Commit Graph

2625 Commits

Author SHA1 Message Date
Timm Bäder
2d43ecf8ca widget: root child before emitting child observer signals
Try to avoid the childen_observer signal handlers screwing up internal
state.

Fixes #3364
2020-11-30 09:21:52 -05:00
Christoph Reiter
b97b95b0c7 gtkwidget: fix crashes with GTK_A11Y=none
at_context can be NULL in that case, so guard against it.

Fixes #3333
2020-11-15 15:59:31 +01:00
Carlos Garnacho
80de00a248 gtk/main: Stack ::active calls
We may have the situation of multiple touchpoints in the same
widget, or combinations with other devices. Stack those ::active
states are preserved on widgets on all but the last pointer/touch
going away.
2020-11-13 00:39:23 +01:00
Matthias Clasen
5d9799d4e3 Merge branch 'ebassi/for-master' into 'master'
ATContext lifetime fixes

Closes #3341

See merge request GNOME/gtk!2811
2020-11-12 12:35:23 +00:00
Emmanuele Bassi
8157717e03 a11y: Add missing fallback for accessible-role getter
If the per-instance accessible role is unset, use the class role.
2020-11-11 19:45:43 +00:00
Emmanuele Bassi
9052f6dafe a11y: Rework ownership and lifetime of GtkATContext
Now that GtkATContext is explicitly realized and unrealized, we should
always create an instance at widget initialization time, and drop it
during the widget finalization. This should make it easier to set up the
initial accessible state of a widget during the instance initialization,
as well as reduce the chances of accidental creation of GtkATContext
instances during the destruction sequence.
2020-11-11 19:45:43 +00:00
Emmanuele Bassi
428ec29598 a11y: Unrealize ATContext on dispose()
By unrealizing the context we avoid additional work during the dispose
phase, in case widget code updates the accessible state. We use
GtkAccessible's API, to ensure we unrealize the right ATContext, instead
of the one we store inside GtkWidgetPrivate.

We drop the ATContext instance inside GtkWidget during finalization, to
mop up eventual vivifications there.
2020-11-11 18:33:15 +00:00
Matthias Clasen
27c4c19333 Don't hide widgets in dispose()
This is leftover code from when widgets were hidden
by default, and was setting them back to their initial
state.

This is getting in the way now, as hiding the widget
updates the HIDDEN accessible state, which ends up
re-creating the at context that we've already disposed
of, leading to memory leaks.
2020-11-11 10:35:54 -05:00
Emmanuele Bassi
c264254d4b a11y: Parse reference lists using varargs
Using GList is a bit lame, and makes the API more complicated to use
than necessary in the common case.

The only real use case for a GList is gtk_widget_add_mnemonic_label(),
and for that we can use the GValue-based API instead.

Fixes: #3343
2020-11-10 14:31:56 +00:00
Emmanuele Bassi
1bdc8b9061 a11y: Proxy GtkWidget:accessible-role
The accessible-role property in GtkWidget has three possible targets:

 - the :accessible-role of a GtkATContext, if realized
 - the accessible_role field of GtkWidgetPrivate
 - the accessible_role field of GtkWidgetClassPrivate

When we set the accessible role of a widget using the GObject property
mechanism, we want to either set the GtkWidgetPrivate.accessible_role
field, if there's no ATContext *or* if the ATContext is not realized.
Conversely, when we get the accessible-role property we want to have a
series of fallbacks in place:

 - if GtkAccessible.get_at_context() returns an ATContext, and that
   ATContext is realized, return the :accessible-role of the context
 - if GtkWidgetPrivate.accessible_role is not WIDGET, return the
   stored accessible role
 - return GtkWidgetClassPrivate.accessible_role

This should help catch the case of getting the accessible role of
widgets that override GtkAccessible.get_at_context(), like
GtkModelButton.

See: #3342
2020-11-10 14:31:56 +00:00
Emmanuele Bassi
d88f7c9459 a11y: Use the tooltip text as the accessible description
It's more common to have a tooltip as the accessible description than an
explicit description. This is also the behaviour of GTK3.
2020-11-04 17:15:13 +00:00
Matthias Clasen
828a067b2b Propagate accessible-role to backend
We want to allow setting the accessible role as long
as the context is unrealized, so we need to propagate
the new role.
2020-11-04 11:53:07 -05:00
Emmanuele Bassi
dafb2b0361 a11y: Allow updating role on unrealized contexts
We can still change the accessible role if an ATContext instance is
unrealized.
2020-11-04 15:22:31 +00:00
Emmanuele Bassi
30210c7087 a11y: Move ATContext to an explicit realization model
We are doing too much work during the construction phase of the
AT-SPI backend for GtkATContext. Instead of having the AtSpiContext
register itself at construction time, let's add explicit realize
and unrealize operations, and connect the ATContext realization to the
rooting operation of a GtkWidget.
2020-11-04 15:02:44 +00:00
Matthias Clasen
3cf26b471b Fix the definition of GtkWidget:can-target
We are setting the value to TRUE initially, but
the property had a declared default of FALSE.
This is messing up the simplification of .ui files
with gtk4-builder-tool, since it thinks it can
omit can-target properties when it really can't.
2020-11-01 21:23:35 -05:00
Matthias Clasen
8df883ed71 Merge branch 'matthiasc/atspi-child' into 'master'
Emit ChildrenChanged signal on hierarchy updates

See merge request GNOME/gtk!2752
2020-10-27 18:24:43 +00:00
Matthias Clasen
0c49f7349d Merge branch 'matthiasc/a11y-buildable' into 'master'
Implement accessible attributes for ui files

See merge request GNOME/gtk!2751
2020-10-27 18:13:44 +00:00
Matthias Clasen
0a3b370805 Cosmetics
Remove some comments that refer to no longer
existing menu code.
2020-10-25 00:31:18 -04:00
Matthias Clasen
1e13acb206 widget: Don't add reordered children
gtk_widget_reposition_after is called both to add new children,
and to reposition existing children. We only want to emit
accessible changes in the former case (since AT-SPI doesn't
have events for reordering).
2020-10-24 11:17:56 -04:00
Emmanuele Bassi
80756322cd a11y: Handle relations in UI files
A bit hacky: we skip parsing values that have a reference or
reference-list type, but we do not error out. Instead, we return a NULL
value, which we catch in the GtkBuildable interface implementation to
get the actual object, and construct a reference list value.

There's still some ickyness around the value type that can only be
solved by having an attribute and role taxonomy.
2020-10-21 14:48:49 +01:00
Emmanuele Bassi
8d46bd1d7c a11y: Notify of changes in the children list 2020-10-21 14:04:49 +01:00
Emmanuele Bassi
7702670d86 docs: Document the "accessibility" custom tag 2020-10-21 13:33:37 +01:00
Emmanuele Bassi
9e4316bf96 a11y: Add an "accessibility" section to the builder XML tree
Accessible attributes are not GObject properties. This means that we
need a custom parser for setting attributes in our UI description files.

The new section is defined as a sub-tree with the `<accessibility>`
element at its root, and elements for each type of accessible
attributes, i.e. properties, relations, and states:

```xml
  <object class="..." id="...">
    <accessibility>
      <property name="label">The accessible label</property>
      <state name="pressed">false</state>
      <relation name="labelled-by">label1</relation>
    </accessibility>
  </object>
```

The name of the attribute is the enumeration value; the value is defined
by the WAI-ARIA specification.
2020-10-21 13:33:37 +01:00
Matthias Clasen
8880e3bd2e Merge branch 'matthiasc/for-master' into 'master'
Matthiasc/for master

See merge request GNOME/gtk!2729
2020-10-21 11:43:58 +00:00
Timm Bäder
0b5b908a74 Merge branch 'wip/exalm/nullable' into 'master'
widget: Add missing (nullable) on binding functions

See merge request GNOME/gtk!2730
2020-10-21 08:17:45 +00:00
Alexander Mikhaylenko
31a072497d widget: Add missing (nullable) on binding functions
Even though they are marked as (skip), they are used in Vala and have wrong
types there atm.
2020-10-21 12:27:51 +05:00
Matthias Clasen
a8baee342c a11y: Fix handling of LABELLED_BY relation
There were several places where we were confusing
GList and GSList and list->data and list->next, causing
a crash in the accessible name computation for buttons
with mnemonic labels.
2020-10-20 21:27:39 -04:00
Benjamin Otte
d9b216e629 widget: Always update the CSS node
Anybody who keeps their own CSS nodes around or wants to order CSS nodes
different from widgets will from now on have to do it manually all the
time.

This is outdated behavior, nobody should be doing either of those two
things.

Also, the correct case is much more common, and not doing it
automatically was causing bugs.

Fixes #3280
2020-10-20 04:50:12 +02:00
Matthias Clasen
75a0bef921 widget: Mark the size as changed when it changes
This notifies the AT context that the widgets size
has changed.
2020-10-19 12:19:55 -04:00
Matthias Clasen
34625142f4 docs: Expand opacity docs
Explain the situation with popovers and opacity.

Fixes: #3246
2020-10-15 22:39:01 -04:00
Matthias Clasen
9c28ecb0da Merge branch 'a11y/atspi' into 'master'
A11y/atspi

See merge request GNOME/gtk!2698
2020-10-15 14:59:24 +00:00
Matthias Clasen
a772d8fcac Merge branch 'wip/baedert/for-master'
Leaving out the gboolean -> bool change.
2020-10-14 15:07:09 -04:00
Timm Bäder
18c4bdad6f widget: Remove gtk_widget_list_devices() from private header 2020-10-14 15:06:12 -04:00
Timm Bäder
ec8f329332 widget: Slightly update opacity docs 2020-10-14 15:06:12 -04:00
Carlos Garnacho
024d832d94 gtkpopover: Add cascade-popdown property/functions
And honor it in gtk_popover_popdown(). By default, a GtkPopover
pops down automatically if a child popover was closed, if this
property is FALSE, the popover will remain opened.
2020-10-13 16:55:53 +02:00
Matthias Clasen
ade9e4b82a accessible: Add gtk_accessible_get_platform_state
As a companion to go with the platform_change api,
add a gtk_accessible_get_platform_state() function
that can be used by backends to get the platform
state.

This is in preparation for making entries inherit
their focus states from the text widget within.
2020-10-12 21:45:41 -04:00
Matthias Clasen
a2d220edfa widget: Notify a11y of focus changes
This is using the new 'platform changes' mechanism
to tell the a11y backends when focus changes.
2020-10-12 15:10:40 -04:00
Matthias Clasen
35f70d5a99 Pass orientation property to accessible
We should set the accessible properties we have,
where they make sense. So set orientation, if the
widget is orientable.
2020-10-12 15:10:40 -04:00
Emmanuele Bassi
658d910967 a11y: Ignore widgets during destruction
There's no point in creating a GtkATContext when a widget is getting
destroyed.
2020-10-12 16:19:32 +01:00
Emmanuele Bassi
8c18480092 a11y: Add a display to GtkATContext
Since we need to check at run time what kind of AT context to use, we
need a hook into the whole GDK backend machinery. The display connection
seems to be the best choice, in this case, as it allows us to determine
whether we're running on an X11 or Wayland system, and thus whether we
should create a GtkAtSpiContext.

This requires some surgery to fix the GtkATContext creation function, in
order to include a GdkDisplay instance.
2020-10-12 16:19:32 +01:00
Matthias Clasen
d2f0e3fa6b doc: Document some apis as widget implementor only
A better solution for this would be nice. For the
time being, just add a sentence to the docs.
2020-09-28 20:51:13 -04:00
Florian Müllner
12189bc10f buildable: Rename set_name() to set_buildable_id()
GtkBuildable's get_name()/set_name() methods may shadow
GtkWidget's methods. Avoid that by renaming the API to
get_buildable_id()/set_buildable_id(), which also reflects
the name of the XML attribute the API refers to.

https://gitlab.gnome.org/GNOME/gtk/-/issues/3191
2020-09-26 02:16:05 +02:00
Matthias Clasen
81793b62d8 docs: Improve gtk_widget_set_font_map docs 2020-09-20 10:59:37 -04:00
Matthias Clasen
d89f7fd046 Drop gtk_widget_new
An abstract type should not have a constructor.
If you want to be generic, just use g_object_new.
2020-09-19 16:58:42 -04:00
Matthias Clasen
67c1600ac2 Document some API as widget-implementation only
Using gtk_widget_insert_before on a complex container
is a *bad* idea; it will mess up the containers bookkeeping
of its children and can easily lead to failure and crashes.
2020-09-14 20:28:44 -04:00
Matthias Clasen
929a56e53c Clean up lots of GTK+ -> GTK
Replace most remaining uses of GTK+ in the docs and
user-visible strings by GTK. Also remove some leftover
"Was added in 3.x" sentences from the docs.
2020-09-12 12:01:04 -04:00
Benjamin Otte
8cb03a8cc0 widget: Add a critical when allocation is too small. 2020-09-09 17:38:37 +02:00
Matthias Clasen
30ab8b3eca Fix picking
My previous change here was too hasty - this code is not
actually transforming points - it is just a convoluted
way to transform the z axis from child coordinates to
the parent.
2020-09-02 15:20:58 -04:00
Matthias Clasen
e36b594867 Replace uses of graphene_matrix_transform_
Replace our uses of graphene_matrix_transform_point,
_point3d and _bounds by our own versions that handle
projective transforms correctly.

This fixes render node bounds being incorrect for widgets
involving projective transforms (e.g. testrevealer swing
transformations), and also fixes picking on such widgets.
2020-09-01 09:18:38 -04:00
Matthias Clasen
44741e1b77 widget: Avoid a crash in crossing event handling
We need to make sure that the crossing data stays
alive until we are done handling it, so take references
on all the widgets in it.
2020-08-26 07:26:49 -04:00
Matthias Clasen
40d7092f4a widget: Queue a draw when we need to
The code in gtk_widget_real_css_changed assumes that
queue_resize > queue_allocate > queue_draw, but the
second one is not really true. These days, we happily
keep reusing the same render node even when the child
allocation is changed.

So, if a css change has flags that tell us we should
redraw, we need to queue a draw, otherwise we might
end up reusing an outdated render node.

This fixes spinners staying visible when they stop
spinning, despite the theme setting their opacity
to 0.
2020-08-23 17:07:37 -04:00
Matthias Clasen
125ed52ccb Merge branch 'new-sysprof' into 'master'
Port profiling to sysprof-collector api

See merge request GNOME/gtk!2457
2020-08-21 23:58:09 +00:00
Matthias Clasen
d4e069a629 Port tracing to the sysprof collector api
Use the new sysprof collector api to do tracing.
2020-08-21 10:55:01 -04:00
Björn Daase
6315cd977c *: Fix spelling mistakes found by codespell 2020-08-21 15:29:34 +02:00
Matthias Clasen
ae7cefd97d Drop style class defines
We document the supported style classes by name,
not by macro name, and these macros don't really
add any value. Drop them for GTK 4.
2020-08-14 07:03:27 -04:00
Matthias Clasen
f561000124 docs: Clean up many dangling links
In many cases, these were references to no longer
existing api, so some rewording was necessary.
2020-08-04 22:11:34 -04:00
Matthias Clasen
ed13e6a41e grid layout: Rename some properties
Rename GtkGridLayoutChild:left-attach/top-attach to
GtkGridLayoutChild:column/row. Update all users.

Fixes: #2967
2020-08-02 17:58:03 -04:00
Matthias Clasen
17208ca7fa Go back to keeping a list of event controllers
This reverts 4b9e5304cc. It caused a number of regressions,
such as event controller ordering changes.
2020-08-02 12:01:06 -04:00
Matthias Clasen
1c16cb751f Revert "Add gtk_widget_prepend_controller"
This reverts commit 692cc42e10.
2020-08-02 12:01:06 -04:00
Matthias Clasen
8c66e2ff66 widget: Drop unnecessary includes
We don't need to include gobjectnotifyqueue.c anymore,
since style properties are gone. This should fix the
build with -Werror.
2020-07-31 10:34:37 -04:00
Matthias Clasen
4d012a6d95 Rename gdk_surface_queue_expose
We don't have expose events anymore; instead, there
is a ::render signal. So rename queue_expose to
queue_render to match.

Update all callers.
2020-07-29 22:40:01 -04:00
Matthias Clasen
b5f256937f Merge branch 'wip/carlosg/device-api-cleanup' into 'master'
Cleanups to device/event API

See merge request GNOME/gtk!2313
2020-07-28 23:49:35 +00:00
Carlos Garnacho
cab1dcb696 gdk: Conflate GDK devices
Make GdkEvents hold a single GdkDevice. This device is closer to
the logical device conceptually, although it must be sufficient for
device checks (i.e. GdkInputSource), which makes it similar to the
physical devices.

Make the logical devices have a more accurate GdkInputSource where
needed, and conflate the event devices altogether.
2020-07-29 01:27:51 +02:00
Emmanuele Bassi
5fc5e52f52 Merge branch 'a11y-tests' into 'master'
Add more a11y tests

See merge request GNOME/gtk!2303
2020-07-28 11:47:14 +00:00
Timm Bäder
f4d9e4032f widget: Compare proper adjusted size to new size when allocating
We were comparing first but then still modifying the adjusted size later
on, leading to unnecessary resizes.
2020-07-28 05:34:11 +02:00
Matthias Clasen
e10aedccc6 widget: Update accessible state
The sensitive property corresponds to the accessible
property disabled, so update it when sensitivity
changes.
2020-07-27 21:18:10 -04:00
Matthias Clasen
bc0182fa3e Merge branch 'ebassi/new-a11y' into 'master'
Accessibility rework

See merge request GNOME/gtk!2239
2020-07-27 11:55:19 +00:00
Matthias Clasen
06c6d62629 listlistmodel: Drop item_type
We are using G_TYPE_OBJECT throughout, so there
is no need for GtkListListModel to do anything else.

Update all callers.
2020-07-26 18:02:51 -04:00
Emmanuele Bassi
82664003c0 a11y: Allow setting the role post-construction
Some widgets have different roles after they are constructed, so we need
to allow changing the role defined by the class. We should still avoid
setting a role after the GtkATContext has been created.
2020-07-26 21:27:03 +01:00
Emmanuele Bassi
a9e0cf77a0 a11y: Update labelled-by when setting mnemonic labels
The ARIA labelled-by attribute behaves the same way as the mnemonic
labels in GTK.
2020-07-26 20:31:15 +01:00
Emmanuele Bassi
572861be27 Allow setting the accessible role at construction
Some widgets have different accessible roles depending on some
parameter, so we cannot set the role at class init time. For those
widgets, we add an "accessible-role" property to GtkAccessible, and we
allow setting it (only) at construction time.
2020-07-26 20:31:15 +01:00
Emmanuele Bassi
3ea069a8d3 Update the accessible state on widget visibility changes
The GTK_ACCESSIBLE_STATE_HIDDEN has the opposite meaning of the
GtkWidget:visible property.
2020-07-26 20:31:14 +01:00
Emmanuele Bassi
25f6da5e89 Have GtkWidget implement GtkAccessible
Each widget type has an accessible role associated to its class, as
roles cannot change during the life time of a widget instance.

Each widget is also responsible for creating an ATContext, to proxy
state changes to the underlying accessibility infrastructure.
2020-07-26 20:31:14 +01:00
Emmanuele Bassi
14faec3ce2 Introduce GtkAccessible
GtkAccessible is an interface for accessible UI elements.

Currently, it doesn't do much except exist as a type; in the future, it
will be the entry point for all accessible state in GTK.
2020-07-26 20:31:14 +01:00
Emmanuele Bassi
c63087a563 Remove ATK
To build a better world sometimes means having to tear the old one down.
        -- Alexander Pierce, "Captain America: The Winter Soldier"

ATK served us well for nearly 20 years, but the world has changed, and
GTK has changed with it. Now ATK is mostly a hindrance towards improving
the accessibility stack:

 - it maps to a very specific implementation, AT-SPI, which is Linux and
   Unix specific
 - it requires implementing the same functionality in three different
   layers of the stack: AT-SPI, ATK, and GTK
 - only GTK uses it; every other Linux and Unix toolkit and application
   talks to AT-SPI directly, including assistive technologies

Sadly, we cannot incrementally port GTK to a new accessibility stack;
since ATK insulates us entirely from the underlying implementation, we
cannot replace it piecemeal. Instead, we're going to remove everything
and then incrementally build on a clean slate:

 - add an "accessible" interface, implemented by GTK objects directly,
   which describe the accessible role and state changes for every UI
   element
 - add an "assistive technology context" to proxy a native accessibility
   API, and assign it to every widget
 - implement the AT context depending on the platform

For more information, see: https://gitlab.gnome.org/GNOME/gtk/-/issues/2833
2020-07-26 20:31:14 +01:00
Benjamin Otte
3078b180fe Replace "gdouble" with "double" 2020-07-25 00:47:36 +02:00
Benjamin Otte
d375dce9f5 Replace "gchar" with "char" 2020-07-25 00:47:36 +02:00
Benjamin Otte
d7266b25ba Replace "gint" with "int" 2020-07-25 00:47:36 +02:00
Matthias Clasen
4786a16696 actionmuxer: Port internal users
Port all internal users of the action muxer
from the GActionGroup interface to the new
action muxer apis.
2020-07-20 08:24:52 -04:00
Timm Bäder
14589c2a59 widget: Save a gtk_css_node_get_style call 2020-07-17 05:49:12 +02:00
Timm Bäder
f678b3804d widget: Remove useless if statement
filter_value is never NULL anyway.
2020-07-17 05:49:12 +02:00
Timm Bäder
f721d2de15 debug: Remove resize highlighting
This has been broken ever since the debug highlighting moved to
inspector overlays.
2020-07-17 05:49:12 +02:00
Matthias Clasen
43b9fc6981 Merge branch 'wip/baedert/for-master' into 'master'
Wip/baedert/for master

See merge request GNOME/gtk!2162
2020-07-01 01:35:56 +00:00
Matthias Clasen
5229069101 widget: Avoid uninitialized access
Initialize all field in the GtkCrossingData struct
when synthesizing crossing events.

Fixes: #2743
2020-06-28 13:42:02 -04:00
Timm Bäder
83543423e2 widget: Don't get the frame clock if we don't use it 2020-06-27 11:47:42 +02:00
Matthias Clasen
913cd91df0 widget: Fix an oversight
g_message comes with a builtin newline, don't add one
to the message.
2020-06-26 22:12:14 -04:00
Matthias Clasen
a26865e741 widget: Add a debug message for consumed key events
Run the application with GTK_DEBUG=keybindings to
get some hints where key events get lost.
2020-06-26 14:08:14 -04:00
Carlos Garnacho
8e549d73a0 gtkwidget: Drop _gtk_widget_[gs]et_shadowed()
This API is meaningless now, with grab-notify being handled internally.
2020-06-24 20:35:23 +02:00
Carlos Garnacho
102d2a36f3 gtkwidget: Avoid use of shadowed flag
This is going away, look up whether there is a grab, and whether
the widget is within it, instead.
2020-06-24 20:35:23 +02:00
Carlos Garnacho
f046e27bb4 gtkwidget: Drop gtk_widget_device_is_shadowed()
A widget no longer needs updating on GTK grabs on itself, drop
this miscellaneous and no longer used API.
2020-06-24 20:35:23 +02:00
Carlos Garnacho
1c5cf17f1c gtkwidget: Drop internal checks in grab-notify
There's no need to check if the widget is already shadowed, we
implicitly have that from the signal argument.
2020-06-24 20:35:23 +02:00
Carlos Garnacho
994f32704b gtkwidget: Drop ::grab-notify signal and vfunc
We don't expect widgets to react directly to these anymore.
Internally we still reset the controllers.
2020-06-24 20:35:23 +02:00
Carlos Garnacho
f401b05e7b gtkwidget: Avoid poking surfaces and events in gtk_widget_list_devices()
This got stuck in ancient times when widgets were windowed, so the devices
in a window to know the devices in that widget would pan out. We do only
want here the devices that are inside the widget, not spread over the
surface, so rewrite this helper function to poke the toplevel foci, and
look they are contained inside the widget.
2020-06-24 11:14:52 +02:00
Carlos Garnacho
42ed8a2748 gtkwidget: Drop _gtk_widget_get_device_surface()
A widget cannot have multiple surfaces anymore, nor needs to
fetch the last event surface as it ought to handle its own.
2020-06-24 10:47:15 +02:00
Carlos Garnacho
29d054c421 gtkwidget: Remove dead code
This variable got unused over time.
2020-06-24 10:47:15 +02:00
Carlos Garnacho
59a8106373 gdk: Drop supports_multidevice API
Crossing events are now detached from widget state, all tricky consequences
from getting multiple crossing events are now somewhat moot. Resort to sending
all generated crossing events, and drop this barely (ever?) used API.
2020-06-23 23:42:53 +02:00
Carlos Garnacho
4fe608e423 gtkwidget: Cancel all gestures in other groups
When a gesture (group) claims a sequence, all other gesture groups
in the same widget should get cancelled. Not just previously claimed
ones, that shouldn't happen actually.
2020-06-23 23:42:53 +02:00
Matthias Clasen
692cc42e10 Add gtk_widget_prepend_controller
Add a variant of gtk_widget_add_controller that
inserts the controller at the beginning, instead
of the end. This will be used in entry completion
to make sure the entry completion key event handling
supersedes the entry one while the popup is open.

Keep this private for now, until we determine if
it needs to be public api.
2020-06-22 12:47:27 -04:00
Matthias Clasen
9faaa5e8be Fix change notification for event controllers
We were adding event controllers at the end, but
announcing a change at the beginning, in
gtk_widget_add_controller. Fix that by emitting
::items-changed for the position where we actually
inserted the controller.
2020-06-22 12:46:12 -04:00
Timm Bäder
424e46849a widget: Don't leak css transform 2020-06-19 05:26:25 +02:00
Timm Bäder
0b13e68bca Avoid calling g_list_model_get_n_items() all the time
These are used in hot paths, e.g. widget construction.
2020-06-19 05:26:25 +02:00
Timm Bäder
1e4dc90d62 widget: don't queue a redraw on css transform changes 2020-06-19 05:26:24 +02:00
Timm Bäder
203bb42cc1 widget: Re-shuffle preconditions in snapshot_child()
The GtkNative type check is more expensive than the mapped one, so avoid
the former for unmapped widgets
2020-06-19 05:26:24 +02:00
Timm Bäder
461497e8d4 widget: Avoid doing GtkSnapshot work just to append one node
We already know that a widget will have literally 1 node, not more.
Avoid doing the GtkSnapshot state stack dance and just append a new
transform node instead.

Seems to give me around 400 more icons in the fishbowl
2020-06-19 05:26:24 +02:00
Emmanuele Bassi
1c856a208f Rename master and slave device
We already use the "logical/virtual" and "physical" names in the
documentation, there's no reason to use loaded terms just because X11
uses them.
2020-06-18 19:22:20 +01:00
Matthias Clasen
37633985bd Drop gtknativeprivate.h
No need to ship empty headers.
2020-06-07 08:52:45 -04:00
Matthias Clasen
3412f0554f gtk: Documentation fixups
Document missing symbols, and other cleanups.
2020-06-05 22:50:30 -04:00
Emmanuele Bassi
51fc382582 Notify accessibility state changes from GtkWidget
Piggybacking on blanket GObject::notify is not very efficient.
2020-06-05 20:35:10 +01:00
Emmanuele Bassi
6e52ef96a4 Move orientable style classes into GtkWidget
It feels slightly wrong to have GtkOrientable operate on widgets, but at
least what happens when an orientable widget changes orientation should
be part of GtkWidget.

This will allow to add more state changes without accessing widget state
from the outside of gtkwidget.c.
2020-06-05 20:35:09 +01:00
Emmanuele Bassi
9606c1eee9 Notify mapped state to the accessible object directly
Do not use signal connections: if we have an accessible object, notify
it directly when getting mapped and unmapped.
2020-06-05 20:32:27 +01:00
Emmanuele Bassi
959c101400 Drop unused AtkRegistry path
We expect widgets to use their own derived GtkWidgetAccessible type,
these days, and given that we hard code the default accessible type of a
GtkWidget to GtkWidgetAccessible, and that we enforce the dependency of
the type passed to gtk_widget_class_set_accessible_type(), the registry
code path is clearly unused.
2020-06-05 20:32:27 +01:00
Emmanuele Bassi
6b096e5c5b Make tooltip properties idiomatic
The tooltip handling in GtkWidget is "special":

 - the string is stored inside the qdata instead of the private
   instance data
 - the accessors call g_object_set() and g_object_get(), and the
   logic is all inside the property implementation, instead of
   being the other way around
 - the getters return a copy of the string
 - the setters don't really notify all the involved properties

The GtkWidgetAccessible uses the (escaped) tooltip text as a source for
the accessible object description, which means it has to store the
tooltip inside the object qdata, and update its copy at construction and
property notification time.

We can simplify this whole circus by making the tooltip properties (text
and markup) more idiomatic:

 - notify all side-effect properties
 - return a constant string from the getter
 - if tooltip-text is set:
   - store the text as is
   - escape the markup and store it separately for the markup getter
 - if tooltip-markup is set:
   - store the markup as is
   - parse the markup and store it separately for the text getter

The part of the testtooltips interactive test that checks that the
getters are doing the right thing is now part of the gtk testsuite, so
we ensure we don't regress in behaviour.
2020-06-05 20:32:26 +01:00
Alexander Larsson
d3e0a1f68c Avoid quadratic slowdown in gtk_widget_add()
If you add a widget to a parent, this will invalidate the css nodes
for parent/siblings. Afterwards, if the parent is mapped, we will
realize the new child. This calls gtk_widget_update_alpha() which
needs the css opacity, so it revalidates the css.

Thus, for each widget_add (while visible) will trigger a full
revalidation of each sibling. If you add N children to a parent that
leads to O(N^2) revalidations.

To demo this I changed gtk-demo to always double the count
(independent of the fps) and print the time it took. Here is the
results (after a bit):

Setting fishbowl count=256 took 3,4 msec
Setting fishbowl count=512 took 10,1 msec
Setting fishbowl count=1024 took 34,1 msec
Setting fishbowl count=2048 took 126,3 msec
Setting fishbowl count=4096 took 480,3 msec
Setting fishbowl count=8192 took 1892,7 msec
Setting fishbowl count=16384 took 7751,0 msec
Setting fishbowl count=32768 took 38097,7 msec
Setting fishbowl count=65536 took 191987,7 msec

To fix this we drop gtk_widget_update_alpha() and just
calculate it when needed (which is only in a single place).
It was really only necessary because we previously set
the alpha on the surface.

With this fix the above becomes:

Setting fishbowl count=256 took 1,0 msec
Setting fishbowl count=512 took 1,9 msec
Setting fishbowl count=1024 took 3,7 msec
Setting fishbowl count=2048 took 7,4 msec
Setting fishbowl count=4096 took 18,1 msec
Setting fishbowl count=8192 took 31,0 msec
Setting fishbowl count=16384 took 66,3 msec
Setting fishbowl count=32768 took 126,7 msec
Setting fishbowl count=65536 took 244,6 msec
Setting fishbowl count=131072 took 492,2 msec
Setting fishbowl count=262144 took 984,3 msec
2020-06-04 16:42:59 +02:00
Matthias Clasen
6652e14662 Update action muxer hierarchy propertly
When a new muxer appears, we must not only make it
inherit from the parent, we also have to update all
children to inherit from it.
2020-06-01 13:52:42 -04:00
Christoph Reiter
2d5cf2b4c4 Drop fallback-c89.c
We require a C compiler supporting C99 now. The main purpose of
these fallbacks was for MSVC. From what I can see this is now all supported
by MSVC 2015+ anyway.

The only other change this includes is to replace isnanf() with the
(type infering) C99 isnan() macro, because MSVC doesn't provide isnanf().
2020-05-31 17:09:23 +02:00
Benjamin Otte
34c6ef5332 widget: Add a hook for resizes
It's private, no APIs, we don't talk about it. But we will start using
it very soon, so we can do size request caching in columns and avoid
sizegroups...
2020-05-30 19:26:46 -04:00
Benjamin Otte
db3e225f09 builder: Make gtk_builder_extend_with_template() work with objects
This will be relevant later when we introduce GtkListItem which is not a
GtkWidget.
2020-05-30 19:26:46 -04:00
Benjamin Otte
7831980c1d listlistmodel: Add gtk_list_list_model_item_moved()
Use it to fix a case that just said g_warning ("oops").

Apparently I had forgotten the case where a container moved a child
in the widget tree.
2020-05-30 19:26:45 -04:00
Benjamin Otte
f1d6a0a4ed widget: Add gtk_widget_get_size()
A little bit of convenience.
2020-05-30 10:48:02 -04:00
Yuri Chornoivan
01bd4cc4e1 Fix minor typos 2020-05-28 11:00:03 +03:00
Timm Bäder
5b77d3fde3 widget: Remove GtkWindow special case in create_render_node() 2020-05-20 17:06:54 +02:00
Matthias Clasen
75d9310986 Change coordinate translation apis to take doubles
Change gtk_widget_translate_coordinates and
gtk_native_get_surface_transform to operate
on doubles. Update all callers.
2020-05-17 17:17:31 -04:00
Benjamin Otte
b353221185 Merge branch 'wip/otte/monitors' into 'master'
various GDK cleanups

See merge request GNOME/gtk!1920
2020-05-17 16:57:52 +00:00
Matthias Clasen
d70feb9165 Apply surface transform in more places
With these fixes, GtkSwitch now works in the presence
of surface transforms.
2020-05-17 02:23:23 -04:00
Matthias Clasen
c90078fd4a widget: Fix gtk_widget_get_surface_allocation
As the name implies, this should be surface-relative
coordinates, so it needs to take the surface transform
into account.
2020-05-17 02:04:22 -04:00
Benjamin Otte
4c7914dc49 display: Remove unneeded getters
Applications can use the listmodel instead.
2020-05-17 07:32:37 +02:00
Benjamin Otte
0c6266fd1a surface: Remove gdk_surface_is_viewable()
It returns the same value as gdk_surface_get_mapped(), so use that
instead.
2020-05-17 00:41:44 +02:00
Matthias Clasen
8912a6eb75 gtk: Handle seatless displays
If you run weston with the headless backend, you get a Wayland
display with no seat, which is just fine by the protocol.

gdk_display_get_default_seat() returns NULL in this case. Various
widgets assume that we always have a seat with a keyboard and a
pointer, since that is what X guarantees. Make things survive
without that, so we can run the testsuite under a headless
Wayland compositor.
2020-05-15 14:11:53 -04:00
Corentin Noël
80473ccd69 widget: Add array annotation to set_css_classes
It expects a NULL-terminated array so annotate it.
2020-05-15 12:01:56 +02:00
Timm Bäder
8a7fb4b133 widget: Make 1 if out of 2 ifs 2020-05-13 11:52:07 +02:00
Timm Bäder
167c6119eb widget: Move adjusted size calculation below early-out goto 2020-05-13 08:07:55 +02:00
Timm Bäder
ef2af521b1 widget: Remove size allocate signal enum member 2020-05-13 08:07:55 +02:00
Matthias Clasen
2a24b8c653 Replace most remaining uses of container api
These are all on GtkBox or enumerating children.
2020-05-11 22:38:21 -04:00
Matthias Clasen
cc28a7b0ef Merge branch 'wip/tintou/doc-fixes' into 'master'
docs: Fix several missing references in the documentation

See merge request GNOME/gtk!1874
2020-05-11 18:40:34 +00:00
Corentin Noël
076b2f11d2 docs: Fix several missing references in the documentation
This fixes several typos and missing references
2020-05-11 19:26:20 +02:00
Matthias Clasen
6d969d1026 Update the docs
Remove various references to gtk_widget_destroy in the docs.
2020-05-11 12:20:59 -04:00
Matthias Clasen
0f10b170f4 widget: Drop gtk_widget_destroy 2020-05-11 12:20:59 -04:00
Matthias Clasen
573b20a8f8 Drop gtk_widget_destroyed
Use g_object_add_weak_pointer or a weak reference instead.
2020-05-11 12:19:39 -04:00
Matthias Clasen
329994291a Merge branch 'wip/baedert/for-master' into 'master'
Wip/baedert/for master

See merge request GNOME/gtk!1867
2020-05-11 14:24:45 +00:00
Matthias Clasen
6372a33838 Clarify docs around focus. 2020-05-11 00:29:41 -04:00
Matthias Clasen
b93631164f Remove lots of focus vfuncs
In many cases, the default widget vfuncs work just fine,
combined with setting focusable.
2020-05-11 00:11:37 -04:00
Matthias Clasen
46bad5d513 widget: Fix gtk_widget_focus_all for !focusable
This was a case that wasn't handle properly when
the focus is entering from the outside, in forward
direction.
2020-05-11 00:11:04 -04:00
Matthias Clasen
edc31a264c widget: Add a :focusable property
Add back a property that determines whether an individual
widget will accept focus or not. :can-focus prevents the
focus from ever entering the entire widget hierarchy
below a widget, and :focusable just determines if grabbing
the focus to the widget itself will succeed.

See #2686
2020-05-10 23:24:48 -04:00
Matthias Clasen
eae4a194b8 Merge branch 'wip/exalm/headerbar-dragging' into 'master'
Extract GtkWindow dragging and titlebar actions

Closes #2689

See merge request GNOME/gtk!1814
2020-05-10 18:15:57 +00:00
Timm Bäder
2455978c78 widget: Move {dis,}connect_frame_clock into their only caller
Makes sense to have these only in (un)realize.
2020-05-10 08:44:20 +02:00
Matthias Clasen
0cd0ef2d26 widget: Don't leak a reference when reordering
When reordering a widget that is alrady under the
same parent, we must not take another reference.
2020-05-09 19:27:56 -04:00
Alexander Mikhaylenko
6c636642df widget: Pass parent to gtk_widget_consumes_motion()
Right now it assumes that the parent is a GtkWindow. This will break when
window dragging is handled by another widget.
2020-05-07 09:44:15 +05:00
Matthias Clasen
57d3719583 Fix an inspector crash
Fix the listmodel semantics of the controller model,
by making get_item return a reference.
This was broken in 4b9e5304cc.
2020-05-06 22:25:51 -04:00
Matthias Clasen
56e590bd31 widget: Drop the ::size-allocate signal
This signal is mainly used for bad things, and we
don't want to emit signals during layout if we can
avoid it.

If you are subclassing, you should either use a
layout manager or override the size_allocate vfunc.

If you are using a GtkDrawingArea or GtkGLArea,
use their ::resize signals to learn about size
changes.

Fixes: #2705
2020-05-06 14:27:45 -04:00
Emmanuele Bassi
cb94c7d69e Explicitly update widget a11y bounds when allocating
The a11y machinery is using signal subscription to get notified of size
changes and notify listeners in turn. This is suboptimal for a couple of
reasons:

 - if something connects to the GtkWidget::size-allocate signal we need
   to emit it; currently, we have an optimization in place that will
   skip the signal emission if there are no handlers, and it would be
   nice to go through the fast path
 - the accessibility implementation is part of GTK, and should not go
   through additional hoops like any out-of-tree API consumer
2020-05-06 14:27:45 -04:00