2016-09-22 17:02:19 +02:00
Rico Tzschichholz
a3f1596069 Add some missing (nullable) annotations
https://bugzilla.gnome.org/show_bug.cgi?id=771826
2016-09-22 14:02:07 +02:00
Timm Bäder
54fdcb3ffa widget: Don't print missing allocation warning by default
This was meant to be silenced unless expicitly requested but
G_ENABLE_DEBUG is defined by default unless --disable-debug is passed to
configure, so use G_ENABLE_CONSISTENCY_CHECKS instead which is only
defined if --enable-debug is explicitly passed.
2016-09-22 09:01:06 +02:00
Lapo Calamandrei
06b2adb107 encodesymbolic: handle circle objects
add circle objects to the injected style for recoloring.
Should avoid randomly colored symbolic icon bits when circles are
in the mix as in network-vpn-acquiring-symbolic for example.
2016-09-21 17:46:19 +02:00
Emmanuele Bassi
b74cd23e29 docs: Add a diagram for GtkListBox nodes
GtkListBox's API reference for CSS nodes should follow the template of
other widget classes.
2016-09-20 23:32:00 +01:00
Jakub Steiner
d8776ae756 Adwaita: sync svg assets with css
- scale slider strokes were lighter on the SVG variants
2016-09-19 22:21:23 +02:00
Olivier Fourdan
4cb1b9645e gtkwindow: Update shadow size on state change
Otherwise, with CSD, we could have a discrepancy where gtk uses the
right values for the shadows whereas the gdk backend still uses the old
values, leading in some cases to invalid or negative min size being
computed (which, in Wayland, leads a protocol error).

https://bugzilla.gnome.org/show_bug.cgi?id=771561
2016-09-19 19:35:50 +02:00
Emmanuele Bassi
08e443e0db docs: Update gtk_window_get_size()
The main corpus of the documentation for gtk_window_get_size() is still
full of X11-isms, so we should port it to something that is more
backend-agnostic. Additionally, having some examples would be nice for
application authors looking at a way to appropriately use this function.
2016-09-17 12:47:39 +01:00
Emmanuele Bassi
bb705837bc Ensure drawing context is set
If somebody decides to use gtk_widget_set_double_buffered() in the
middle of a draw() then there's the risk of calling end_draw_frame()
with an invalid pointer.

Some overeager compilers may warn about the double_buffered bit field
changing values and leading to a potentially uninitialized variable.

In order to avoid compiler warnings or crashes, we can simply store the
value of the double_buffered bit field at the beginning of the rendering
and use that instead of the actual bit field.

https://bugzilla.gnome.org/show_bug.cgi?id=771463
2016-09-15 10:17:24 +01:00
Timm Bäder
5ea69a136b widget: Only warn about missing allocation if G_ENABLE_DEBUG
Not all occurrences of this warning can be fixed today, so put it behind
a G_ENABLE_DEBUG flag since it still shows legitimate problems even if
some of them are false positives.
2016-09-14 18:06:54 -04:00
Matthias Clasen
29faa2db44 Redo focus handling in treeview once more
The fix for bug 767468 had some unintended side-effects. This is
an attempt at doing the same fix (don't grab focus when we are
grab-shadowed), while avoiding the breakage, by using GTK+'s
internal tracking for grab-shadowed-ness.

https://bugzilla.gnome.org/show_bug.cgi?id=770508
2016-09-11 11:47:55 -04:00
Matthias Clasen
3f102c27aa gtk-launch: Add help string for --version
This was missing for no good reason.
2016-09-11 10:51:19 -04:00
Christian Kellner
d4bf215611 placesview: keep reference during network fetching
Analogous to (un)mount operation, we now keep a reference around
during the ongoing operation and make use of the destroyed flag
to check if we are still alive or if we have been cancelled as
a result of the widget being destroyed.

https://bugzilla.gnome.org/show_bug.cgi?id=764979
2016-09-09 16:55:47 -04:00
Christian Kellner
7cdc04c8d2 placesview: override destory to cancel ongoing ops
Since we hold on to a reference during (un)mount operations, we
don't trigger the cancellation of operations in finalize anymore.
Instead we now override the GtkWidget's destroy() and cancel any
ongoing operations there.

https://bugzilla.gnome.org/show_bug.cgi?id=764979
2016-09-09 16:55:47 -04:00
Christian Kellner
0e5b0b75c5 placesview: properly recover from cancellation
The current code wrongly assumes that cancellation can only happen
as a result widget finalization, and consequentially does not
properly recover from it. Therefore if the operation is cancelled
as a result of user interaction, the entry is will stay disabled
and the spinner will keep spinning. This is fixed by removal of
the early bail out in case of cancellation.

https://bugzilla.gnome.org/show_bug.cgi?id=764979
2016-09-09 16:55:47 -04:00
Christian Kellner
3ed25941ac placesview: hold onto ref during (un)mount ops
During mount and unmount opertions we keep a reference to the
GtkPlacesView around, so we have a valid view for the callback
code, even in the case that othe external references have been
dropped (i.e. the containing window gets destroyed).

https://bugzilla.gnome.org/show_bug.cgi?id=764979
2016-09-09 16:55:47 -04:00
Tristan Van Berkom
c39ad84970 Revert "about dialog: Set a max-content-height on credits"
This reverts commit 6af5033386.

Scrolled window behavior of propagating child natural sizes
has now been made optional, so there is no need to work around
this by setting a hard coded maximum content height anymore.
2016-09-06 21:29:03 +09:00
Matthias Clasen
e51fd7cb18 css: Complete the cross-fade implementation
We need to actually compute the images. Otherwise,
bad things happen when we try to draw.

https://bugzilla.gnome.org/show_bug.cgi?id=767391
2016-09-06 06:38:50 -04:00
Matthias Clasen
c2ab49e963 Fix whitespace 2016-09-06 06:38:50 -04:00
Matthias Clasen
c30ee622fe css: Set meaningful fallback color
This code was passing out-of-range values into
gdk_cairo_set_source_rgb, which cannot really make
anything better.
2016-09-06 06:38:50 -04:00
Tristan Van Berkom
eaa6ca4a49 GtkScrolledWindow: Bug 766569 - fix return types in added APIs
In gtkscrolledwindow.c, the return type of _get_propagate_natural_width()
and _get_propagate_natural_height() were accidentally gint instead of
gboolean, fixed to match the type correctly declared in the header file.
2016-09-05 16:08:27 +09:00
Matthias Clasen
953b82a584 label: take padding into account when measuring
When doing height-for-width measurements, we need to take
padding into account. This was causing various allocation
errors.

https://bugzilla.gnome.org/show_bug.cgi?id=770849
2016-09-04 15:41:11 -04:00
Matthias Clasen
61e8c69460 label: Fix indentation
Trivial, but annoying.
2016-09-04 15:41:11 -04:00
Matthias Clasen
5b21c7d79f range: Ensure we don't underallocate the highlight gadget
This was causing warnings in HighContrast.

https://bugzilla.gnome.org/show_bug.cgi?id=770614
2016-09-04 09:43:51 -04:00
Matthias Clasen
9ddd40f61e Fix a variable name misspelling
No semantic change here; the misspelling just annoyed me.
2016-09-04 08:21:49 -04:00
Joaquim Rocha
f9658a2faf listbox: Select the last row on page down when there are few rows
When pressing page down doesn't retrieve a valid row (because the list
box has few rows), the last visible one should be selected instead.

https://bugzilla.gnome.org/show_bug.cgi?id=770703
2016-09-03 13:52:39 +02:00
Joaquim Rocha
adabec7d25 listbox: Avoid crashing on page down if the list has few rows
The code always assumed that getting a row at a certain 'y' was
possible but if the list box has more empty space than rows then a
valid row may not be retrieved.

https://bugzilla.gnome.org/show_bug.cgi?id=770703
2016-09-03 13:52:39 +02:00
Timm Bäder
857997be8c popover: Skip first frame when transitioning
Similar to what we do in GtkStack.
2016-09-03 12:24:58 +02:00
Jakub Steiner
63c18bba51 Adwaita,HC: fix icon name
- while we don't use steppers anymore, for some reason they are still
  defined in the theme and if you sacrifice a chicken and jump on one
  leg at full moon, you can enable them.

https://bugzilla.gnome.org/show_bug.cgi?id=769498
2016-09-02 19:11:29 -04:00
Timm Bäder
ce1ae60da2 popover: Propagate drawing 2016-09-01 15:20:12 +02:00
Timm Bäder
2c04941c5c popover: Fix hfw/wfh size request
Since we're using _get_rect_coords in size-allocate when allocating the
size of the child widget, use the newly introduced _get_rect_for_size to
calculate the difference between the passed size_for and the one we're
supposed to pass on to the child widget.
2016-09-01 15:16:58 +02:00
Timm Bäder
7f1ab32e60 popover: Introduce _get_rect helper for a size
When calculating the requested size of a popover, we need to do the
exact same same thing _get_rect_coords did, but not for the
current popopver allocation. Add _get_rect_for_size that can be used for
this purpose
2016-09-01 15:15:18 +02:00
Timm Bäder
4f7ec1f1c0 gtk{info,search}bar.ui: Remove some leftover attributes 2016-09-01 12:32:51 +02:00
Arnaud Bonatti
533de822ba Use 'focused' everywhere.
https://bugzilla.gnome.org/show_bug.cgi?id=770624
2016-09-01 07:14:30 +02:00
Arnaud Bonatti
3c3a11591b Fix typos.
https://bugzilla.gnome.org/show_bug.cgi?id=770624
2016-09-01 07:14:30 +02:00
Tristan Van Berkom
0984d1622d GtkScrolledWindow: Bug 766569 - Make propagation of natural child sizes optional
Making propagation of child natural sizes mandatory (or default, even) was
evidently a mistake as this causes dynamic content in a scrolled window
to resize it's parent when the scrolled window is competing for space
with an adjacent widget.

This patch instead adds API to control whether natural width and
height of the child should be propagated through the scrolled windows
size requests.
2016-08-31 14:44:50 -04:00
Lionel Landwerlin
ea1295c9ab gtk: window: don't forget to realize popovers of subwindows
https://bugzilla.gnome.org/show_bug.cgi?id=770307
2016-08-29 14:01:10 -04:00
Jonas Ådahl
f98c9fec6c window: Add Wayland implementation of gtk_window_export_handle
https://bugzilla.gnome.org/show_bug.cgi?id=769788
2016-08-29 13:20:44 -04:00
Benjamin Otte
e1a03ead7a Use NULL for generic marshallers in g_signal_new()
glib will use the correct marshaller automatically. And as a side
effect, we also get all glib optimizations, like a va marshaller.
2016-08-29 16:20:54 +02:00
Matthias Clasen
6af5033386 about dialog: Set a max-content-height on credits
This keeps the credits section from making the dialog grow
when there are lots of credits.

https://bugzilla.gnome.org/show_bug.cgi?id=770458
2016-08-29 10:01:20 -04:00
Carlos Soriano
cfe4c9e0f4 gtkplacessidebar: fix signal marshal
We were using __VOID for the SHOW_OTHER_LOCATION signal that
uses flags named SHOR_OTHER_LOCATION_WITH_FLAGS.
However, if a signal uses flags the marshal needs to use __FLAGS.

This patch addresses this using VOID__FLAGS as the marshaler parameter.

Thanks to Jan Steffens for pointing this out.

https://bugzilla.gnome.org/show_bug.cgi?id=770550
2016-08-29 15:01:12 +02:00
Benjamin Otte
35c96c2c1f Add braces to appease gcc
-ftrack-macro-expansion=0 doesn't like if statements without braces when
evaluating indentation levels.
2016-08-25 23:02:27 +02:00
Benjamin Otte
c8b301f1ea uimanager: Correctly indent code 2016-08-25 23:02:27 +02:00
Benjamin Otte
da8cef16dc a11y: Correctly indent code 2016-08-25 23:02:27 +02:00
Matthias Clasen
f53706b003 cell renderer text: Allow entries to shrink
Set a small max-width on entries used for editing cells, so they
adapt to small columns and don't overlap the next column.

https://bugzilla.gnome.org/show_bug.cgi?id=770374
2016-08-25 07:53:32 -04:00
Matthias Clasen
f66a76d998 Document window-icon related api as 'may not work'
Some platforms simply don't have window icons (such as Wayland).
2016-08-24 14:43:58 -04:00
Matthias Clasen
ae636ba0c2 notebook: Update arrow state when switching pages
Otherwise, we may end up showing clickable arrows that don't
do anything.

https://bugzilla.gnome.org/show_bug.cgi?id=770332
2016-08-24 14:10:35 -04:00
Matthias Clasen
07e9b399d5 scale: Document interaction between rounding and autoscrolling
This is a bit of a gotcha, so better document it.
2016-08-24 12:58:54 -04:00
Matthias Clasen
6163b3b2b5 Allow activating the desktop portal for inhibit
We allow it everywhere else, and there is nothing wrong with it.
2016-08-24 10:55:44 -04:00
Carlos Garnacho
e9771d642f GtkPadController: Make ring/strip actions be activated with the axis value
This way these axes may be used in detail by the implementors of pad GActions.

https://bugzilla.gnome.org/show_bug.cgi?id=770026
2016-08-23 21:01:45 +02:00
Carlos Garnacho
b24a865d9b GtkEventController: Add some comment describing the filter_event private vfunc
It might not be entirely clear what the boolean return value means.

https://bugzilla.gnome.org/show_bug.cgi?id=770026
2016-08-23 21:01:45 +02:00
Carlos Garnacho
e66a821a74 GtkPadController: Notify actions back to the windowing on wayland
This allows applications to provide descriptions of the actions performed
by each pad feature. pass the GtkPadActionEntry labels for this.

https://bugzilla.gnome.org/show_bug.cgi?id=770026
2016-08-23 21:01:45 +02:00
Carlos Garnacho
a0b9586465 gtk: Add GtkPadController
This GdkEventController is a helper object to handle pad events,
it allows setting a mapping to action names, to be triggered in
the given action group.

In order to help on places where advanced mapping/configurability
of pad features is not desirable, this controller also allows
passing a NULL pad device, meaning it will listen on all pads,
and/or passing -1 on mode/index, so an action applies to all
modes/features (eg. strips/rings).

https://bugzilla.gnome.org/show_bug.cgi?id=770026
2016-08-23 21:01:45 +02:00
Carlos Garnacho
3a6d0ff2ab gtk: Add minimal handling of pad events
No real handling is yet performed, to be done through a GdkEventController

https://bugzilla.gnome.org/show_bug.cgi?id=770026
2016-08-23 21:01:44 +02:00
Philip Withnall
d3c204c774 gtkbindings: Add an example for gtk_binding_entry_add_signal()
Otherwise the way the varargs are supposed to be used is completely
opaque.

https://bugzilla.gnome.org/show_bug.cgi?id=770236
2016-08-23 11:51:13 +01:00
Philip Withnall
92a95c7de7 gtkbindings: Clarify that widgets need has-focus for bindings to work
https://bugzilla.gnome.org/show_bug.cgi?id=770242
2016-08-22 18:06:39 +01:00
Lapo Calamandrei
a6409458f0 Adwaita: increase border radius on button.circular...
...to allow huge circular buttons.
See https://bugzilla.gnome.org/show_bug.cgi?id=770166
2016-08-22 14:17:34 +02:00
Timm Bäder
5c696a7ee3 popover: Clarify transitions-enabled deprecation
https://bugzilla.gnome.org/show_bug.cgi?id=769706
2016-08-20 20:54:20 +02:00
Timm Bäder
73b949173e stylepropertyimpl: Remove double assignment 2016-08-20 09:54:44 +02:00
Timm Bäder
a985e62b25 Use gtk_popover_popdown/popup where appropriate
https://bugzilla.gnome.org/show_bug.cgi?id=769706
2016-08-16 11:49:26 -04:00
Timm Bäder
a6b9b3648d GtkPopover: Deprecate transitions-enabled
The effect of transitions-enabled=true can now be
achieved using gtk_popover_popup/popdown and the effect
of transitions-enabled=false can be achieved using
gtk_widget_show/hide.

https://bugzilla.gnome.org/show_bug.cgi?id=769706
2016-08-16 11:49:26 -04:00
Timm Bäder
1f7b6c1d6f GtkPopover: Add gtk_popover_popdown/popup
Since not chaining up in gtk_widget_show/gtk_widget_hide is not allowed,
we can't just implicitly delay the hiding in GtkPopover's hide
implementation. Fix this by introducing gtk_popover_popup() and
gtk_popover_popdown() to show or hide a popover with transition and
revert GtkPopover's show/hide implementation to apply their effect
without the transition.

https://bugzilla.gnome.org/show_bug.cgi?id=769706
2016-08-16 11:49:26 -04:00
Matthias Clasen
4e418ed71d file chooser: align date formatting with nautilus
We were treating 'yesterday' not quite the same way.

https://bugzilla.gnome.org/show_bug.cgi?id=769568
2016-08-11 12:22:01 -04:00
Timm Bäder
0d17421ffd popover: Make some public API NULL-safe
Instad of potentially dereferencing a NULL-pointer, use the NULL-safe
gtk_popover_get_instance_private before the g_return_if_fail
precondition check.
2016-08-10 16:37:17 +02:00
Timm Bäder
57ef631216 label: Remove some useless casts
These were casts of a const GdkEvent* to GdkEvent* even though the
function they are passed to takes a const GdkEvent*.
2016-08-10 16:37:17 +02:00
Ignacio Casal Quinteiro
b0f793c8e7 Use g_snprintf instead of snprintf.
snprintf is not compatible with msvc.
2016-08-10 12:45:39 +02:00
Matthias Clasen
5fefc7a714 headerbar: Fix size allocation
The GtkHeaderBar gadget implementation was subtly broken: it called
gtk_widget_set_allocation both in gtk_header_bar_size_allocate (with
the actual allocation) and in gtk_header_bar_allocate_contents (with
the content allocation of the main gadget). Dropping the second call
fixes the render node conversion for GtkHeaderBar.
2016-08-07 15:08:44 -04:00
Colomban Wendling
64521345b9 GtkMenuButton: use :toggled instead of :clicked
:toggled is triggered on :clicked, so using :toggled lead to the menu
to be popped up at the same time, while allowing to use the toggle state
and avoiding any need to a hack to prevent recursion, which somehow
wasn't enough for double emission of GtkMenuToolButton:show-popup.

https://bugzilla.gnome.org/show_bug.cgi?id=769287
2016-08-04 23:31:42 +02:00
Timm Bäder
2f9b22243a csslookup: Remove useless cast
The style passed to _gtk_css_lookup_resolve is already a
GtkCssStaticStyle.
2016-08-04 20:40:34 +02:00
Руслан Ижбулатов
5e6c1928b4 W32: Prefer the deadkey combinations that the OS uses
Pick the W32 API for possible deadkey+<something> combinations
and prefer these to other sources of deadkey combos.
Specifically, if W32 API supports at least one combo for a particular
deadkey, only use that data and do not attempt to do other, unsupported
combinations, even if they make sense otherwise.
This is needed to, for example, correctly support US-International
keyboard layout, which produces a combined character for <' + a>
combo, but not for <' + s>, for example.

This is achieved by stashing all the deadkeys that we find in
an array, then doing extra loop through all virtual key codes and
trying to combine them with each of these deadkeys. Any combinations
that produce a single character are cached for later use.

In GTK Simple IM context, call a new GDK W32 function to do a lookup
on that cached combination table early on, among the "special cases"
(which are now partially obsolete).

A limitation of this code is that combinations with more than
one deadkey are not supported, except for combinations that consist
entirely of 2 known deadkeys. The upshot is that lookups should
be relatively fast, as deadkey array stays small and the combination
tree stays shallow.

Note that the use of ToUnicodeEx() seems suboptimal, as it should
be possible to just load a keyboard library (KBD*.DLL) manually
and obtain and use its key table directly. However, that is much more
complicated and would result in a significant rewrite of gdkkeys-win32.
The code from this commit, though hacky, is a direct addition to
existing code and should cover vast majority of the use-cases.

https://bugzilla.gnome.org/show_bug.cgi?id=569581
2016-08-04 16:37:19 +00:00
Timm Bäder
60f3c02080 shortcutlabel: Make parameter names match 2016-08-04 13:33:07 +02:00
Timm Bäder
50a0c5f242 widget: remove some unneeded function prototypes 2016-08-04 13:09:08 +02:00
Timm Bäder
6032e7b216 filechooserwidget: Remove unused enum 2016-08-04 13:09:01 +02:00
Timm Bäder
41c8446631 label: Destroy the popup menu on unmap 2016-08-04 13:07:23 +02:00
Matthias Clasen
2de6b0800d menu: Fix up new properties
The new positioning-related properties had some quality of
implementation issues, such as incorrect initial values and
excessive change notification. This broke the notify test.
2016-08-02 23:57:37 -04:00
Emmanuele Bassi
831d8b1261 Update the docs for gtk_window_get_position()
Drop mentions of GnomeClient, and add mentions of different windowing
systems instead of just assuming that we're using X11.
2016-08-02 12:29:26 +01:00
Christoph Reiter
4556d0f0fe gtkbox: remove unnecessary queue_compute_expand()
The expand child property does not have any effect on the
expand state of the GtkBox, so queuing a compute_expand
when changing it is not needed.

https://bugzilla.gnome.org/show_bug.cgi?id=769162
2016-08-01 09:02:04 -04:00
Christoph Reiter
a72f1c76c8 gtktable: don't try to propagate expand related child props in compute_expand()
It tried to set the expand state if either xexpand/yexpand where true.
Due to a missing queue_compute_expand when adding a child it actually
only computed the expand state in case a child queued after being added
or in case a child had the expand property set (see optimization in
gtk_widget_set_parent)

In my case this broke layouts as a child of GtkCombBox started setting
an exand flag with 3.20 which queued a compute_expand, which in turn
propagated an expand child props set for a cell in the same table up
and overrode the expand child prop of a parent GtkBox.

This removes the custom compute_expand implementation to match the
behaviour of GtkBox (don't propagate child prop expand flags
but let child expand flags override the child props) and not get random
expand behaviour depending on whether and when child widgets set their
expand state.

https://bugzilla.gnome.org/show_bug.cgi?id=769162
2016-08-01 09:02:04 -04:00
Matthias Clasen
b01e975667 Port GtkPrintOperationPortal to gtk_window_export_handle 2016-07-28 13:01:22 -04:00
Matthias Clasen
936c1649de Port gtk_show_uri_on_window to gtk_window_export_handle
As part of this, make it use the new
g_app_info_launch_default_for_uri_async, since we need a way
to unexport the window handle afterwards.
2016-07-28 13:01:22 -04:00
Matthias Clasen
237df66a33 Port GtkFileChooserNativePortal to gtk_window_export_handle 2016-07-28 13:01:22 -04:00
Matthias Clasen
18aa05110f Put window exporting behind a display protocol agnostic API
Introduce a private API meant for abstracting how to get a handle
of a window that can be shared with other processes. The API is
async, since some implementations will require that. Currently,
only X11 is supported, which doesn't.

Based on a patch by Jonas Adahl.
2016-07-28 13:01:22 -04:00
Benjamin Otte
5ccc0e40f5 css: Fix case where we didn't convert pt => px
Fixes gtk_widget_override_font() not behaving properly since commit
df08fc91bd.

https://bugzilla.gnome.org/show_bug.cgi?id=768902
2016-07-28 11:42:54 -04:00
Benjamin Otte
f422208040 css: Rename functions
I don't want to use a generic function for font sizes as font sizes are
special. Plus, the function is only used in one place.
2016-07-28 11:42:54 -04:00
Murray Cumming
0fd69ce99a Fix tiny typo. 2016-07-28 11:12:54 +02:00
Lapo Calamandrei
15b2dbf9a8 Adwaita: use a transition to animate checks and radios
see https://bugzilla.gnome.org/show_bug.cgi?id=762260
2016-07-27 23:00:35 +02:00
Georges Basile Stavracas Neto
ddee89f4a3 shortcut-label: add 'disabled-text' property
When there's no useful shortcut accelerator set,
GtkShortcutLabel doesn't show any useful information.

To work around that, add a new property to set the
text to be displayed when there's no accelerator
available.

https://bugzilla.gnome.org/show_bug.cgi?id=769205
2016-07-27 16:30:43 -03:00
Georges Basile Stavracas Neto
7543cd8ce4 shortcut-label: make it public
GtkShortcutLabel is a widget that displays a single
shortcut accelerator or gesture in the user interface,
and is currently used by the shortcuts window.

This widget, however, has public value as other applications
also may want to expose their own shortcuts. For instance,
it'll be useful for the Keyboard panel on Control Center and
the new shortcut editor in Pitivi, among others.

This patch exposes GtkShortcutLabel as a public widget,
and adds the necessary documentation.

https://bugzilla.gnome.org/show_bug.cgi?id=769205
2016-07-27 16:30:31 -03:00
Carlos Garnacho
8e6a68c5fc GtkScrolledWindow: reset scroll history if it contained no real history
Scroll history must refer to a timespan for the values to be valid, otherwise
we return FALSE, in this case the stored event(s) should be discarded anyway.
2016-07-27 20:29:38 +02:00
Carlos Garnacho
9eb356d9f2 GtkScrolledWindow: Always uninstall scroll cursor when starting deceleration
It could be the case that the last scroll event is received long after any
previous scroll event, in this case the last scroll event discards all "old"
scroll events, and scroll_history_finish() returns FALSE because there's no
time/offset deltas in the scroll history.

This is desired so we don't trigger the deceleration effect if there was no
effective velocity, we still must reset the installed scroll cursor, so take
it out of this if() condition.
2016-07-27 20:28:23 +02:00
Matthias Clasen
45957100b0 Revert a mistaken change
I thought I needed ot rearrange the ordering of the animation-direction
values for the parser, overlooking the fact that we already parse them
backwards to address this very problem.
2016-07-26 00:21:48 -04:00
Sébastien Wilmet
af5ef152b1 docs: GtkWidget::style-updated vs GtkStyleContext::changed
Explain the difference between those two signals.

Add "Since: 3.0" for GtkStyleContext::changed, since that signal has
been added in commit 9f84e101bf, present
since 2.91.6.

https://bugzilla.gnome.org/show_bug.cgi?id=769047
2016-07-25 09:26:41 -04:00
Sébastien Wilmet
3432587f89 docs: add missing info for gtk_widget_get_style_context()
It is important to know whether the returned object can or cannot
change, for a certain widget. For example to connect to the
GtkStyleContext::changed signal.

https://bugzilla.gnome.org/show_bug.cgi?id=769047
2016-07-25 09:26:41 -04:00
Matthias Clasen
f089d7503b GtkApplication: some cleanups
Always return an error if we fail to get a dbus proxy; the callers
are only looking whether error is set, not whether the return value
is NULL.

Use the same function for the inhibit proxy as well, and clean up
the sm_proxy in finalize.
2016-07-25 09:11:21 -04:00
Simon McVittie
f65c116d2a Don't apply GDK_HINT_RESIZE_INC to GDK_WINDOW_STATE_TILED windows
This matches the behaviour of Mutter, Metacity and traditional X11
window managers on the window manager side, and is what we want
for at least gnome-terminal. I can't think of any reason why we'd
want incremental resize in any other tiled window.

Signed-off-by: Simon McVittie <smcv@debian.org>
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=760944

https://bugzilla.gnome.org/show_bug.cgi?id=755947
2016-07-25 09:00:01 -04:00
Matthias Clasen
8db8891c66 Use g_clear_object in a few more places 2016-07-25 08:32:08 -04:00
Lapo Calamandrei
bab906f838 Adwaita: use margins for entry image spacing
there used to be a padding, since it was the only option in the
past, that makes little sense now, hence use margins.
2016-07-24 19:11:27 +02:00
Lapo Calamandrei
bc070a70a5 HC: export unfocused_insensitive_color...
...to reflect Adwaita changes.
2016-07-24 18:59:58 +02:00
Lapo Calamandrei
f84b8c1c26 Adwaita: export yet another named color for insensitive bg/fg...
...which was needed, but not there, use responsibly.
2016-07-24 18:04:51 +02:00
Timm Bäder
dd09e252c9 Fix a few documentation parameter names 2016-07-23 14:07:47 +02:00
Matthias Clasen
82fd72a477 Opt in to structured logging
Define G_LOG_USE_STRUCTURED, so that all our g_debug, g_warning,
etc calls directly use structured logging and provide source information.
2016-07-22 23:13:20 -04:00
Lapo Calamandrei
1b0ade8bdd Adwaita: dim calendar uninteresting days some more
see https://bugzilla.gnome.org/show_bug.cgi?id=769003
2016-07-22 14:10:56 +02:00
Lapo Calamandrei
80b28a17f7 Adwaita: clean calendar styling
remove some non working props and generally simplify and clean up
the code, in the process address:

https://bugzilla.gnome.org/show_bug.cgi?id=769003
2016-07-20 22:48:25 +02:00
William Hua
8701e34f74 port to new gtk_menu_popup_at_* () functions
https://bugzilla.gnome.org/show_bug.cgi?id=756579
2016-07-19 09:38:54 -04:00
William Hua
b03361366a gtkmenu: add gtk_menu_popup_at_* ()
Adds the following functions:

gtk_menu_popup_at_rect ()
gtk_menu_popup_at_widget ()
gtk_menu_popup_at_pointer ()

https://bugzilla.gnome.org/show_bug.cgi?id=756579
2016-07-19 09:38:54 -04:00
Руслан Ижбулатов
2465493237 Try to delete any existing W32 status icons when DPI changes
Fixes some or all of the issues reported in
https://github.com/hexchat/hexchat/issues/1423
2016-07-18 10:32:04 +00:00
Matthias Clasen
793d5291d9 link button: Use gtk_show_uri_on_window
This gives slightly better behavior in the sandboxed case and
makes no difference otherwise.
2016-07-13 11:24:16 -04:00
Patrick Griffis
6184edc318 GtkFileChooserNativePortal: Use correct response code for accept
As per the documentation of GtkFileChooserNative.

https://bugzilla.gnome.org/show_bug.cgi?id=768756
2016-07-13 07:02:15 -04:00
Chun-wei Fan
d8815f3e4e gtk/gtkfilechoosernativeportal.c: Don't use g_autoptr()
This code is also built by non-GCC/CLang compilers, so use the normal
ref/unref steps for the GVariant.

https://bugzilla.gnome.org/show_bug.cgi?id=768659
2016-07-13 11:26:35 +08:00
Matthias Clasen
500c9dbe96 Remove a few mentions of GtkVBox in the docs
We only have GtkBox nowadays.
2016-07-12 21:58:11 -04:00
Emmanuele Bassi
8f1fd7d964 docs: Fix typo. 2016-07-12 18:09:20 +01:00
Emmanuele Bassi
c0dae6c146 docs: Attempt a better explanation for gtk_widget_destroy()
Clarify the nature of this function, and the expectations after it's
been called.
2016-07-12 12:44:23 +01:00
Piotr Drąg
3831b2a725 placesview: fix spelling of WebDAV in a translatable string 2016-07-12 13:31:21 +02:00
Razvan Chitu
9747312e6c placesview: fix open action for locations without a mount or volume
In the "Other Locations" view, locations can be opened from the context menu
based on their mount or volume. However, some locations, like "Computer", do not
have either of those so they cannot be opened from the context menu. In order to
fix this, the file associated with the location can be used as well.

https://bugzilla.gnome.org/show_bug.cgi?id=768657
2016-07-12 12:56:04 +03:00
Lapo Calamandrei
53a1f0ba78 Adwaita: remove old popover checkbutton/radiobutton style
remove some now disfunctional style regarding checkbuttons and
radiobuttons inside popovers which shouldn't be needed anymore.
2016-07-09 00:56:32 +02:00
Lapo Calamandrei
725d5b1477 Adwaita: remove some check and radio related cruft
remove superfluous padding which was there before the css node
conversion of the widget.
2016-07-08 22:11:07 +02:00
Matthias Clasen
3756a1eca2 Fix up the documentation of GtkTextView::move-viewport
Pointed out in https://bugzilla.gnome.org/show_bug.cgi?id=768546
2016-07-08 10:26:22 -04:00
Matthias Clasen
25fed03420 Add portal support to GtkPrintOperation
Make GtkPrintOperation talk to org.freedesktop.portal.Print when
running in a sandbox.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:09:05 -04:00
Matthias Clasen
56ebfc6ca5 print job: Add api to send data from an fd
This will be useful for portalized print support.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:09:02 -04:00
Matthias Clasen
cc97e55f8a page setup: Add api to serialize to a GVariant
This will be useful for portalized print support.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:57 -04:00
Matthias Clasen
42ae48a23e Add api to serialize a GtkPaperSize to a GVariant
This will be useful in portalized printing.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:53 -04:00
Matthias Clasen
a3764e27a4 print settings: Add api to serialize to a GVariant
This will be useful for portalized print support.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:47 -04:00
Matthias Clasen
04cec61aa1 Support the Inhibit portal
xdg-desktop-portal now has a portal for inhibiting session status
changes. We don't need to use it if we can talk to the session
manager, but if can't, try org.freedesktop.portal.Inhibit.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:37 -04:00
Matthias Clasen
4600209962 Implement combobox apis for GtkFileChooserNativePortal
https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:33 -04:00
Matthias Clasen
5839c13843 Implement combobox apis in GtkFileChooserWidget
https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:28 -04:00
Matthias Clasen
6fc7485077 GtkFileChooser: Add abstract api for comboboxes and checkbuttons
This commit adds API for adding combo boxes and check buttons to
GtkFileChooser, and getting the selected value back in ::response.

In contrast to gtk_file_chooser_set_extra_widget, these APIs are
abstract and suitable for implementation in GtkFileChooserNative.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:21 -04:00
Matthias Clasen
2b77eaa1ae Adapt to simplified file chooser portal api
OpenFiles is gone, just set the 'multiple' option instead.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:17 -04:00
Matthias Clasen
445d12e5cb file chooser portal: Port to new Request API
The portal API has changed, we need to adapt.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:11 -04:00
Matthias Clasen
6cf71ed6bb Use the portal when sandboxed
Use the sandbox helper api to find out whether to use
the file chooser portal.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:08:05 -04:00
Matthias Clasen
3b3e1eca2b portal: Sent more data along
Send the current_name, current_folder or current_file fields to
the portal.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:07:59 -04:00
Matthias Clasen
0acb6a0e98 portal: Send file filters to the portal
https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:07:54 -04:00
Matthias Clasen
35181e793d Add api to serialize GtkFileFilter to a variant
This will be used to send filters over D-Bus in future commits.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:07:49 -04:00
Alexander Larsson
01d0bac83e Initial version of file chooser portal support
Implement GtkFileChooserNative for sandboxed applications
by talking to org.freedesktop.portal.FileChooser. Currently,
this supports OPEN and SAVE mode.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:07:42 -04:00
Matthias Clasen
e47957406d about dialog: Use gtk_show_uri_on_window
This allows us to handle things slightly better in
the portal case.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:07:11 -04:00
Matthias Clasen
c6416aec74 Add gtk_show_uri_on_window
The gtk_show_uri API doesn't let us specify a parent window. With
portals, there may be an intermediate dialog, for which it is nice
to have parent window information, to place it properly.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:07:07 -04:00
Matthias Clasen
96d8afbccb Add a portal helper
Instead of open-coding checks for sandboxing in multiple
places, add a helper function that caches the information.

https://bugzilla.gnome.org/show_bug.cgi?id=768499
2016-07-08 00:06:14 -04:00
Owen W. Taylor
b840a59766 Change the priority of the window-close idle to G_PRIORITY_DEFAULT
If we have an application that never goes idle (or takes a long time to
go idle), the close buttons in CSD decoration don't work properly.
While it's not clear why the usage of an idle was added in the first
place, keep on using it to avoid unexpected reentrancy problems, but
change the priority to G_PRIORITY_DEFAULT.

https://bugzilla.gnome.org/show_bug.cgi?id=768485
2016-07-06 09:50:21 -04:00
Matthias Clasen
3c4be3c912 Fix parsing of some css properties
Our property parser stops at the first match when looking for
enums, so we need to order our values so that we don't end up
with prefixes of longer names being found first.

I noticed this when the parser tried to interpret
background-blend-mode: color-burn; as "color, with junk at the end".

It also affects animation-direction, which is also fixed here.
2016-07-03 17:54:19 -04:00
Matthias Clasen
34986231f4 css parser: Improve an error message
Say what property we are dealing with when we complain about the
value.
2016-07-03 17:46:30 -04:00
Georges Basile Stavracas Neto
369db4a406 css: add background-blend-mode support
CSS supports blend modes, in which a series of layers are
merged together according to the given operation or set of
operations.

Support for blend modes landed on Cairo, which exposes all
the commons and also the exquisites blend modes available.
Adding support for blend modes, then, is just a matter of
using the available Cairo operations.

This patch adds the background-blend-mode CSS enum property,
and adapts the background rendering code to blend the backgrounds
using the available blend modes when they're set.

https://bugzilla.gnome.org/show_bug.cgi?id=768305
2016-07-03 17:24:47 -04:00
Matthias Clasen
c8a74a1f50 Simplify the xfce session manager support
We don't have to check name owners twice; reuse the information
we already have.
2016-07-03 17:22:14 -04:00
Eric Koegel
3c7cd7ac23 GtkApplication: Add support for the Xfce session manager
Xfce4-session-manager added support for managing dbus based clients.
This patch adds support for checking if Xfce session manager is
around after trying the gnome one.
https://bugzilla.gnome.org/show_bug.cgi?id=693203
2016-07-03 14:56:48 -04:00
Matthias Clasen
a43fce058c print dialog: Apply initial capabilities
I was struggling to understand why calling
gtk_print_unix_dialog_set_manual_capabilities (...,0)
was not having the expected effect of hiding the preview
button. The initial capabilities were not applied at all.
2016-07-02 18:08:38 -04:00
Chun-wei Fan
51444b7909 gtk/gtkstylecascade.c: Fix formatting
My previous patch on this file did not take into the account of the
formatting, sorry!
2016-06-30 18:00:47 +08:00
Ray Strode
386d17db6d headerbar: don't throw a warning if title widget is hidden
commit 0015ebc4a8 reworked
some of the titlebar size allocation code.  Those changes
inadvertently introduced a warning when the application sets
the headerbar title widget to be hidden.

This commit fixes that warning.

https://bugzilla.gnome.org/show_bug.cgi?id=768184
2016-06-29 09:27:33 -04:00
Timm Bäder
92de947d5e GtkWindow: Check for GtkWidget-window-dragging in multipress gesture
This partly reverts 9f5b9c0e07, which
removed the check for GtkWidget-window-dragging in the multipress
gesture. This check is still needed for widgets which have this style
property set (e.g. menubars and toolbars) can maximize the window on
double click -- but those widgets which have it set to FALSE shouldn't
maximize the window.
2016-06-28 21:42:55 +02:00
Ondrej Holy
0c1cc9832f gtkfilesystem: Improve heuristics to detect remote filesystem
Use G_FILE_ATTRIBUTE_FILESYSTEM_REMOTE to detect remote filesystems
instead of hardcoded list of filesystem types.

Bump required GLib version accordingly.

https://bugzilla.gnome.org/show_bug.cgi?id=767965
2016-06-28 10:55:06 +02:00
Timm Bäder
9f5b9c0e07 GtkWindow: Fix dragging on non-titlebar widgets 2016-06-27 19:23:12 +02:00
Timm Bäder
d52f6ff710 widget: Don't unnecessarily export function
_gtk_widget_get_translation_to_window is only used in gtkwidget.c
2016-06-27 14:43:53 +02:00
Chun-wei Fan
5f525839da gtk/gtkstylecascade.c: Declare variables at beginning of block 2016-06-27 11:58:42 +08:00
Chun-wei Fan
cd44f6d875 Fix build of commit f23e99b
Commit f23e99b made use of snprintf(), which is not universally available.
Fix this by using g_snprintf().
2016-06-27 11:53:30 +08:00
Lapo Calamandrei
ebccc8b23f Adwaita: color the selection of entries with error...
...or warning style class applied.
This particular style bit wasn't converted to the saner 3.20 way
so `entry:selected` used in place of `entry selection`.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=768025
2016-06-25 18:29:51 +02:00
Lapo Calamandrei
d74e4af1a2 Adwaita: restrict infobar styling to, well, infobars
we used to style infobars by using the .info, .question, .warning
and .error selectors directly, which used to be ok when we had just
styleclasses all over the place, now it needs to be more specific
or it interferes with everything with those styleclasses applied
like entries.
2016-06-25 18:17:59 +02:00
Ray Strode
0015ebc4a8 headerbar: support expand property for children
The header bar currently ignores the expand property on its
children. This commit changes the code to honor that property.

It divvies up any free space and distributes it equally to packed
children (with any left over space given out a pixel at a time
on a first come, first serve basis).

This commit also adds support for the title widget to be made
expandable.

It accomplishes this by using up the padding the title widget
is centered with.

https://bugzilla.gnome.org/show_bug.cgi?id=724332
2016-06-24 20:58:41 -04:00
Ray Strode
847a43c431 headerbar: compute nominal size of sides up front
In order to support the expand property on children, we're
going to need to look at the size of the packed children
on each side of the title widget, up front, before allocating
them (to compute how much extra allocation each expanded child
gets).

This commit lays the groundwork for that analysis by splitting
the size calculation of each side of the header bar outside
of the loop that allocates each child of the header bar.

https://bugzilla.gnome.org/show_bug.cgi?id=724332
2016-06-24 20:24:43 -04:00
Ray Strode
eccfce50d6 headerbar: simplify some size allocation code
With a headerbar, the widget in the center may be a label, constructed
internally, or a custom widget, constructed externally.  The size
allocation code needs to handle either case the same way for the most
part.  There's more than one place in the code that checks which of
the two widgets to use and does some operation on the selected one.

This commit simplifies the code by checking up front which one is the
center (title) widget and storing that in a temporary variable,
This allows reducing duplicated logic later on in the function.

https://bugzilla.gnome.org/show_bug.cgi?id=724332
2016-06-24 20:20:22 -04:00
Ray Strode
e0bebba470 headerbar: simplify calculation in size allocation code
There's some extraneous MIN() calls that have predetermined answers.

This commit drops them and then simplifies a few redudant checks into
one MIN call.

https://bugzilla.gnome.org/show_bug.cgi?id=724332
2016-06-24 20:19:32 -04:00
Matthias Clasen
8430be75ed Fix an array overrun
When I introduced the new input source enum value for
track points, I forgot that there is a fixed-size array
with names for those in the inspector.
2016-06-23 15:47:47 -04:00
Ray Strode
78eccd0db9 headerbar: don't use GTK_PACK_* as indices into array
The size allocation code maintains an array of two elements,
to track the allocation of children packed into the two sides
of a header bar. Sometimes this array is indexed with 0 and 1,
and sometimes its indexed with GTK_PACK_START and GTK_PACK_END.
The latter happen to have the values 0 and 1, respectively, but
that's not really obvious.

For clarity, this commit changes the code to index those arrays
consistently, sticking to 0 and 1 across the board.
2016-06-23 14:32:33 -04:00
Ray Strode
01eda209e1 headerbar: don't allocate both label and custom title
It's only possible to have a label or a custom title, not both.
The size allocate code confusingly treats them as independent.

That is confusing, because, as the code is written, it makes it
look like the space for the custom title isn't getting accounted
for.

This commit else-ifies some parts of the size allocate code for
clarity.
2016-06-23 14:32:33 -04:00
Emmanuele Bassi
40ee61a686 gtk: Keep Firefox working in the DrawingContext world
Firefox does a bunch of interesting things with GTK.

If the top-level GtkWindow does not have a "csd" style class associated,
Firefox will happily draw the contents of the container used to render
HTML and XUL directly on the top level's GdkWindow; on the other hand,
if a "csd" style class is found, the MozContainer will create a new
child window, and draw on it.

Then, Firefox will proceed to disable double buffering on both the
top-level window and the MozContainer (unless they are backed by the
same GdkWindow, in which case only the top-level will be
single-buffered) *and* it will add a GDK_EXPOSURE_MASK flag to the
MozContainer events for good measure (even if this is only needed for
GTK+ 2.x).

After landing the GdkDrawingContext API in GdkWindow, GTK enabled
automatic double buffering on all top-level windows backed by a native
surface, ad most users of single buffering rely on child widgets instead
of top-levels, and we'd still like to have the same double buffering
behaviour for all top-levels on all backends. Obviously, with Firefox
disabling double buffering on the top-level window, the change broke
their drawing mechanism.

Ideally, Firefox could be fixed to not disable double buffering on the
top-level window when MozContainer has a separate GdkWindow — i.e. the
CSD case — but since we did introduce a slight change of behaviour in
fringe users of the GTK+ API, let's keep backwards compatibility with
the old code for a little while longer, and create an intermediate Cairo
context unbound from the GdkDrawingContext, like we used to do until
GTK+ 3.20.
2016-06-22 11:41:59 +01:00
Carlos Garnacho
f44b9ef1f0 GtkColorEditor: Keep ref on pre-popup focus widget
And ensure it's still visible before returning the keyboard focus to it.
Because of the extra ref, add a dispose handler that will ensure the
ref is lost (by popping down), although this should be already ensured
through other paths (eg. when the popup widget loses visibility).

This fixes a possible crash in dispose paths, where we might be restoring
focus on an already destroyed widget, and at a time where, if the toplevel
is being itself disposed, no new focus should be set.

https://bugzilla.gnome.org/show_bug.cgi?id=767849
2016-06-21 17:56:34 +02:00
Carlos Garnacho
ed227f9e7b GtkPopover: Honor widget margins when calculating tail position
If there are widget margins set, the whole popover will be displaced.
However the calculation of the tail position doesn't have this into
account, ending up with the tail being detached from the popover if
the margin grew too big.

We should not render the arrows invariably next to the GdkWindow edge,
but optionally displaced inside it depending on the widget margins.
Fixes the gtk3-demo "Popovers" demo case, whose GtkEntry popovers set
widget margins for some reason.

https://bugzilla.gnome.org/show_bug.cgi?id=767851
2016-06-21 17:54:42 +02:00
Christian Hergert
3724592efb listbox: short-circuit if no sort function is set
do_sort will crash if sort_func is not defined. Instead of adding a check
there in the hot path, just check for sort_func before invalidating the
sort of the underlying GSequence.
2016-06-20 18:51:42 -07:00
Matthias Clasen
8a2812217b shortcuts: Try harder to disambiguate search items
I noticed that some of the gestures did not show up in the
search results in the builder example in gtk3-demo, because
they share the same title and don't have an accelerator to
disambiguate. Include the shortcut type to handle this case.
2016-06-19 23:24:44 -04:00
Carlos Garnacho
e33e23a6d9 treeview: Ensure the treeview has the implicit grab before grabbing focus
The cellrenderer signals might be taking the grab somewhere else, at which
point it's dubious we should attempt to take the keyboard focus into the
treeview.

This concretely breaks popovers triggered from cellrenderer signals on
button press, because the treeview will attempt to grab focus
inconditionally then.

https://bugzilla.gnome.org/show_bug.cgi?id=767468
2016-06-19 20:54:18 -04:00
Matthew Leeds
51799d41e4 GtkActionHelper: Change a message to a warning
It's almost certainly a programmer error if an action isn't
activatable because its target and parameter type don't match.
This commit changes the existing g_message to a g_warning.

https://bugzilla.gnome.org/show_bug.cgi?id=767705
2016-06-19 20:06:13 -04:00
Philip Chimento
76bacfde6e style cascade: Allow cascades with more ancestors
Previously a style cascade's parent could not have a parent itself. That
represented the two levels at which you could add a style provider: at
the screen level, with gtk_style_context_add_provider_for_screen(), and
at the style context level, with gtk_style_context_add_provider().

This commit changes no functionality, but this change will be necessary
for adding style providers in the future that apply to a subtree of the
widget tree. It relaxes the requirement that a style cascade's parent
must not have a parent, since in the future a style context may be
affected by any number of parent widgets' style contexts.

https://bugzilla.gnome.org/show_bug.cgi?id=751409
2016-06-19 19:47:05 -04:00
Matthias Clasen
1c6b878e09 shortcuts window: Fix handling of significant xml chars
We are using markup in the labels, so we need to escape things
like < and >.

https://bugzilla.gnome.org/show_bug.cgi?id=767795
2016-06-19 17:49:03 -04:00
Emmanuele Bassi
c3f4fe334d Deprecate gtk_widget_send_expose()
We have various replacements for what this function does, and we are not
calling it internally any more.
2016-06-10 16:13:27 +01:00
Tristan Van Berkom
a9a59ac55e scrolledwindow: Fix calls to gdk_window_move_resize()
This is a bit of fallout from 34feba1, now that we resolve
the has_indicators value earlier than realize, it becomes
possible to call gdk_window_move_resize() before realization.

Just added the appropriate checks.
2016-06-10 21:16:53 +09:00
Stefan Sauer
32675db490 docs: point to suggested api for deprecated functions
We deprecated a bunch of _focus_on_click() functions. Point to the new API
in GtkWidget.
2016-06-09 22:44:44 +02:00
Emmanuele Bassi
c9916fbc58 docs: Add deprecation notice for GtkButton.set_focus_on_click()
To match the deprecation annotation on the function declaration.
2016-06-09 21:35:36 +01:00
Emmanuele Bassi
c5d0522a23 Deprecate the gdk_window_begin/end_paint family of functions
They are replaced by the more appropriate gdk_window_begin_draw_frame()
and gdk_window_end_draw_frame() functions.

https://bugzilla.gnome.org/show_bug.cgi?id=766675
2016-06-09 17:45:40 +01:00
Emmanuele Bassi
dda6a0d385 Associate the drawing context to the Cairo context
Instead of associating the GdkWindow that created the GdkDrawingContext
we can directly bind the Cairo context to the GDK drawing context.

Cairo contexts created via gdk_cairo_create() go back to not having a
GdkWindow associated to them, like they did before we introduced the
gdk_window_begin_draw_frame() API.

https://bugzilla.gnome.org/show_bug.cgi?id=766675
2016-06-09 17:45:40 +01:00
Emmanuele Bassi
a7ef37da2a Add GdkDrawingContext
Instead of giving out Cairo contexts, GdkWindow should provide a
"drawing context", which can then create Cairo contexts on demand; this
allows us to future proof the API for when we're going to use a
different rendering pipeline, like OpenGL.

https://bugzilla.gnome.org/show_bug.cgi?id=766675
2016-06-09 17:45:40 +01:00
Emmanuele Bassi
2c7b21718f Simplify the widget rendering entry point
Now that GDK has the appropriate API, we can simplify the widget drawing
code into a single function.

https://bugzilla.gnome.org/show_bug.cgi?id=766675
2016-06-09 17:45:40 +01:00
Tristan Van Berkom
34feba1e7c scrolledwindow: Resolve whether we use overlay scrollbars early.
Widgets should support size requests before being realized in general,
otherwise this can cause flicker/resize at initial display time as
the toplevel window can make a request before realize/allocate.

This also makes the added testsuite/gtk/scrolledwindow.c test work again,
this was broken because we only ever calculate whether we are going
to use overlay scrollbars once the scrolled window is realized (and
the test case does not realize any window).
2016-06-09 14:57:10 +09:00
Tristan Van Berkom
d7e242eec0 scrolledwindow: Bug 766569 - General size request fixes.
This patch does a couple of things:

  o Removes the obscure 'extra_width' and 'extra_height' variables
    making the request code exceedingly difficult to read

  o Fixes the max-content-size properties introduced in bug 742281
    so that they do not grow the minimum request.

  o Cleanup of request code in general:
    - min/max content sizes are clamped around the child request as needed
    - scrollbar requests are only added in one place, after child request
      sizes are calculated and without the extra_width/height thing.
2016-06-09 14:57:10 +09:00
Jakub Steiner
6de0ecb78e HC: focus outline more prominent
- make visible regardless of colors used
- make larger and more inset

https://bugzilla.gnome.org/show_bug.cgi?id=767310
2016-06-08 08:38:19 -04:00
Lapo Calamandrei
199ad2c14b Adwaita: .osd button.color
remove box-shadow from colorbutton colorswatch when an ancestor has
the osd styleclass applied.
2016-06-08 13:39:39 +02:00
Lapo Calamandrei
a0ee8f7f0a Adwaita: centralize outline-color
define outline-color generally using alpha(currentColor, 0.3),
which avoids setting it in a bunch of different places.
2016-06-08 13:28:07 +02:00
Kjell Ahlstedt
ff5304ee13 Make gtk_scrolled_window_remove() smart
If gtk_scrolled_window_add() has added a GtkViewport,
  gtk_container_remove (GTK_CONTAINER (scrolled_window), child_widget);
or
  gtk_container_remove(GTK_CONTAINER(scrolled_window),
                       gtk_bin_get_child(GTK_BIN(scrolled_window)));
removes both the added child widget and the viewport.

https://bugzilla.gnome.org/show_bug.cgi?id=710471
2016-06-08 10:30:22 +02:00
Matthias Clasen
3f3a226090 Update GtkLabel::angle docs
We can have wrapped or ellipsized labels at an angle nowadays.
Still no selections though.
2016-06-07 23:40:32 -04:00
Matthias Clasen
f23e99b063 scale, spinbutton: Avoid -0.0
This is a very longstanding bug; time to finally put it to rest.

https://bugzilla.gnome.org/show_bug.cgi?id=118959
2016-06-07 23:04:40 -04:00
Matthias Clasen
7f06f2818a Ensure that GtkRange allocates enough space for the value
This is a long-standing problem of GtkScale.

https://bugzilla.gnome.org/show_bug.cgi?id=766372
https://bugzilla.gnome.org/show_bug.cgi?id=578626
https://bugzilla.gnome.org/show_bug.cgi?id=79229
2016-06-07 21:28:44 -04:00
Matthias Clasen
dce4003c20 Fix preconditions in set_max_content_size setters
We must allow resetting the properties to -1.
2016-06-07 16:37:09 -04:00
Benjamin Otte
775e277ab9 widget: Add classes to widget path even if no style context exists yet
This removes leftover code from when classes where added to the style
context.
Now that they get added directly to css nodes, the classes can exist
without a style context.

https://bugzilla.gnome.org/show_bug.cgi?id=767312
2016-06-07 16:12:18 +02:00
Carlos Garnacho
0d407bcd6c searchenginetracker: Fix prefix searches
With the FTS5 query syntax, when using quotes to delimit the search phrase
the '*' token must happen after the quote, or will otherwise be considered
a character to match, go through the tokenizer, and end up ignored in
result.
2016-06-07 00:56:38 +02:00
Matthias Clasen
289fc9640b Revert "scrolledwindow: Bug 766569 - Report child natural size unconditionally"
This reverts commit 0943c9f6b2.

The commit caused unexpected breakage in gtk3-widget-factory,
and also broke the just-added max-content-size properties.
2016-06-06 14:35:46 -04:00
Tristan Van Berkom
03742e83fb scrolledwindow: Bug 767238 - Fixing up for max content sizes
Needed to adjust this again after applying commit 4e5ecb7
for bug 742281. Now that we also have max content size properties,
pushed the addition of possible scrollbar sizes to after the
clause which clamps the child request size into min/max content
sizes.
2016-06-06 12:04:58 +09:00
Phillip Wood
72ea348ad6 Fix hiding popover when focus moves outside
Commit a01fe14 changed the behaviour of popovers when the focus leaves
them to stop child popovers being hidden when the focus leaves their
parent. However they are now a bit too reluctant to hide - if the
focus passes to an unrelated popover the first popover is not
hidden. Also if the focus passes to another widget that does not
perform a gtk grab then the popover isn't hidden until the user
presses a non-movement key or clicks outside the popover.

The solution is to go back to checking if the focused widget is a
descendant of the popover, but to include popovers and their related
widgets in the ancestry chain.

https://bugzilla.gnome.org/show_bug.cgi?id=765595
2016-06-05 22:54:00 -04:00