The former can be called individually on each sequence, and the latter will
always call the former on all currently active sequences, so only implementing
resetting on cancel() works for both cases. Also, chain up on subclasses
implementing cancel.
This fixes clicking on nautilus' file list after popping up a menu, as broken
grabs are one of those situations where sequences get cancelled individually,
the "current button" wasn't properly reset, and further clicks with button != 3
were ignored.
This used to be done before the gestures port, and was removed
accidentally, so keep the motion_notify_event handler just for
this, and fallback to having those events handled by gestures
too.
This way plain clicks can be handled in gtkmain through the usual delivery mechanism,
and get possibly handled too by widgets holding a GTK+ grab. If window dragging is to
be started, the sequence will be claimed (and a grab will happen afterwards), notifying
properly the grabbing widget that event delivery was interrupted.
This makes it possible to dismiss popovers by clicking on window headerbars, while
still making it possible to drag the window with the popover opened.
Weston numbers its touch sequences ids starting from 0, thus simply
setting the GtkEvents touch.sequence to the touch id value typically
causes gdk_event_get_event_sequence to return NULL. Unfortunately this
confuses other parts of GDK.
As both weston & mutter keep the sequence id between 0..max_dev_touches
-1 simply use + 1 to keep the id > 0. While this isn't entirely correct
(compositor could send -1 as the touch id), this keeps the touch id in
gtk tied to the touch id from weston which is useful for debugging. A
more thorough solution could be done when it turns out this is an issue
in practise
https://bugzilla.gnome.org/show_bug.cgi?id=731371
The non-zero default default-border was causing buttons to shrink as
the focus moves around them. Themes which want a default-border should
define it explicitly.
The code is actually prepared for that, the gesture was initially limited
to only handling GDK_BUTTON_PRIMARY because it only used to handle row
activation.
This gesture acts only on events from the bin window, and checks that
either the pressed row is draggable, or the conditions for rubberband
selection apply.
A multipress gesture takes care of autosizing on double click, and
a drag gesture is used for both column dragging/resizing (only one
can happen at a time).
When placing tooltips, the csd shadow will get 'pushed up' and
may end up underneath the pointer. We don't want this to cause
the tooltip to be hidden, because that leads to flickering, so
ignore the shadow when finding the widget under the pointer.
For csd override-redirect windows, we don't set up resize handles,
but we were not ignoring the margin in all places, causing some
size calculations to go wrong.
... from per style data to only existing once per style context. This is
technically an API break because it no longer allows getting different
style properties between save()/restore() pairs, but I don't think this
was ever intended to work that way, as the style property API was to be
used and is used via gtk_widget_get_style().
And it simplifies code a lot.
We used to accept the same syntax for text-shadow and icon-shadow as
we accept for box-shadow. However, box-shadow does accept a spread and
the inset keyword while the others should not.
The signal needs to be emitted after the text insertion as at-spi gets
the text to compute the inserted text due to the AtkText::insert-text
signal not containing it.
Also adjust position to reflect changes to the offsets.
https://bugzilla.gnome.org/show_bug.cgi?id=731429
This is old code from dating back many years. Nowadays, we can
just use css drawing and csd windows to achieve much the same
effect.
Themes will need some adjustment for this change.
https://bugzilla.gnome.org/show_bug.cgi?id=731187
This commit makes it possible to use client-side decorations for
override-redirect windows by calling _gtk_window_request_csd()
before realizing the window. Since the wm won't do interactive
resizing for us in this case anyway, don't bother creating
the border windows we use for this purpose on regular toplevels.
To make this accessible to themes, we set a "csd" style class
on client-side decorated windows. With this, .window-frame.csd.menu
can be used to define the shadow for csd menus, and .menu can be
used to define a border for menus under non-composited wms.
https://bugzilla.gnome.org/show_bug.cgi?id=731187
This tests just a few basic things for now. Mainly, that we don't
emit redundant notifications for enum, flags, int and boolean
properties. It also checks that we do emit the expected notifications
when the value actually changes. This is checked for string, double
and float properties as well.
There is a large number of exceptions in the test, and a lot more
checks that could be done. One class of exceptions is all the places
where we have -set booleans to go along with another property. We
should have a dedicated test for these pairs. Another class of
exceptions is where naked objects created by g_object_new () just
don't have the full functionality - e.g. a tree selection without
a tree view does not work very well. We set up the instance object
better for these situations.
A few properties here are special, and can't benefit from it:
those which are just shorthands, like ::margin and ::expand,
and those that have explicit -set properties, like::hexpand
and ::vexpand.
This one is a little tricky, since we override this property
in many places, and you cannot add flags when overriding. So,
all places where this is overridden will have to make sure to
notify explicitly.
When we expanded the GtkLicense enumeration in 3.12, we forgot
to update the limit check in gtk_about_dialog_set_license_type.
Caught by testing property notification for enum properties.
The reftest is testing "transparent" works as expected by drawing a
purple background once with purple and once with transparent and
expecting the same result. This works fine unless anti-aliasing happens
at rounded corners. The overdraw of the 2nd background changes the
antialiased pixels.
Fix this by explicitly setting the border radius to 0.
Also reindent the file to make it more readable.
The a11y headers are now listed with the main Makefile.am of GTK, and
GDK introduced deprecated headers, so we need to account for them for the
build of the introspection files.
Currently, due to the way that Visual Studio 2010+ projects are handled,
the "install" project does not re-build upon changes to the sources, as it
does not believe that its dependencies have changed, although the changed
sources are automatically recompiled. This means that if a part or more
of the solution does not build, or if the sources need some other fixes
or enhancements, the up-to-date build is not copied automatically, which
can be misleading.
Improve on the situation by forcing the "install" project to trigger its
rebuild, so that the updated binaries can be copied. This does trigger an
MSBuild warning, but having that warning is way better than not having an
up-to-date build, especially during testing and development.
Otherwise, it is pretty hard to keep track of what
classes are missing from the list. As a consequence,
several of the existing classes were missing. Add those
at the same time.
We were applying response based heuristics, even if the button
is explicitly put in the headerbar. That broke button placement
in some epiphany dialogs, such as the Cookies one. Therefore,
restrict the heuristics to action widgets that are added through
gtk_widget_add_action_widget() or <child type="action">, where it
is not possible to specify placement explicitly.
There are plans to add session-dependent defaults to GSettings
(based on the newly standardized XDG_CURRENT_DESKTOP); until
then, the WM uses a different schema for its button-layout
setting in classic mode. So for the time being, do the same
and pick the alternative schema when XDG_CURRENT_DESKTOP
indicates that we are in a classic session.
(It's not pretty, but hopefully won't be with us for too long ...)
https://bugzilla.gnome.org/show_bug.cgi?id=731273
Pick up the setting from the org.gnome.desktop.wm.preferences schema
if available. It is slightly more involved than other settings, as
the actual button names used in the schema differ from the ones we
use, so we need an additional translation step.
https://bugzilla.gnome.org/show_bug.cgi?id=731273
When combining action child type with <action-widgets> to assign
response ids, we were not properly updating buttons that ended up
getting added to the headerbar before they have a response id.
Fix this by reapplying the headerbar button setup after parsing
<action-widgets>, and make sure to also update the suggested-action
style class.
If called when already popped down, warnings would be issued due
to priv->grab_pointer being unexpectedly NULL, this would happen
in regular operation when selecting items in appears-as-list mode.
So both add a NULL check for priv->grab_pointer, and bail out early
if the popup window is already hidden.
GtkKineticScrolling implements the actual physics laws for friction
and springs. When created, position/velocity/boundaries/constants are
given, so at every gtk_kinetic_scrolling_tick() it returns the current
position, and whether the system is in rest.
https://bugzilla.gnome.org/show_bug.cgi?id=729608
A multipress gesture takes care of link handling, and char/word/all
selection mode on selectable labels. A drag gesture is used for both
text selection and DnD checks on selectable labels.
A multipress gesture takes care of clicks, and where those happened.
If the click is meant to move the slider while pressed, the drag gesture
takes over, dragging the slider while active.
Widgets becoming insensitive won't receive further events, but there
could be chances the controllers don't get properly notified and reset
in those situations.
The signal handler for the visibility of menu items changing had the
wrong signature, resulting in the GParamSpec from the notify signal
being treated as a boolean (which was always true). This resulted in
items being added over and over and never being removed.
Fix that...
Since commit 872fbfac the GtkWindowGroup was split out from the GtkWindow
sources, so include gtkwindowgroup.h to avoid C4013 warnings/errors
(implicit declaration of ...) on gtk_window_group_add_window().
We were setting the next-tab properly, but were trying to
read it off the wrong object. Now, going from a cell renderer
attribute mapping to the model, or from an action-name to the
action actually works.
The touch_event handler was missing those when emulating pointer events
for the widgets that get GDK_TOUCH_MASK set, but have no specialized
touch handlers.
This code is a product of early stages in the gestures branch, where
capturing would have an effect outside grab boundaries. But this isn't
really the case, so every gesture outside the grab scope must be reset
to avoid keeping stale data.
This tests both a sequence being claimed early to be then denied
(and handled deeper in propagation chain), and a sequence being
claimed late in the capture phase (and thus being cancelled deeper
in the propagation chain)
Before this change, a sequence being claimed deep in the event propagation
chain would make the sequence go denied on every ancestor, regardless of
previous state.
To make things more consistent, only deny the sequence if it was previously
claimed, so the behavior is the same for gesture groups within the widget
than for those outside the widget.
The gestures testsuite has been updated to reflect this new behavior.
It might happen that a gesture claims a sequence before any other gesture
in its group even handled a single event from that sequence. In that case,
ensure the state is set accordingly right when the sequence is handled in
those.
The "group" gesture testcase has been updated to observe this behavior.
The CSS editor was feeling a little sluggish, because it was
reparsing and reapplying the CSS on every keystroke. Add a small
delay, to make this feel smoother.
Like the GDK and GTK portions, use autotools scripts to generate the
complete projects for gtk-inspector as sources there seem to change from
time to time.
It might be so that this, like the a11y sources, will be referenced from
the main Makefile.am of GTK directly, but just do this so that the
projects can build properly.
This test check that resizing the window when expanding
the expander yields the same end result as having the
expander expanded to begin with. The test uses the inhibit
mechanism introduced in the previous commit.
This adds an inhibit api that code from the reftest module
can use to delay the taking of the snapshot. Also refactor
the code in gtk-reftest to use the inhibit mechanism for
its own delaying of the snapshot until after the first
expose.
Every button press/release event reaching the the multipress gesture in GtkWindow
and happening in the "title" region must be handled, regardless of the event widget.
Children there wanting the event(s) for themselves are (and were always) expected
to stop event propagation.
So the only place to check for the event widget's "window-dragging" style property
is the "content" region, which matches the pre-gestures behavior.
This fixes some issues with sequences being mistakenly claimed (and events not
propagated further) on situations it shouldn't.
The multipress gesture must react to either direct events on the
GtkWindow (special cased through _gtk_widget_check_handle_wm_event),
or bubbled events from child widgets. Ensure bubbled events go
through the gesture, those are fed manually to make sure events are
only handled once, in either one or other place. The implicit grab
will ensure that doesn't change mid-action.
Otherwise the event is possibly handled, but still propagated further anyway.
Ensure the event is consumed by claiming the current sequence on the
GtkGestureMultiPress::pressed handler.
::row-activated only used to be triggered by GDK_BUTTON_PRIMARY, so make
the multipress gesture handling this now to be only triggered by that same
button.
https://bugzilla.gnome.org/show_bug.cgi?id=731020
This gesture was only meant to react on GDK_BUTTON_PRIMARY (either
through real pointer events, or implicitly assumed from touch events),
as it used to behave before gestures. Otherwise the gtk_drag_begin*()
call assumes being triggered by button 1, and the drag misbehaves
because that button isn't really in the state mask.
https://bugzilla.gnome.org/show_bug.cgi?id=731016
When going from attribute mapping to model, it makes most sense
to go directly to the data tab, and when going from an action
name to the owner, we want to show the actions tab. Make it so.
This reftest makes use of the new feature to add signal handlers.
It adds a libreftest.so module containing all the code for the reftests.
When adding a test named reftest.ui, please keep code contained in a
source file names reftest.c and add that file to Makefile.am.
https://bugzilla.gnome.org/show_bug.cgi?id=730833
When connecting signal names, gtk-reftest now allows you to use a colon
in the signal handler name like so:
module:function_name
where module is a module loaded from the same directory (or the .libs
subdirectory for compatibility with uninstalled libtool) as the running
test and the function is resolved in that module. Of course, normal
function names work as before.
The extra condition here that caused the current child to
not redraw during reordering was introduced in f383e1f1
during the port to ::draw, but was not explained in the
commit message, and removing it has no obvious negative
effect.
https://bugzilla.gnome.org/show_bug.cgi?id=730767
Just calling get_type() does not ensure that the signals, properties
and everything else gets set up properly. Ensure it is, by calling
g_type_class_ref() before using the type. This fixes the testcase
added in the previous commit.
Those might trigger the destruction of some widget that would dispose the
event controller while the event is still being handled, so keep an extra
ref on the controller during event processing.
Without this information introspection-based consumers don't realize
they can include context information, but instead think that they
receive an extra gpointer argument (which they don't know how to
handle).
https://bugzilla.gnome.org/show_bug.cgi?id=730745
This will hopefully help resolve the circular dependency between
libgtk linking against inspector/libgtkinspector and inspector/
needing extract-strings from gtk/.
I didn't preserve the EXEEXT decorations in this operation -
automake gave me stern warnings about it, so I just dropped them
all. Somebody who cross-builds GTK+ will have to reconstruct this.
This prevents some of our generic object implementation tests
from working with gesture objects. Instead, add g_return_if_fail
checks in all the gesture constructors.
Mainly doing s/TARGET/BUBBLE/ on the fully ported widgets, but GtkTreeView
where the double click handler has moved to GTK_PHASE_TARGET so it runs
parallelly to the still existing event handlers.
Previously, there would be globally just a capture and a bubble phase,
with the event just going down the hierarchy once, and the up once.
GTK_PHASE_TARGET actually meaning "run within event handlers", so in
a hierarchy of 3 widgets, emission would be:
Capture(C)
Capture(B)
Capture(A)
Target(A) (if event handlers allow)
Bubble(A)
Target(B) (if event handlers allow)
Bubble(B)
Target(C) (if event handlers allow)
Bubble(C)
This commit changes this behavior and uses GTK_PHASE_TARGET in a less
misleading way, running only on the widget that was meant to receive
the event. And GTK_PHASE_BUBBLE has taken over the execution place of
GTK_PHASE_TARGET, so the emission remains:
Capture(C)
Capture(B)
Capture(A)
Target(A)
Bubble(A) (if event handlers allow)
Bubble(B) (...)
Bubble(C) (...)
As it was, GTK_PHASE_BUBBLE was useful for running event controllers
paralelly to event handlers, without modifying a single line in those.
For those mixed scenarios, Any of the other phases will have to be
used at discretion, or the event handlers eventually changed to chain
up and let the default event handlers in GtkWidget to be run.
The events to those are fed outside the regular event propagation scheme,
through _gtk_window_check_handle_wm_event(), so set the controller to
GTK_PHASE_NONE so events aren't processed first manually, and then
automatically.
Event controllers now auto-attach, and the GtkCapturePhase only determines
when are events dispatched, but all controllers are managed by the widget wrt
grabs.
All callers have been updated.
Go back to respecting GtkMisc::xpad/ypad. Not doing so breaks
the misc-alignment reftest. As long as we still derive from
GtkMisc, we may as well do this.
The icontheme lookup code has a special-case that prefers builtin icons
if the theme name is "hicolor". This is problematic for our reftests,
which run in a barebones environment with not settings.
Drop the ref on the action muxer in finalize, and also make sure
shutdown() tears down the muxer setup done in startup().
When GtkApplication adds itself to a muxer, it causes the muxer to take
a ref on the GtkApplication. This has to be undone in shutdown() to make
sure the GtkApplication doesn't end up holding a ref on itself.
https://bugzilla.gnome.org/show_bug.cgi?id=730383
And handle the fact that drawing bounds are now handled by this API and
the corresponding gtk_widget_get_clip().
Also add _gtk_widget_supports_clip() function to check if a widget has
been ported to the new world.
Now that labels and images no longer use x/yalign in their code,
we need to make gtk_misc_set_alignment set the h/valign for these
widgets, to keep it stumbling along until its final demise.
https://bugzilla.gnome.org/show_bug.cgi?id=730613
Those get in the middle more than help on these widgets, the widget
is already packed with clickable areas and having handles (and their
invisible clickable area around) hovering above don't help, plus the
purpose in most likely numeric values is a bit doubtful.
All touch events are either consumed by the up/down panels, or
the swipe gesture, all GtkEntry handling of touch events on the text
window is avoided, so handles to not appear anymore.
the "bubble" phase used to run before event handlers before GTK_PHASE_TARGET
was added, in order to keep phases in the expected order, move GTK_PHASE_BUBBLE
to be run (still invariably) after event handlers.
The only behavioral change should be wrt widgets wanting mixed event handler/
gesture handling, they could previously attach the gesture to the bubble phase
and check for gtk_gesture_is_active() in the event handler to bail out, they'll
have to use GTK_PHASE_CAPTURE for that purpose from now on.
The handle is still centered horizontally, but the extra vertical
space wasn't taken into account, leading to misplacing the dragging
point (and the handle) during motion events.
GtkPaned may just capture pointer events because the child widget
doesn't happen to have GDK_TOUCH_MASK set, resort to checking the
device in that case.
Dragging is all handled by a GtkGesturePan now, matching the
paned orientation.
On touch events, a wider area is listened for, so touch events
don't need to be as accurate to initiate dragging, if no dragging
is truly initiated in this case, events are just forwarded for
child widgets to handle.
A pan gesture is used to handle switch dragging, which is only triggered
by horizontal panning movements. A multipress gesture handles the cases
where clicking without dragging happens, just toggling the switch.
A pan gesture is optionally attached if there is only one scrolling direction, the pan
gesture orientation is changed so movements tangential to the scroll direction get
scrolling cancelled (The pan gesture is automatically denied when that happens, and
that state change spreads to the others gestures in the group). If the pan direction
happens in the expected directions, no cancellation happens, and scrolling eventually
takes place.
Multiple calls are supposedly allowed to change the phase (although
unlikely to happen), so remove the g_return_if_fail() checking whether
the controller was already added.
Presses alternatively show and dismiss the popover, the popover is still
always shown invariably after any dragging happens (either text selection,
or dragging a text handle)
Presses alternatively show and dismiss the popover, the popover is still
always shown invariably after any dragging happens (either text selection,
or dragging a text handle)
Similarly to GtkTextView, a GtkGestureMultiPress gesture handles
button/touch presses to initiate one selection mode or other, and
a GtkGestureDrag is used to handle text selection and DnD checks.
The code from button press/release, motion, and grab_notify handlers
has been shuffled into the actions triggered by those gestures.
A GtkGestureDrag is used for color selection, removing also the
need to track the pointer state in widget data. The GDK grab performed
just to set the crosshair cursor has been replaced by a call
to gdk_window_set_device_cursor(), which will be unset if the
drag operation is finished, or cancelled due to the implicit grab
being broken.
When the pointer cursor is updated on CSW, lookup for either a device
cursor, or a global one. It would previously lookup for windows with
a global cursor, and then check if it had a device cursor, which would
skip windows with only device cursors set, and unexpectedly set the
global cursor.
All "exclusive" gestures listen for either pointer events, or
"pointer emulating" touch events, so only a single sequence at
a time can make these run.
This signal will always be paired with a ::pressed signal, unless
the sequence is cancelled, or the controller is reset. the n_press
argument in the signal always matches the ::press signal one, even
if GtkGestureMultiPress::stopped was emitted in between.
The current sequence (as per gtk_gesture_single_get_current_sequence)
is used to find out the coordinates. And only emit ::pressed if the
gesture began through a GDK_BUTTON_PRESS/TOUCH_BEGIN (eg. not due to
an extra touch being lifted)
Just call the controllers on that phase if the default widget handlers
are run.
For compatibility reasons, in the touch event handler, let the pointer
emulating touch be transformed to a pointer event as usual, in order to
have widget handlers a chance to run at all. If they have to be managed
by a controller in that phase, it'll have to be through the default pointer
event handlers.
This phase is meant to run in the default widget handlers, as opposed
to externally as in the bubble/capture phase. This will be most usually
the expected phase for every controller replacing code in event handlers
in GTK+, just so invocation and triggering order is kept unaltered.
That may happen separately from grab-notify, and also due to external
reasons, so ensure all sequences are cancelled if a grab is taken
in some GdkWindows that would obscure events on the controller.
We can end up with _gtk_widget_remove_controller getting called
while we are iterating over the list in _gtk_widget_run_controllers.
To avoid trouble, only mark the event controller as dead by
setting data->controller to NULL, and defer the actual freeing
and list manipulation to the loop in _gtk_widget_run_controllers.
Update other places that operate on controllers to handle
data->controller being NULL.
Make it really sure that the event is only emitted after every gesture
that consumed the button press is done with the sequence.
The event must only be emulated if a gesture in the capture phase happened
to consume the event, be cancelled, and
Sequences may be cancelled within the ::sequence-state-changed handler, which
would change the points hashtable as it's being iterated in this function. So
iterate over a list of sequences and let the hashtable change freely.
The propagation phase property/methods in GtkEventController are gone,
This is now set directly on the GtkWidget add/remove controller API,
which has been made private.
The only public bit now are the new functions gtk_gesture_attach() and
gtk_gesture_detach() that will use the private API underneath.
All callers have been updated.
Within a widget, if a gesture accepts a sequence, it would previously
cancel every other gesture that not in the same group. Change this to
only cancelling gestures that previously claimed the gesture, and let
gestures with state=NONE for that sequence remain like that.
This enables late recognition of gestures, even on the presence of
another gesture group that was more eager at claiming the gesture.
One usecase is user-defined panning gestures on scrolledwindows,
if ::capture-button-press is TRUE (eg. the default), the gesture is
claimed early in order to consume the button press, but that would
tipically make every other gesture group deny the sequence. With
this change, the pan gesture can keep state=NONE, and later claim
the sequence for itself if the panning gesture is recognized.
Also, do not propagate state=DENIED to every gesture in the widget,
that was unintended.
If no match is found with the gesture widget when poking the event
window parents, bail out safely instead of falling in an infinite
loop. This was seen on Mutter.
The utility of those signals is somewhat dubious now that there is
gtk_gesture_group(), so make that the only way to coordinate gestures.
The cooperation model offered by gtk_gesture_group() is flexible
enough,
Listen for notify::sequence-state-changed on the controller, so the
only way to manipulate a sequence state are gtk_gesture_set_sequence_state()
and gtk_gesture_set_state().
Also, make use of gesture groups, so the sequence state is set at once
on all gestures pertaining to a single group. Within a widget, if a sequence
is claimed on one group, it is made to be denied on every other group.
This API eliminates the need for overriding
GtkWidget::sequence-state-changed virtually everywhere. Grouped
gestures share common states for a same GdkEventSequence, so the
state of sequences stay in sync across those.
A multipress gesture is used to control all this, replacing
single/double click custom code, and triggering window dragging
when the multipress is stopped, yet active (ie. the sequence remains
pressed).
This gesture is used by gtk_drag_source_set() to determine
whether dragging moved past the threshold. The gesture events
are handled via the usual ::event callbacks, so we don't mess
up with callers expecting that to happen in a signal handler.
If the sequence gets claimed somewhere else in the event widget
stack, the DnD gesture will be cancelled.
With gtk_gesture_get_point() returning events in
gtk_event_controller_get_widget() coordinates, we no longer need
to compensate for the overshoot here.
GtkEventController may be certainly useful to keep event
handling self-contained in other places than gestures, but
the current widget API is highly related to gestures, so
just using GtkGesture as the argument there will be quite
more convenient. The other places where GtkEventController
make sense as a base object will better provide their own
hooks.
Gestures attached with this phase will expect callers to have it
receive events through gtk_event_controller_handle_event(), but
the gesture will still be notified of sequence state changes,
grabs, etc...
Translate events meant for other widgets/windows, so gtk_gesture_get_point()
always returns coordinates based on the gtk_event_controller_get_widget()
allocation.
If a gesture has denied sequences (so those are presumably handled above/below
the widget), it shouldn't attempt to handle extra touches, even if those end
up matching the expected number of touches.
Gestures should always receive one of such events in order to be activated,
and the propagation mechanism will ensure they do so if the original event
was caught up the widget hierarchy by another gesture that is now declining
the sequence.
If the captured touch begin or button press event have been consumed
for the given sequence, propagate it upwards if the sequence goes from
claimed to denied, so the widgets on the way to the event widget receive
a coherent event stream now that they're going to receive events.
The policy of sequence states has been made tighter on GtkGesture,
so gestures can never return to a "none" state, nor get out of a
"denied" state, a "claimed" sequence can go "denied" though.
The helper API at the widget level will first emit
GtkWidget::sequence-state-changed on the called widget, and then
notify through the same signal to every other widget in the captured
event chain. So the effect of that signal is twofold, on one hand
it lets the original widget set the state on its attached controllers,
and on the other hand it lets the other widgets freely adapt to the
sequence state changing elsewhere in the event widget chain.
By default, that signal updates every controller on the first usecase,
and propagates the default gesture policy to every other widget in the
chain on the second. This means that, by default:
1) Sequences start out on the "none" state, and get propagated through
all the event widget chain.
2) If a widget in the chain denies the sequence, all other widgets are
unaffected.
3) If a widget in the chain claims the sequence, then:
3.1) Every widget below the claiming widget (ie. towards the event widget)
will get the sequence cancelled.
3.2) Every widget above the claiming widget that had the sequence as "none"
will remain as such, if it was claimed it will go denied, but that should
rarely happen.
This behavior can be tweaked through the GtkWidget::sequence-state-changed and
GtkGesture::event-handled vmethods, although this should be very rarely done.
The kinetic scrolling feature is now implemented using a
GtkGestureDrag and a GtkGestureSwipe, plus a GtkGestureLongPress
in order to denying the sequence, so it is possibly handled
underneath.
A controller can be optionally hooked on the capture or the bubble
phase, so the controller will automatically receive and handle events
as they arrive without further interaction.
Now, all captured events run from the toplevel to the deepmost widget,
regardless of GTK+ grabs. This makes captured events more useful to
event controllers if used together in the hierarchy with widgets doing
old fashioned event handling and GTK+ grabs.
This gesture handles any number of clicks, ensuring multiple presses
stay within thresholds and timeouts. When anything of that happens,
the gesture is reset and press count starts from 1 again.
Optionally, the gesture can be given a rectangle, used in in presses > 1
to ensure the consecutive presses happen on user imposed areas.
This gesture implementation recognices swipes on any direction.
The "swipe" signal has the X/Y velocity vector components, so
those can be used for direction guessing and velocity thresholds.
This test just checks that all the icon names that GTK uses are present
in the default icon theme.
As icon names are not checked programmatically and we do not want to run
into missing-icon icons in the code, this test seems necessary.
For now, it's just a stub that tests stock icons.
The change to take out unneeded NULL checks requires some care
at startup: we check both adjustments when any of them changes;
we need to do those checks in the same order in which we create
the scrollbars, otherwise we'll try to get the adjustment of
the vscrollbar when we just set the up the hscrollbars' adjustment.
When the color editor is visible, there is no way for the
application to know about the changes that are happening.
Fix this by emitting property notification for the "rgba"
property.
http://bugzilla.gnome.org/show_bug.cgi?id=708037
This avoids a bunch of policy problems with deciding how to lay
out the window menu under different WMs.
For now, we use the special event _GTK_SHOW_WINDOW_MENU, but we
hope to have this standardized in wm-spec quite soon, as KDE wants
it as well.
To make the icontheme test run successfully when installed,
we need to use the correct test-framework-provided location,
and we need to install the test theme without stripping its
subdirectory structure.
We see an active link when creating the menu, but by the time
the menuitem is activated, we've received a leave notify that
makes the label clear its active link. Instead, give the
menuitems a direct reference to the link that is active when
the menu is created.
Problem pointed out by Tim Baedert
We know the objects in a size group are always widgets, so we
can avoid hard-to-track down problems with weak references by
just cleaning up when the object gets destroyed. There is still
a chance that we show a widget as part of the group after it
has been removed, but size groups simply have no signals that
would let us avoid that.
Add a tab that shows size groups of a widget. The properties
of the size group are available here, as well as the widgets
that are part of the size group. We highlight the widgets in
the application when their row in the inspector is hovered.
In contrast to the flashing, where we blink the widget a
few times, this is explicitly turned on and off.
It will be used for indicating widgets that are part of
a size group, in the next commit.
The focus widget might be unset, just to be set again on a widget inside
the popover. Have the popover wait till the focus is actually moved outside
before dismissing.
Move away from cell editing, and use a popover instead. This makes
it easier to e.g. use a color chooser - there's just not enough room
in a cell for many things.
Much of this code is adapted from tests/prop-editor.c.
The child properties in GtkAssistant are somewhat broken, since
they are not on direct children - but that is no reason to crash
if somebody does ask for child properties of direct children.
When a model is sortable, but the the column is not currently used
for sorting, we want to reserve the space for showing the sort
indicator. But we currently set the icon to 'missing-image', which
is not great to show all over the place. So, just set the opacity
to 0.
With the keybinding, it is possible that users may trigger the
inspector unintentionally. Show a dialog that informs them about
whats going on and gives them a chance to back out.
The warning dialog can be bypassed with the
org.gtk.Settings.Debug inspector-warning setting.
All the globals we care about should appear before doing anything
else, up-front, so a single round-trip after adding the registry
should be more than enough.
Update visual.c to use Windows themes rather than the stock Raleigh theme,
and avoid hardcoding data paths for Windows (and Mac). As the dlfcn.h
functions are only used when Python is enabled, move its inclusion there[*].
Also ensure that variables are declared on the top of the block.
[*] Python support Windows needs to be investigated, as POSIX signal
handling is used there.
https://bugzilla.gnome.org/show_bug.cgi?id=730236
Add project files to build the GtkInspector sources, as gtk-inspector is a
required portion for GTK+. "Install" the
org.gtk.Settings.Debug.gschema.xml gsettings schema file as well, so that
people can trigger GtkInspector as they develop and test their GTK+-based
programs.
Since you can't take grabs on unmapped windows, GtkMenu takes a grab on
the menu in a convoluted way: it first grabs another window, shows the
menu window, and then transfers the grab over to the GtkMenu widget.
For normal menubars, this is perfectly fine, as the first window it grabs
is our toplevel, and that gets picked up in our transient path. For
GtkMenuButton or other spurious uses of gtk_menu_popup, it creates a new
temporary input-only window which it takes the grab on, known as the "grab
transfer window". Since this window isn't a transient-for of our new menu
widget window, the grab isn't noticed when we go to show it, and thus the
menu ends up as a new toplevel.
Add a special hack to GtkMenu and the Wayland backend which lets us notice
this "grab transfer window", and include it in our grab finding path.
It's sort of terrible to have to hack up the widgets instead of just the
backend, but the alternative would be an entirely new window type which is
managed correctly by GDK. I don't want to write that.
The entire UI is constructed with templates, so the wrapper
constructors are never called, except for gtk_inspector_window_new,
which gets called from the GTK+ code.
Show the actions that are added to GtkApplication and
GtkApplicationWindows, as well as action groups that are
inserted elsewhere with gtk_widget_insert_action_group.
https://bugzilla.gnome.org/show_bug.cgi?id=730095
Moving the inspector into libgtk lets use reuse internals without
having to add public API for everything or inventing awkward private
call conventions.
https://bugzilla.gnome.org/show_bug.cgi?id=730095
Add two new icon lookup flags, GTK_ICON_LOOKUP_DIR_LTR and _RTL,
which tell GtkIconTheme to look for icon variants which have a
-ltr or -rtl suffix. GtkIconHelper adds these lookup flags when
looking up icons.
Note that due to the way this overlaps with symbolic icon lookup,
directional variants of symbolic icons must be called -symbolic-rtl, not
-rtl-symbolic.
https://bugzilla.gnome.org/show_bug.cgi?id=729980
Provide API to query the matrix instead of API that applies the matrix
directly. This makes the API more flexible.
See the commits implementing shadows.
This test is a bit brittle because it doesn't properly rely on CSS
properties but needs to use widget style properties to turn of extra
sizing from widgets.
It might break in the future when porting widgets to draw properly.
When forcing regular or symbolic icons, fall back to the default
specified icons. This ensures that when no symbolic icon is present, an
icon will still appear - the regular one.
GTK_ICON_LOOKUP_FORCE_REGULAR and GTK_ICON_LOOKUP_FORCE_SYMBOLIC can be
used to force a regular or symbolic icon to be loaded, even if the icon
names specify a different version.
This is intended to support the CSS property -gtk-icon-style.
The values can be:
"requested" - the style as requested
"regular" - use a regular full-color icon
"symbolic" - use a symbolic icon
The property defaults to "requested", so no changes should be seen
unless CSS overrides it.
It is also inherited, so that using this CSS
.toolbar { -gtk-icon-style: symbolic; }
is enough to force the whole toolbar to use symbolic icons.
The value implements the 2D parts of CSS transforms. See
http://www.w3.org/TR/css3-transforms/
For the specification.
All it does is give us an expressive way to define Cairo matrices (and
their transforms)
This allows using icons from the icontheme as images in CSS. The
reasoning is that this allows to give the image control about how it's
scaled (by using the icon theme's scaling method. So we can get crisp
images at different resolutions.
Replace them by GtkWidget h/valign. The only remaining uses
are those where a size group is involved; they can't be replaced
until GtkLabel stops looking at GtkMisc alignment for size
allocation.
Wayland's mechanism tells us all of our new states, rather than
telling us which ones were added and removed. Add a new private
interface so that we can simply specify the new states as a
bitfield directly rather than having to compute which ones were
added and removed.
Make the relative_to widget the parent for a GtkPopover's
GtkActionGroup. This, for example, makes the menu model of a
GtkMenuButton find action groups attached to the button.
https://bugzilla.gnome.org/show_bug.cgi?id=729915
Returns NULL in case of a duplicated tag name in the tag table. It is
still a programmer error to duplicate a name, but if it happens the
behavior is a little nicer (and hopefully doesn't crash).
https://bugzilla.gnome.org/show_bug.cgi?id=614717
The user doesn't need to check the return value, because if FALSE is
returned it is a programmer error. But it permits a nicer behavior for
gtk_text_buffer_create_tag() in case of failure.
https://bugzilla.gnome.org/show_bug.cgi?id=614717
Instead of manually doing it everywhere, just call set_object (NULL)
from the selection_changed handler. Fix all the set_object() functions
to deal with NULL.
Nice to have a quick way of testing this everywhere.
The implementation is not quite perfect: due to the way text
direction works in GTK+, widgets that appear in the inspector
window while we are flipped will inherit the flipped direction
instead of the fixed direction of the inspector window.
The widget-tree was not safe against object just going away.
Fix this by using row references instead of iters where
necessary, and by using weak refs to clean up when objects
die.
It does not really add much over the sensitive/insensitive rendering
that is already indicating which widget is mapped. At the same time,
set up signal handlers so we can update that when the widget changes.
Not only was the property list connecting to notify::bla for
each property individually, it was also leaking the signal
handlers when the selected object changed. Fix both.
When showing the objects in the tree, use the property name
as the name thats shown in the list. This makes it easier
to differentiate e.g. hadjustment and vadjustment in a
GtkScrolledWindow.
Add a tab that shows available signals for each object. For now,
we only show if each signal has handlers connected or now. More
functionality will be added later.
The list of toplevels also includes hidden combobox popups
and the like, so we have to be a little careful. To ensure
the right choice, we now pick the first visible window
that is not a GtkInspectorWindow.
Instead, we want to let GTK+ open a window whose life-cycle
it can control. We just ensure that all our types are registered
when the module is loaded, so GTK+ can find them.
The positioning of the highlight window was not reliable; instead
just use a after-handler for the draw signal, in the same way that
drag highlights are drawn by GTK+ itself.
And copy the code for grabbing a widget via pointer from testgtk;
that code is known to work.
Quoting the spec:
If the cascaded value of a property is the unset keyword,
then if it is an inherited property, this is treated as
inherit, and if it is not, this is treated as initial.
Spec in question:
http://dev.w3.org/csswg/css-cascade/
Also use unset in the reset-to-defaults.css we use to reset css in
reftests.
In the GTK_IMAGE_ICON_NAME case, instead of keeping around the icon_name,
create a GThemedIcon and use that. This way, we can reuse the code paths
for the GTK_IMAGE_GICON case.
We have to use the internal accessor to the visible flag, instead of
calling gtk_widget_set_visible(), from within the show() and hide()
implementations - otherwise we'll recurse.
https://bugzilla.gnome.org/show_bug.cgi?id=721069
We now use a narrower trough and paint the optional text
beside the trough instead of inside it. This makes for a much
cleaner appearance and more readable text.
The documentation for GtkFileChooser references the button which allows
the user to create new folders as the New Folder button, although the
label actually says Create Folder.
Update the documentation to say Create Folder instead of New Folder.
https://bugzilla.gnome.org/show_bug.cgi?id=729475
This small refactor makes the code more readable when adding more
transition types that have left, right, up, and down variations.
It adds inline boolean functions to tell information about transition
types (avoiding long if clauses) and changes long chains of "else if
(transition_type == ...)" into switch statements. Both are only likely
to get longer as more transition types are added.
https://bugzilla.gnome.org/show_bug.cgi?id=726676
Added GTK_BUILDER_ERROR_INVALID_PROPERTY and GTK_BUILDER_ERROR_INVALID_SIGNAL
error codes
ObjectInfo: Use a GType instead of a char * for the class name.
PropertyInfo: Use a GParamSpec instead of a char * for the property name.
SignalInfo: Use signal id and detail quark instead of a detailed signal name string.
This not only save us a few malloc in each case but lets us simplify the code
and report unknown properties and signals as a parsing error instead of just
printing a warning.
Fixed memory leaks in parse_object(), parse_template() and parse_signal() functions.
Parameters value where strduped before the last posible return and not freed.
...unless they are labeled.
To implement this, keep track of the nesting depth of section boxes,
starting from 0 for the toplevel and the submenu ones, and only
insert unlabeled separators if the depth is at most 1.
When the adjustment changes (due to e.g. a mouse wheel scroll) we update
the prelight. The part that un-prelights the previous prelight was
broken by the the pixel cache, as it called update_prelight in the
middle of the scrolling operation, where the windows were moved
but the tree_view->priv->dy was not changed to the new value. This
caused the updates to the pixel cache to go to the wrong place.
We fix this by fully doing the scroll before we update_prelight().
https://bugzilla.gnome.org/show_bug.cgi?id=728284
GtkMenuTrackerItem::visible was removed a few commits ago.
It is not necessary to bind visible anyway, since the menu
tracker will insert and remove items as their visibility
changes.
Rendering doesn't do much about clipping drawing operations to the window shape,
although invalidation applies the shape to every window, leaving possibly trails
of "overrendered" content. So ensure the shape portions get invalidated too when
the window is moved/resized.
https://bugzilla.gnome.org/show_bug.cgi?id=729095
Now that popovers may snap to any side with enough space, make enough
room on every side when requesting size, so that there's no w/h differences
at the time of setting the child allocation.
https://bugzilla.gnome.org/show_bug.cgi?id=729097
When activating an item using mnemonics GtkRecentChooserMenu does not select a
item before calling the item-activated signal thus
gtk_recent_chooser_get_current_uri() always return the last selected item
instead of the activated one.
Fixes Bug 495105 "Open recent file keyboard shortcuts do not work correctly"
The real reason for the problem that the just reverted change
was supposed to address is that testfilechooser uses show_all
to tease out places where composite dialogs don't properly
protect their internals against unintended showing. Well,
lets do that, then.
In iconic mode, model buttons will be styled like regular icon
buttons, preferring to show only the icon if one is set, falling
back to showing the label.
https://bugzilla.gnome.org/show_bug.cgi?id=727477
When rendering iconic sections, we want to use icons for verbs,
and we want to differentiate these in the menu model, to keep
the icon attribute reserved for nouns.
https://bugzilla.gnome.org/show_bug.cgi?id=727477
Instead of using GtkMenuTracker to flatten the sections into a single
linear menu, handle the sections ourselves by nesting boxes.
Each section gets an inner and outer box. The inner box numbers its
children in the way that the tracker instructs. The outer box
containes the inner box and the separator, if appropriate.
Having the two separate boxes will allow us to change the orientation of
the inner box if we want to pack widgets horizontally within a section.
Add the possibility of a GtkMenuTracker that performs no section
merging. Instead, it will report an item in the form of a separator for
subsections. It is then possible to get a separate tracker for the
subsection contents by using gtk_menu_tracker_new_for_item_link().
We have some API in GtkMenuTracker and GtkMenuTrackerItem that is
specifically designed to deal with submenus.
Generalise these APIs to take a 'link_name' parameter that we always
give as G_MENU_SUBMENU for now. In the future, this will allow creating
trackers for other types of links, such as sections.
Make this a property just like all of the other things and make the APIs
for accessing it non-private (but add a note that they are not intended
to be used).
This is not a great name to use in themes; instead, add the
menuitem class, so themes can use .button.menuitem for styles
that are specific to menuitem-like buttons.
We were not really handling all cases correctly here. We want
the suggested-action style class to only be set on headerbar
buttons, and it should be set on the default widget. Ensure
this by syncing the suggested-action style class with the
default style class. As a side-effect, setting has-default
on an action widget in ui files will now have the expected
effect.
https://bugzilla.gnome.org/show_bug.cgi?id=728846
GtkDialog has convenience API for adding action widgets that are
either placed in the action area or the headerbar. This commit
makes the same functionality available from GtkBuilder ui files
by specifying "action" as the child type.
https://bugzilla.gnome.org/show_bug.cgi?id=728846
When constructing from a menu model, use popovers by default.
This change has the potential to cause some size problems for
applications with big gear menus, so we're doing it early in
the cycle to uncover and fix those.
It may happen that the received clipboard data is empty, but
if it's of type image/bmp, gtk+ will crash:
gdk_property_change: 00030AD4 GDK_SELECTION image/bmp REPLACE 8*0 bits:
... delayed rendering
gdk_selection_send_notify_for_display: 00030AD4 CLIPBOARD image/bmp
GDK_SELECTION (no-op)
_gdk_win32_selection_convert_to_dib: 1252003C image/bmp
Program received signal SIGSEGV, Segmentation fault.
0x749a9f40 in msvcrt!memmove () from C:\Windows\syswow64\msvcrt.dll
Thread 1 (Thread 2248.0x1b34):
target=0xc07b) at gdkselection-win32.c:1292
at gdkevents-win32.c:3498
wparam=8, lparam=0) at gdkevents-win32.c:232
message=773, wparam=8, lparam=0)
at gdkevents-win32.c:263
C:\Windows\syswow64\user32.dll
C:\Users\rugoosse\AppData\Local\virt-viewer\bin\libpangocairo-1.0-0.dll
wparam=0, lparam=-1687549457)
at gdkevents-win32.c:248
C:\Users\rugoosse\AppData\Local\virt-viewer\bin\libpangocairo-1.0-0.dll
https://bugzilla.gnome.org/show_bug.cgi?id=728745
According to what i've been told, modelbuttons are supposed to look
flat, like menu items.
This is basically an improved copy of the menuitem styles.
https://bugzilla.gnome.org/show_bug.cgi?id=728810
This prevents checkmarks in modelbuttons from being styled as buttons
(apparently, modelbuttons apply "focused" to their children, unlike
normal buttons).
https://bugzilla.gnome.org/show_bug.cgi?id=728808
The events are routed through a new slave device with type
GDK_SOURCE_TOUCHSCREEN, minimal tracking of touches is done
to keep the state for each of those.
https://bugzilla.gnome.org/show_bug.cgi?id=728426
The master pointer/keyboard pair should never disappear or be
inconsistent. The seat capabilities are now reflected through
slave devices, those may come and go freely as the seat
capabilities change. This also enables adding further capabilities
to handle eg. touch.
https://bugzilla.gnome.org/show_bug.cgi?id=728426
When making the entire window a drop target, as file-roller does,
we don't want to draw the drag highlight around the CSD window
decorations.
https://bugzilla.gnome.org/show_bug.cgi?id=728526
Please enter the commit message for your changes. Lines starting
This should allow theme developers to use a very small width for
the resize handle, but still let users easily move the handle by
defining a wider resize area.
The additional resize area follows the "margin" style property.
https://bugzilla.gnome.org/show_bug.cgi?id=728073
Binding an object sensitive property with a check button active property will look like this:
<object class="GtkButton" id="button">
<property name="sensitive" bind-source="checkbutton" bind-property="active"/>
</object>
This is based on the original work done by Denis Washington for his GSoC project
This closes Bug 654417 "[GSoC] Add <binding> element to GtkBuilder syntax"
Get monitor on which the most of the window is located (nearest monitor if
window is not on screen), get its work area (area not occupied by taskbar or
any other bars) and use that for maxsize.
Previous default of 30000 meant that windows maximized onto full screen,
even covering the area where taskbar is.
https://bugzilla.gnome.org/show_bug.cgi?id=726592
When selecting a swatch in a GtkColorChooserWidget the previously
selected swatch and the currently selected swatch are not redrawn. This
can leave the old swatch still marked with a checkbox even though a new
swatch has been selected.
Redraw the swatches after changing the selection.
https://bugzilla.gnome.org/show_bug.cgi?id=727487
The documentation for the GtkWidget::size-allocate signal is missing the
description of the "allocation" parameter. Add the missing description
to the parameter.
https://bugzilla.gnome.org/show_bug.cgi?id=726179
Use GSlice to allocate all types of segments:
- char
- toggle
- mark
- pixbuf
- child widget
Char segments are a bit more complicated because the length of the text
is determined at run time and stored in the 'byte_count' field. If the
text is long, GSlice will call the system malloc() anyway, so it's
better to always use GSlice for GtkTextLineSegment.
Toggle segments are also freed in gtktextbtree.c, hence the function
_gtk_toggle_segment_free() (for a later commit it would be nice to
rename those functions with the _gtk_text prefix).
https://bugzilla.gnome.org/show_bug.cgi?id=727908
There is a possible confusion with the sentence:
"@iter is inside a natural-language word"
The iter should be viewed here as the pointed character (i.e. on the
right of the iter), not as a position between two characters.
Instead of improving the documentation, another solution would have been
to change the implementation so it is interpreted as an iter position
inside a word, i.e. between two characters that are part of a
natural-language word. But maybe some applications rely on the current
implementation.
https://bugzilla.gnome.org/show_bug.cgi?id=727908
- As the tests show, some of the functions have a strange and
inconsistent behavior for corner cases.
- Rename test_full_buffer() -> test_search_full_buffer() because
textiter.c is used for other GtkTextIter unit tests.
https://bugzilla.gnome.org/show_bug.cgi?id=727908
We are using the actionbar in the middle of the window, which
is not really what it is designed for. To avoid the unfinished
appearance on the sides, move it into the frame that we have
around the stack below. This fixes the sides, but gives the top
a double stroke. Too bad.
The text view is resizing several times after the window is mapped.
Not setting a hscrollbar-policy of never avoids that, and a
scrollbar still doesn't appear. Magic
The incremental loading was broken by GtkIconHelper - queuing a
redraw is no longer sufficient to cause GtkImage to redraw with
the new pixbuf contents.
Pointed out by Jasper St. Pierre.
The keynav dialog is transient to the example window; since the
example window is now modal, we need to make the keynav dialog
modal as well, so it can receive input.
Problem pointed out by Jasper St. Pierre.
We are keeping references on the widget we are handling as we
are iterating up, but that doesn't protect us against the entire
tree being axed from inside gtk_widget_handle_event.
https://bugzilla.gnome.org/show_bug.cgi?id=727644
Commit d05191a010 change the height
requisition to be completely dependent on the PangoLayout, but that
breaks when the font has special characters with different metrics.
Use the maximum between the two instead.
https://bugzilla.gnome.org/show_bug.cgi?id=728054
The use of border-width-set here was an attempt to differentiate
between explicitly set (from code / ui files) border width from
theme changes. But when we are calling gtk_window_set_border_width
to apply the theme value, the -set property gets set, and all
further theme changes are ignored. This has the effect of only
letting the default value of these properties get applied.
Fix this by unsetting border-width-set after applying theme values.
Because GTK does not invalidate windows that aren't mapped, we cannot
update the pixel cache when the window it handles isn't mapped. So we
add API to call when GDK windows get mapped/unmapped.
https://bugzilla.gnome.org/show_bug.cgi?id=726475
This commit makes it possible for GtkSwitch to indicate when
the underlying state changes with a delay, causing the switch
to temporarily go 'out of sync' with the underlying change.
https://bugzilla.gnome.org/show_bug.cgi?id=725648
This commit adds API for dealing with multi-selection. It is identical
to the flow box API for this purpose. The implementation is still limited
to single-selection, and will be updated in subsequent commits.
It can happen that we get a size request when the main widget
is still NULL. Currently we hit a critical in this case, and
stumble on. We can do better.
Opening a new tab in nautilus is hitting this case.
https://bugzilla.gnome.org/show_bug.cgi?id=727643
We are getting bug reports from people who are irritated that
dialogs now have 'double headers' under any wm but gnome-shell.
As an example, xfwm4 seems to do ok with csd windows, and
on balance it seems better to have some invisible border issues
than to have double headers.
https://bugzilla.gnome.org/show_bug.cgi?id=727414
Add support to build the introspection files for GdkWin32, as done recently
in the autotools builds and clean up the NMake Makefile for building the
introspection files a bit.
For some reason, gdk_win32_display_manager_get_type() was not exported in
gdk-3.0.lib, force its export, so that the GdkWin32-3.0.gir can be built
properly with the Visual C++ builds. This is a known problem that some
symbols in static libraries that are linked into a DLL in Visual C++, even
if they were marked with __declspec(dllexport) via _GDK_EXTERN.
* Makes listbox background white instead of default grey
* Uses gradients or W32 theme parts to draw sexy selection/prelight
rectangle instead of changing selected/prelighted item background
* Removes blanket button text color, allowing buttons to inherit
text color from their parents. Non-normal buttons DO get specific
text color though. This partially fixes text color propagation from
listbox rows to their children.
https://bugzilla.gnome.org/show_bug.cgi?id=727244
The arrow is part of the background image drawn by the theme, don't draw
GTK's own arrow. This also applies to non-entry comboboxes.
On XP the arrow-and-nothing-else part does not exist (W32 XP theme draws a
white rectangle) and has to be subtituted for a simple dropdown button even
for non-entry comboboxes.
https://bugzilla.gnome.org/show_bug.cgi?id=727035
It seems that the dec-button no longer has a "bottom" selector-thingy to
select for. Use "last-child" instead.
Use .vertical to style vertical instance differently, as last-child/first-child
meaning is inverted there.
Increase (towards 0) margins a bit, to avoid clipping the pre-light rectangle.
Use dir(rtl) selector to style horizontal instance differently for RTL locales,
where first-child is "inc".
https://bugzilla.gnome.org/show_bug.cgi?id=727022
...on Windows 8+ and when the system setting for non-Unicode programs do
not match the language version of Windows by falling back to using Pango.
This ensures that the correct font is used during these scenarios, so that
we minimize the risk of seeing garbled characters for texts that the system
code page does not support due to system peculiarties. There might be a
way to support gtk-font-name handling using the native Windows APIs
directly on Windows 8+, but that needs to be investigated.
https://bugzilla.gnome.org/show_bug.cgi?id=726298
GTK_TREE_VIEW_TIME_MS_PER_IDLE is currently 30 milliseconds, meaning
that validate_rows will validate rows up until all the validations have
taken over 30 msecs. So it's likely to block redrawing via the clock
frame update mechanism, as that tops at 16.66 milliseconds per frame
(1/60th of a second).
Stop validating rows if we've spent more than 3/5 of our allotted budget
for inter-frame processing, so as to avoid blocking.
In the future, we would probably want to calculate how long we would
have left until the next frame, especially if higher priority idles
and timeouts have already consumed a portion of that allotted time.
https://bugzilla.gnome.org/show_bug.cgi?id=726871
A "typo" led to using a wrong GtkTreePath when converting the path of the
virtual root to check the ancestors, which would lead to either no checks being
performed, or maybe segfaulting when using an invalid path as result.
https://bugzilla.gnome.org/show_bug.cgi?id=722058
Signed-off-by: Olivier Brunel <jjk@jjacky.com>
These AtkRelation types are added automatically for widgets with a label
specified (e.g. via gtk_label_set_mnemonic_widget, gtk_frame_set_label,
and gtk_frame_set_label_widget). When such specification is absent, the
accessible relationship must be manually set.
https://bugzilla.gnome.org/show_bug.cgi?id=726996
Commit faba7df4fe changed the logic in
apply_emblems() so that GtkIconInfo->emblems_applied would be set to
TRUE even in case there was no emblem info available, which confuses the
theme cache.
This commit changes the logic back, so that NULL is returned from
apply_emblems_to_pixbuf() when there are no emblems available, fixing
the bug.
https://bugzilla.gnome.org/show_bug.cgi?id=726830
The compositing that is meant here is really specific to the
X11 Composite extension, and does not apply to Wayland.
This is very rarely used functionality anyway, and none of
the other backends support it.
This way, we don't create lots of cell accessibles when creating the
first one (because surely one is the parent/child of another who again
is a parent/child of another who again....)
Nobody was reffing those related object in the first place and that
was causing random crashes.
And if somebody had reffed those related objects, they'd have caused
reference cycles.
https://bugzilla.gnome.org/show_bug.cgi?id=726838
This fixes an issue where the theme-provided border-width prevents
dialog contents from lining up properly with the headerbar. To make
this work in message dialogs, we have to explicitly set the border-
width of the action area to 0.
In select-folder mode, we are putting the directory name into the
entry ourselves. Then the entry appends a /. If we react to this
'spontaneous' change of the entry by clearing the list selection,
this will in turn make us clear the entry. We don't want that.
https://bugzilla.gnome.org/show_bug.cgi?id=726855
Setting windows undecorated was broken by some of the recent
shadow width changes. We need to ensure that shadow width is
zero for undecorated windows, then things work again.
If the delete event ends up destroying the widget, unsetting
priv->delete_event_handler will happen on invalid memory, so
unset it before the widget is possibly destroyed.
https://bugzilla.gnome.org/show_bug.cgi?id=726825
Theoretically, we apply the shape mask client-side ourselves
with an ARGB32 pixmap and intersect it to get a union shape,
but I don't particularly care enough to write that code.
Realistic application code using bounding shapes in 2014 is
quite rare.
Widgets should only call set_realized() after having created and
registered their GDK windows. In this case, the creation of the style
context (or more exactly: figuring out the scale factor for it) requires
knowing if the widget is already realized. Which it isn't.
https://bugzilla.gnome.org/show_bug.cgi?id=726717
expected a valid semicolon
sounds kinda not so great. Make it say
expected semicolon
Unless somebody can tell me the difference between valid and invalid
semicolons?
We had a margin of 30 before/after the text. Put that add around
the icon as well, and separate the icon and text by 30 pixels.
This does not affect the appearance of message dialogs without
icons.
It seems that some backends implemented get_root_origin wrong
and returned the client window coordinates, not the frame window
coordinates. Since it's possible to implement generically for all
windows, let's do that instead of having a separate impl vfunc.
Lots of code, including dragging code in GtkWindow, use these
fields. Setting them to 0 causes lots of strange and weird bugs.
Use the same "hack" from query_device_state of just using
win_x / win_y for now. We'll convert this to the proper fake root
coordinate system used by get_root_coords in the next commit.
If a textview had lateral windows that might displace the text window, the
handles and popovers would appear displaced. Those lateral windows aren't
affected by RTL/LTR settings, so just checking for left/top is ok here.
And the counterpart to unmaximize when dragging a maximized window, if
touch devices aren't going to use EWMH moveresize, having this one at least
makes things feel a bit less awkward.
https://bugzilla.gnome.org/show_bug.cgi?id=709914
Sadly, EWMH moveresize mechanism can't work with touch devices for two
reasons:
1) As a mutter implementation detail, the device is queried in order
to check whether the dragging button is still pressed. Touch devices
won't report the button 1 being pressed through pointer emulation.
2) Even bypassing that check, on X11 touch events are selected prior
to sequences being started, either through XISelectEvents or
XIGrabTouchBegin, no late registering through active grabs is allowed,
as WMs do on reaction to EWMH moveresize messages.
So for the time being, make touch devices fallback on emulated window
dragging, which at least allows for moving windows.
https://bugzilla.gnome.org/show_bug.cgi?id=709914
We want to present a clean, rounded top when there is nothing
else to show, but many dialogs in applications rely on showing
information in their title, so add a label and show the title
when it is not empty.
Add gdk_device_get_last_event_window(), and use to implement the window
tracking we need for synthesizing crossing events for sensitivity changes
and gtk grabs, rather than keeping the information in qdata and updating
it based when GTK+ gets events.
https://bugzilla.gnome.org/show_bug.cgi?id=726187
Stop ignoring various crossing events from grabs:
Enter events with type GRAB/GTK_GRAB/GTK_UNGRAB/STATE_CHANGED:
Ignoring these events was added as a workaround for synthesized
events not having the right coordinates (see bug 555109) but
now they do have the right coordinates. (see bug 704456)
Leave events with types types GTK_GRAB/GTK_UNGRAB:
Ignoring these events was added because since we were ignoring
the enter events as above, ignoring the leave events meant we
could lose the prelighted row in a grab-triggered leave/enter
pair. (See bug 653676. It's also now impossible to
reproduce the leave events that were reported in that bug as causing
problems.)
Leave events of type GRAB.
Ignoring these events was added without a ChangeLog entry in 2001,
possibly to keep the prelight from flashing when activating menus.
But ignoring these events could lead to stuck prelighting, and we don't
do it for any other widgets.
https://bugzilla.gnome.org/show_bug.cgi?id=726209
The bug this patch is fixing is that currently if you have a GtkPopover in
clicking off the popover to dismiss it on a GtkTreeView (which triggers
a synthetic enter event on the GtkTreeView) will leave the GtkTreeView
in a confused state until the user moves the mouse again.
That doesn't make sense.
And it causes issues, because when holding down the tab key, we
show/hide a lot of windows and cause a lot of map/unmap events that
stall the event pipeline.
We did not set an input shape on the window, so the region outside
the invisible border where we draw the outer edges of the shadow
were still part of the window, as far as clicks and cursors were
concerned. Fix this by setting an input shape that makes all clicks
outside of the resize borders go through to the underlying window.
https://bugzilla.gnome.org/show_bug.cgi?id=726125
eb1ab0dac2 removed support for authentication
based on crypt()-hashed passwords but it didn't remove the header.
Finish up with the removal.
This allows the broadway backend to build on FreeBSD (which has no
crypt.h).
https://bugzilla.gnome.org/show_bug.cgi?id=726149
And let GtkPopover use it as its GtkAccessible implementation, this
accessible sets the POPUP_FOR relationship to the relative-to widget,
and keeps track of changes there.
https://bugzilla.gnome.org/show_bug.cgi?id=725864
As those are internal children, there's no signal that GtkWindowAccessible
could catch when those are added or removed, so make GtkWindow use the private
GtkContainerAccessible methods to add/remove the child accessible when that
happens.
https://bugzilla.gnome.org/show_bug.cgi?id=725864
window->x / window->y are in "root window coordinates", e.g. relative
to the topmost toplevel. However, the coordinates in get_xdg_popup are
relative to the passed-in surface, so we need to do the reverse
translation here.
As discussed on desktop-devel-list [1], "There should be an intuitive,
consistent, immediate way to jump to the widgets that live in the
header bar." F10 has been suggested for this as it is already used to
active menubars.
F10 will focus the custom titlebar widget if the window has one and it
isn't already focused. If the titlebar widget doesn't exist or is
already focused then F10 focuses the menubar if there is one.
[1] https://mail.gnome.org/archives/desktop-devel-list/2014-February/msg00176.htmlhttps://bugzilla.gnome.org/show_bug.cgi?id=725141
It turns out popovers are already smart enough to cope with this
situation, so let popovers be internal children so things that rely
on gtk_container_forall(), like DnD, work without modifications.
https://bugzilla.gnome.org/show_bug.cgi?id=725727
We have a hack in the XSETTINGS code to substitute gtk-xft-dpi
with gdk-unscaled-dpi unless the screen has a fixed window scale,
in which case we just use gtk-xft-dpi.
But if the screen is changed to have a fixed window scale, then
the substituted value of gdk-unscaled-dpi will stick around until
the next (coincidental) change to XSETTINGS. To fix this, force
an immediate reread of the XSETTINGS property when
gdk_x11_display_set_window_scale() is used.
https://bugzilla.gnome.org/show_bug.cgi?id=725754
It may be unusual, but handlers of day-selected may want to transfer
focus somewhere else, without getting it reset back right after by/to
the calendar. This makes popovers demo work on the calendar again, for
one...
Make the popover temporarily undo the GTK+ grab, so it remains modal
to its window, but does not attempt to steal focus on other non-modal
windows that get the focus.
This was most confusing with keyboard navigation, as the focus would
remain stuck on the popover, and not move to the newly focused window
after the popover was dismissed. It didn't have as much effect on
pointer operations as only the first click would be consumed in order
to hide the popover.
There are early returns in this method before the completion timeout
is set later on, so set the source to 0 to avoid trying to remove it
later again.
https://bugzilla.gnome.org/show_bug.cgi?id=725824
This is not necessary for the popover itself, but helps tooltips
code confine the widget lookup within the popover if the pointer
is inside it, otherwise the widget lookup may turn out wrong for
motion events, starting the tooltip widget lookup from the toplevel
window, mistakenly triggering tooltips on the natural window
descendants (ie. the widget below the popover)
https://bugzilla.gnome.org/show_bug.cgi?id=724785
gnome-terminal is still using this setting, so we'll let
applications override it for another cycle. It is no longer
backed by a system-wide setting, though, and it will still
go away eventually.
This partically reverts b26c74e5da
gnome-terminal is still using this setting, so we'll let
applications override it for another cycle. It is no longer
backed by a system-wide setting, though, and it will still
go away eventually.
This partically reverts 7e3a494fac
Add documentation for GtkTreeView::move-cursor
Add links to GtkTreeModel::row-inserted and GtkTreeModel::row-deleted
in the documentation for gtk_tree_view_set_reorderable ().
https://bugzilla.gnome.org/show_bug.cgi?id=725560
If we aren't using a header bar then put a fake titlebar
box on it so we can round the corners.
One of the advantages of this is so that the styling of the dialog
is completely within one theme framework. This prevents skew between
the theming expectations from the window manager and GTK+.
https://bugzilla.gnome.org/show_bug.cgi?id=725345
GOBJECT_INTROSPECTION_CHECK macro can be missing if introspection
is not installed, so this way the following error is prevented:
"gtk/Makefile.am:1324: error: HAVE_INTROSPECTION does not appear
in AM_CONDITIONAL"
https://bugzilla.gnome.org/show_bug.cgi?id=723438
This is a web service provided by Google that allows people to
share their printers (https://www.google.com/cloudprint/learn/).
In addition to being able to print to printers shared on Google Cloud
Print, there is an equivalent of "Print to file" in the form of "Save to
Google Drive".
The cloudprint module uses gnome-online-accounts to obtain the OAuth 2.0
access token for the Google account.
Currently it can discover available printers, get simple details about
them such as display name and status, and submit jobs without any
special options.
https://bugzilla.gnome.org/show_bug.cgi?id=723368
Restore the drawing order in GtkFixed to what it was in 3.8. With the
GDK drawing changes this will not be correct in some cases (un-windowed
children can now overlap windowed children and native children overlap
everything), but fixes Eclipse drawing.
https://bugzilla.gnome.org/show_bug.cgi?id=725089
The point of GdkWindowImpl::get_root_coords is to translate the passed
in coordinates against the passed-in impl window. For a child window,
in fact, window->abs_x and window->abs_y already track the child
window's coordinates against the impl window.
If we pass in a child window, and backends don't explicitly get the impl
window from it, we'll double-count the child window.
Really, we should *always* be passing impl windows to backends, and
never child windows. However, I'm a bit worried for regressions late
in the cycle if we want to fix up the rest of the callers, like
gdk_window_get_geometry, so I'm only going to touch get_root_coords
for now after careful review of all the backends.
GtkWindow calls set_shadow_width then maps the window, meaning
that we never set the margin. Save it when we set and then set
it when we create the XDG surface.
In the unlikely case that there is another GPL released in the future
it would be best if we link directly to the 3.0 version of the
license description instead of the alias to the latest
version.
Old code assumed that AT-SPI would keep track of references and
therefore tried to only hold weak references. On the other hand it also
tried to keep objects alive so it referenced objects very randomly. All
of that lead to cycles and leaking.
As AT-SPI does not keep track of objects at all, the treeview now does.
The refcounting looks as follows:
GtkTreeViewAccessible
=> creates per row/column
GtkTreeViewAccessibleCellInfo
=> which references 1
GtkCellAccessible
If there is only one cell, this accessible is a
GtkRendererCellAccessible, otherwise a GtkContainerCellAccessible is
created and that accessible holds references to the
GtkRendererCellAccessibles that are created for every cell renderer.
https://bugzilla.gnome.org/show_bug.cgi?id=554618
Instead of destroying the surface in the backend if this is
unable to resize, let the core code do it, and do it properly.
Based on a patch by Benjamin Otte.
https://bugzilla.gnome.org/show_bug.cgi?id=725172
GObject-Introspection was recently changed to support acquiring the name of
the DLL from a library (.lib, etc) that was passed into g-ir-scanner on
Windows, like the *nix builds, instead of directly passing in the name of
the DLL.
This updates the introspection build process, so that introspection files
for GTK+ can continue to be properly built.
The code in GDK is incredibly broken and nobody is quite sure what's
right-side-up and what's upside down, but this breaks mutter-wayland
now, so let's remove it. It might leak, but we should probably do a
full restructuring of GDK drawing to fix it.
Like in other backends (except X) we can't resize cairo image surfaces
so let's sync the code here with what the other backends do.
This prevents the painting machinery above us to paint on the wrong
buffer.
https://bugzilla.gnome.org/show_bug.cgi?id=724968
The new() function is supposed to return an empty fully initialized
GtkCssKeyframes object, while the alloc() function just allocates and
initializes static values. So alloc() can be used for copying or
resolving keyframes.
Fixes a memleak when resolving keyframes.
With the code as written, use-popover has to be set first,
before the model. To avoid this ordering dependency, re-set
the model when use-popover changes.
The convention we follow is that the PROP_foo define should
match the property name. Therefore, change PROP_MODEL to
PROP_MENU_MODEL to match "menu-model".
gtk_tree_view_remove_column was first removing the column from
its list, then call gtk_tree_view_column_unset_tree_view, which
would then call gtk_container_remove to remove its button from
the treeview. But the treeview remove implementation relied
on the column being still in the list in order to recognize
the button as 'special', so in effect the button was never
properly removed and thus, leaked.
Fix this by callling unset_tree_view before removing the
column from the list.
https://bugzilla.gnome.org/show_bug.cgi?id=724891
c287845240 was trying to fix
the memory leak caused by popovers begin destroyed in
gtk_window_destroy before chaining up to gtk_widget_destroy,
which unrealizes the window, and would clean up the popover
windows if the popovers were still around.
Fix this in a better way by moving the popover destruction
after the chaining up, so we unrealize first, and then
destroy the popovers.
Also, make _gtk_window_remove_popover unrealize the popover,
for symmetry with _gtk_window_add_popover.
This should fix
https://bugzilla.gnome.org/show_bug.cgi?id=724921
Normally, a GtkAboutDialog is shown using the convenience
API. But if you manually construct one and show it by calling
gtk_widget_show_all() on it, the license tab would show up
uninvited. Fix that.
https://bugzilla.gnome.org/show_bug.cgi?id=724411
Add some flexibility in the property sheets for one building GTK+ that it
also searches for a settable installation path of Python, in addition to
searching the PATH for an installation of the Python interpretor. This
currently defaults to Python 2.7.x, which is normally installed in
c:\python27 on Windows by default. Also tell people in the README.txt's
for the Visual Studio builds
In practice this shape is only used to outline the popover when it is
above native windows, in the most normal full-csw case the shape won't apply
visibly, so popovers will still be able to cast a shadow there.
If there are native windows below the popover, the shape will exclude the
shadow, so there are no alpha contents above the window. One worst case that
might happen is that the popover lays above patches of native/client-side
windows, so the shadow could come and go around the border. But first let's
see whether that happens often or visibly enough before adding something more
convoluted.
The update of the needs-attention state is done via its own property,
so it doesn't need to be done via visibility changes.
This patch is largely the result of inspecting the code due to a warning
and not a result of testing. So if issues pop up that bisect back to
this patch, that's why.
And document the fact that the popover will get destroyed if
a NULL relative-to is given on a parented popover, if no extra
references are kept.
For gtk_popover_new*(), a NULL relative-to will leave the widget
as a floating object, to be sunk by a later call to
gtk_widget_set_relative_to().
https://bugzilla.gnome.org/show_bug.cgi?id=724407
The recursion in map needs to follow the actual physical
widget tree, otherwise we violate invariants. The generic
container map implementation uses gtk_container_forall to
operate on the children, and thus is not suitable for
containers where the children are inside some internal
container.
We should only eat button release events when the label is
actually selectable, since the comment indicates that we
want to eat the release events belonging to press events
that triggered a selection. This fixes problems with actions
on parent widgets that are triggered by button release,
as seen in this bug:
https://bugzilla.gnome.org/show_bug.cgi?id=724541
Suprisingly, this bug has been there for a very long time.
I'm fixing it now because we now use a custom search entry
in the app chooser dialog, and this is causing the templates
cleanup test to fail.
The with_separators argument does not really make sense
for popovers, it was just copied from the menu implementation.
Drop it now, before it becomes part of the public API.
Dragging windows was not working on widgets in the titlebar
region unless they had the window-dragging style property
set. Fix this by looking at the region for motion notify
events as well as for buton press events.
The code setting up the button has been move a little later
in the dialog construction, with the effect that the entry
is already insensitive when we set up the binding.
We had already set the image to be hidden in the .ui file.
This patch removes the image altogether, and deprecates the
property, setter and getter.
If an image is explicitly put with the setter, it is still
shown, so to not break existing users of this API.
Based on a patch by Jon McCann.
Try to do a better job of keeping example content
from being too wide. It is often rendered as <pre>
text so the only time we can wrap it is in the source.
It is best to full break lines at all punctuation and
to try to keep the width under 70 chars or so.
We don't want the size request to change as icons come and
go (thinking e.g. about the caps lock warning). Just make
sure that we have enough room for showing the icons.
Improve the algorithm to determing popover placement:
If it fits in the preferred direction or its opposite,
do that, otherwise use the direction that causes the least
of the popover to be cut off.
Add api to allow explicitly setting a GtkPopover instead of
a GtkMenu as the popup of a GtkMenuButton. Also, add api to
instruct the menu button to construct a popover when given
a menu model.
We set the style class "menu-button" on the button only when
it pops up a menu, to allow different treatment for the active
state of the button in the two cases.
https://bugzilla.gnome.org/show_bug.cgi?id=723878
cairo_rectangle_int_t replaced by GdkRectangle whenever it is used.
Also, rect parameter in public method gtk_popover_set_pointing_to
made const.
Bug #723394
Replace GTK_TYPE_WIDGET with more specific GTK_TYPE_FLOW_BOX_CHILD
for GtkFlowBox::child-activated. This matches signature of
child_activated slot in class struct.
https://bugzilla.gnome.org/show_bug.cgi?id=723716
The properties are declared read-write, but only the setter
was hooked up. This was leading to criticals in test apps using
the prop-editor.c code. Complete the implementation by adding the
getter side too.
This adds a new function, gtk_popover_new_from_model, which creates
a popover and populates it with suitable content according to the
menu model. The current implementation uses GtkModelButton for the
individual items, and a GtkStack for submenus.
https://bugzilla.gnome.org/show_bug.cgi?id=723014
Similar in spirit to GtkModelMenuItem, this private GtkButton subclass
can connect to a GtkMenuTrackerItem and present itself as either a
regular button, a check button, or a radio button. Activation and
state tracking is done through the GAction that is associated with
the menu tracker item.
https://bugzilla.gnome.org/show_bug.cgi?id=723014
Popovers aren't direct children of the widget they point to, but yet
they act as children of it, so do the same with state propagation,
so the flags that propagate across the hierarchy reach popovers too.
Anytime ::grab-notify comes across, the popover visibility and GTK+
grab ownership are checked, so the popover is hidden when it loses
the GTK+ by any reason.
Previously we did a semi-successful job at ignoring it. Unfortunately
this job was bad enough that we could lose the direction.
We still allow passing in the enum values, because we want code like
this to work:
set_state_flags (get_state_flags() | SOME_FLAGS)
Heavy duty can prevent this idle function from being called before
the window is destroyed, so make sure that the source is removed
when the window is finalized.
https://bugzilla.gnome.org/show_bug.cgi?id=723771
This hideous hack is necessary so far because the main users of
GtkScaleButton are also clutter users, so the GtkScaleButton popover
will be very likely shaped against a clutter-enabled native window.
https://bugzilla.gnome.org/show_bug.cgi?id=723556
Popovers no longer sets a shape, unless this function is called. This
function exists so widgets that are potentially placed on top of other
native windows can get a popover that's nicely shaped, even if it has
no border shadow around.
https://bugzilla.gnome.org/show_bug.cgi?id=723556
Since realize does a lot of the heavy lifting of setting up
csd, we have to re-realize the window if we go from no-custom
titlebar to a custom titlebar or vice versa.
https://bugzilla.gnome.org/show_bug.cgi?id=722919
When gtk_window_set_titlebar is called, we need to set up
client-side decorations properly, and the easiest way to do
so is to realize the window again. Really, you should call
set_titlebar before the window is realized.
https://bugzilla.gnome.org/show_bug.cgi?id=722919
10b5ec20 made sure not to set focus_child to NULL all the way up to the
top, but only up to the common ancestor. However, it would never set it
on the common ancestor itself, which would therefore remain with a
focus_child set when it shouldn't.
A manifestation of the bug: focus column headers of a treeview, press Tab.
Now pressing Shift+Tab will go to another widget and not the column
headers, and Tab will (appear to) do nothing, all because the treeview
still has a focus_child set to column headers after a grab_focus().
Signed-off-by: Olivier Brunel <jjk@jjacky.com>
https://bugzilla.gnome.org/show_bug.cgi?id=723402
We can't destroy buffers if they're in-use by the compositor. Well,
technically we can, but that is considered undefined by Wayland and
mutter won't cope with it very well -- it simply kills the client.
To solve this, we need to delay the destroy operation until the
compositor tells us that it's released the buffer. To do this, hold
an extra ref on the cairo surface as long as the surface is in-use
by the compositor.
This reverts commit ba6128f8af.
This change breaks emacs drawing entirely. Since GtkFixed is
somewhat of a legacy widget anyway, lets just not bother doing
this modernization there, at least for now.
Showing tooltips on top of a transient popup does not work
out well, and is not really necessary here. At the same time,
remove the unnecessary repetitions of properties. In particular,
setting the label of the buttons here defeats the scale buttons
use of symbolic icons.
https://bugzilla.gnome.org/show_bug.cgi?id=723181
This prevents warnings like
(gtk3-demo:14948): Gdk-CRITICAL **: _gdk_frame_clock_thaw: assertion 'GDK_IS_FRAME_CLOCK (clock)' failed
(gtk3-demo:14948): Gdk-CRITICAL **: gdk_frame_clock_get_timings: assertion 'GDK_IS_FRAME_CLOCK (frame_clock)' failed
We need to do this, as the compositor might have already sent us a frame
event, in-flight, at the same time we destroy our window. In this case, we'll
receive the then-in-flight "done" event, and then warn as we try to look
up the frame clock on a destroyed window.
disconnect_by_func() is slow, and this becomes particularly evident
when disposing a number of widgets (and their associated style
context) at once, such as when using a language binding which
uses a GC.
https://bugzilla.gnome.org/show_bug.cgi?id=723183
It seems that alternate implementations of GtkFileChooserWidget
never materialized. The split between GtkFileChooserWidget and
GtkFileChooserDefault is awkward. The immediate problem is that
it makes it difficult to document the keybinding signals. So it
makes sense to drop the abstraction and just have one thing.
https://bugzilla.gnome.org/show_bug.cgi?id=723157
The button press/release handlers did invariably return FALSE, even
though it shouldn't if a row was found on the event coordinates. Also,
use GDK_EVENT_* defines for the return values.
The minimal size if no child widget was present/shown was far too small
to have enough room for the arrow width plus border radii, so
gtk_render_frame_gap would spew warnings about the gap being out of
boundaries.
Fixes issues seen in
https://bugzilla.gnome.org/show_bug.cgi?id=723031#c2
If the grab is released during button press, the button release is
just then sent to the widget below the pointer. Depending on the
widget implementation, this could already trigger actions if the
widget does not perform any kind of button state tracking. It is
safer to ungrab on button release so no extra actions are possibly
triggered, and the behavior is uniform across widgets.
But the opposite situation may also happen, that a popover is
shown/grabbed on a button press event, so it'd get the sole button
release event after being shown, so prepare for that case by making
popover ignore single button release events with no preceding button
press.
Fixes issues seen in
https://bugzilla.gnome.org/show_bug.cgi?id=723031#c2
Removable USB drives or memory sticks should be powered down
when the eject button is pressed. For this, we need to call
g_drive_stop() instead of g_drive_eject(), provided the drive
can be stopped.
https://bugzilla.gnome.org/show_bug.cgi?id=723121
Add printer_compare() function for comparing printers according
to their names and locations. It is possible to search by multiple
keys separated by space or tabulator using logical conjunction.
Based on patch by William Hua.
https://bugzilla.gnome.org/show_bug.cgi?id=692931
The root window is a fairly X-centric concept, and it
really has no place in the GtkWidget API. Plus, this
is a rarely-used one-line convenience function with
poor documentation.
Put dialogs and utility windows in the same level as normal and
toolbar windows so that Gtk can control their stacking instead of
forcing them, rather unnaturally, to be on top of all other windows,
even other application windows, even when another application has
focus.
Instead of setting "use-header-bar" from gtk_about_dialog_new(),
override its default value by inspecting the properties at constructor()
time. This makes it work from bindings as well, since they don't use the
convenience C constructor.
https://bugzilla.gnome.org/show_bug.cgi?id=722574
GtkWindow has 4 (!) APIs for setting window icons, and we
have to try them all in the right order to find the right
icon. This commit makes it so, and keeps the icon list
manipulation inside gtkwindow.c by adding a private API
for getting a single icon at the right size.
https://bugzilla.gnome.org/show_bug.cgi?id=722515
This is needed to show the gtk3-demo icon and is needed for the "Multiple
Views" demo to work. Hmm, why couldn't I do a for loop for a batch in a
property sheet ? :|
It was only done so the background would connect visually to the popover
tail, but then it brings aliasing issues when the border is drawn over the
background. Instead, overdraw the tail, so it also fills the gap left by
the border.
If pointing_to starts falling outside of the parent scrollable allocation,
the popover will be automatically hidden, and shown back again when
pointing_to scrolls back to visibility.
If the rect a handle points to starts falling outside of the parent
scrollable allocation, the handle will be automatically hidden, and
shown again when the rectangle is visible too.
When all popovers are removed on destroy(), if a popover is nested into
(eg. with relative_to within) another popover, the removal of one can
lead to the other being removed while the hashtable is being iterated,
which would lead to undefined behavior in further iterations.
Then, use a GList to store popovers, iterating can be made more resilient
on these situations, and unless on pathological cases there's not going
to be as many of those popovers as to cause performance decreases at the
times those are iterated.
The popovers may return keyboard grabs to previous widgets, so if
called after unsetting the focus, the window may be left with a
dangling GtkWidget that would cause crash at later dispose() calls.
With only get_preferred_width and get_preferred_height implemented,
we end up calling the GtkBin height_for_width implmementation, which
knows nothing about the margins and paddings that GtkPopover needs.
As a result, a listbox added to a popover was getting cut off
at the bottom.
This property is TRUE by default, when a popover is modal, it
will automatically set a GTK+ grab on the popover, and grab
the keyboard focus into the popover.
The GtkBuilder window containing the complex popover UI was left
dangling, and with a dangling pointer to its former child, causing
crashes on gtk_grab_notify() after the popover was destroyed.
Popovers are strange in the sense that they aren't attached to a
parent directly, they rely on the relative_to widget so the toplevel
is shared, and when they have a parent, it is the toplevel itself,
not relative_to. This also means that there are conditions where the
popover loses it's parent, so they must survive unparenting.
The previous code would be floating the last reference as soon as the
parent is gone, but it was non-obvious who'd own that reference. So
fix this situation by granting the ownership of popovers to their
relative_to widget, an extra reference may be held by the toplevel
when the popover has a parent, but the popover object will be
guaranteed to be alive as long as the parent lives.
This way, memory management of popovers is as hidden from the user
as regular widgets within containers are, users are free to call
gtk_widget_destroy() on a popover, but it'd eventually become
destructed when relative_to is.
When a popover is focused, the focus is forwarded so the first
child what would get the focus actually gets it. Also, implement
correct focus chain, so the keyboard focus stays within the popover
when navigating with keyboard.
This makes it possible to move/resize client-side decorated windows that are
otherwise obscured by a GTK+ grab somewhere else, either a popover within the
window itself or a modal dialog above the window.
This offers the same behavior, but GDK_WINDOW_TEMP windows aren't used
anymore, involving less translations from/to root coordinates, plus no
glitches in having handles snap to content as windows move.
If there is a GTK+ grab on the popover, ensure that it's removed when it's
unmapped. If no GTK+ grab was performed on the popover, this function will
do nothing.
Those functions aren't as useful anymore, hiding/showing can be
controlled by setting the widget visibility, and grabbing can be
achieved by performing a GTK+ grab.
GdkWindows are gone now from the API, the pointed_to rectangle
is from now on relative to the widget allocation. GtkTextView
and GtkEntry were updated to adapt to this change.
This is not as necessary now that bubble windows are popovers, if
a modal behavior is wanted on popover contents, a GTK+ grab on the
popover widget will suffice.
Popovers are transient floating widgets that are confined to the
window space. These have their own GdkWindow that is set on top
of the regular window contents, so they can be used for popup menu
alike UIs with custom popup/popdown/grabs behavior.
The color chooser test is constantly running into the
problem that the custom color setting is not empty.
Avoid that by using the memory settings backend.
This new method allows getting a widget from a GtkStack when we know its
name, and will also return NULL if there is no widget going by that
name.
Usage example would be to check if a child with a given name exists
before calling gtk_stack_set_visible_child_name().
https://bugzilla.gnome.org/show_bug.cgi?id=722588
It would be nicer if we could have the tests specify what environment
their expected output was created in, then we could test multiple
scenarios. For now, just fix the setting to avoid test failures.
When running on quartz, it is no longer expected for applications to
provide their own application menu. Instead, they should simply ensure
that they provide "app.about", "app.preferences" and "app.quit" actions
(which many apps are already doing).
A default menu will be shown that looks like the one presented by all
other Mac OS applications, containing menu items for the above actions,
as well as the typical "Hide app", "Hide Others and "Show All" items and
the "Services" submenu.
If an application does explicitly set an application menu (via
gtk_application_set_app_menu()) then it will be respected, as before.
https://bugzilla.gnome.org/show_bug.cgi?id=720552
Add support for extracting strings from GMenu markup in GtkBuilder
files.
This means that we have to support translatable <attribute/> tags.
Unfortunately, <attribute> is also used by GtkTreeViewColumn and
GtkLabel for other purposes, and those other purposes use a value=''
(XML) attribute, so we must accept (and ignore) that in order not to
have errors. Nothing will happen in those cases because they do not
also specify translatable='yes', so we ignore them.
https://bugzilla.gnome.org/show_bug.cgi?id=720552
Add a new private API to GtkApplication akin to
gtk_widget_insert_action_group().
We'll use this to insert a few extra actions at the app level with a
separate namespace for the special items in the Mac OS application menu.
https://bugzilla.gnome.org/show_bug.cgi?id=720552
Add a private hack to allow the insertion of the name of the application
into the label of menu items.
If it appears in the label of any menu item, "%s" will be replaced with
the name of the application.
We will use this for the "Hide myapp", "Quit myapp" and "About myapp"
labels typically found on Mac OS programs.
https://bugzilla.gnome.org/show_bug.cgi?id=720552
Allow the possibility for items to be marked with a special attribute and
expose this via GtkTrackerMenuItem. For internal use only.
We will use this to implement the special 'Hide', 'Hide Others' and 'Show All'
items and the 'Services' submenu in the Mac OS application menu.
https://bugzilla.gnome.org/show_bug.cgi?id=720552
By default, Mac OS scans menus as they are opened, updating the
sensitivity of each item in the menu.
The current code in gtkapplication-menu-quartz disables this behaviour,
preferring to manually control the sensitivity of each item in the menu
(when told by the tracker that it has changed internally).
Change the way that this works to more closely follow the usual Mac OS
regime.
This will allow us to construct a typical "application menu" on Mac OS
containing the items that are typically found there ("Hide", "Hide
Others", "Show All", "Services") and have the OS automatically update
their sensitivity.
https://bugzilla.gnome.org/show_bug.cgi?id=720552
Add a custom title had the side-effect of showing the widget.
That is not right, adding children and managing their visibility
should be independent. The headerbar size allocation code also
made the assumption that a custom title is always visible.
With these changes, GtkHeaderBar should be usable in situations
where the centering functionality is not required, and it is
important to freely pack content at both ends, such as in nautilus.
https://bugzilla.gnome.org/show_bug.cgi?id=722340
Use a GtkHeaderBar for the credits and about buttons.
It makes less sense here than in other places to go back to
the buttons on the bottom, considering we only have a close
button, so we always use a header bar.
https://bugzilla.gnome.org/show_bug.cgi?id=720059
This commit introduces a private convenience API that derived
dialogs can call in their instance init. This is necessary to
make the setting work as intended in the face of 3rd party
dialogs derived e.g. from GtkFileChooserDialog, which are
created with g_object_new.
This setting will let us keep traditional appearance
of dialogs on platforms where this is expected.
The new setting is called gtk-dialogs-use-header, backed
by the Gtk/DialogsUseHeader xsetting.
This change makes it possible for GtkDialog to pack
its action widgets into a header bar, instead of the
traditional action area. This change is controlled
by the use-header-bar construct-only property.
https://bugzilla.gnome.org/show_bug.cgi?id=720059
Stop trying to deal with "theoretical possibilities".
We can't possibly continue to be a faithful GActionGroup implementation
across dispose because dispose has a side effect of removing everyone's
signal handlers.
The code that we ran after the dispose chainup to do all of the fancy
signal emulation was therefore dead. The test that aimed to verify this
was buggy itself due to an uninitialised variable, so really, it never
worked at all.
We keep the re-ordering of the chainup from the original commit to avoid having
trouble with GtkActionMuxer and keep the checks in place that will prevent an
outright segfault in the case that someone else tries to use the interface
post-dispose.
https://bugzilla.gnome.org/show_bug.cgi?id=722189
With proper notifications, plus an accessor method for that state. This
allows client to just listen to notify::is-maximized instead of tracking
window-state-event.
https://bugzilla.gnome.org/show_bug.cgi?id=698786
This leads to disastruous results, since each menu is itself
in a GtkWindow, so holding down the menu key leads to a neverending
cascade of menus on top of menus.
https://bugzilla.gnome.org/show_bug.cgi?id=722106
Theming code gets confused when computing the spacing for 0px wide dots
and then divides by 0. And then cairo complains and stops drawing
anything forever out of spite and then we end up with a single color
screen.
https://bugzilla.gnome.org/show_bug.cgi?id=721800
Rename testsrules_msvc.mak to detectenv_msvc.mak and remove some package-
specific stuff from it, to reflect on the nature that this NMake Makefile
is shared.
The current approach of building the introspection files for GTK works, but
is often cumbersome as one needs to set many environmental variables before
launching a solution file, which runs a Windows batch script to generate
the .gir/.typelib files. It was also possible to hand-run the batch script
from the Visual Studio command prompt, but even more environmental
variables need to be set.
This changes the approach to build the introspection files using an NMake
Makefile (but elimating from the Visual Studio Project Files the part to
build the introspection files) to:
-Make it clearer to the person building the introspection files what
environmental variables are needed, specifically for PKG_CONFIG_PATH and
MINGWDIR and CFG (formerly CONF). Setting stuff like VSVER, PLAT and BASEDIR
is no longer required, which was a bit clunky.
-Allows some more easier flexibility on the build of the intropsection files.
Make sure the needed public headers for GTK master is "installed", and re-
order some items so that it is easier when the headers lists are
automatically acquired from the various Makefile.am's.
The EWMH defines _NET_WM_MOVERESIZE_SIZE_KEYBOARD and
_NET_WM_MOVERESIZE_MOVE_KEYBOARD for operations that are not
initiated by a button-press event. Allow using these by passing
a button of 0 to gdk_window_begin_move/resize_drag.
The window-dragging code had a number of issues: The code was
starting a drag on every button press, never bothering to cancel
them. This leads to the odd hand cursor occurring between the two
clicks to maximize. We relied on GDK's multi-click detection, which
gives us triple-clicks when we really want sequences of double-clicks.
Lastly, we didn't propery restrict double-click handling to the primary
button, so e.g. if you had a window on an empty workspace, double-right
click on the titlebar would maximize it, which is not intended.
This commit solves all three problem by a doing our own double-click
detection, and only starting a drag when the pointer goes out of
'double-click range'. We change the way dragging is implemented for
menubars and toolbars to just letting events bubble up, so they
get the same behaviour as the titlebar. To make this work, we
have to select for pointer motion events in a few more places.
Two changes that sneaked in during the GtkApplication port
made it so that the window would not let you shrink it again
after you've made it larger. This also yielded very surprising
results when unmaximizing the window: it would come back to
have a minimum width slightly larger than the screen, making
maximization fail from then on.
The behaviour of gtk_text_view_add_child_in_window() used to be
quite broken. It scrolled with the window during scrolling, then
jumped to the absolute position when the widget resized. Furthermore,
in 3.10 we broke the first feature, making it always be fixed.
The "proper" way to handle this is to always follow scrolling. This
is what the only user so far (gedit) wants, and if you want some
kind of overlay you should use GtkOverlay instead.
So, this changes the behaviour to something that is internally consistent
and works. I.e. all added widgets scroll with the textview as needed.
https://bugzilla.gnome.org/show_bug.cgi?id=711826
Cook up some silly cases to test out the hidden-when='' attribute.
- make sure hidden-when='action-missing' shows/hides items based on
actions being created and destroyed
- make sure hidden-when='action-disabled' shows/hides items based on
actions being enabled and disabled
- make sure hidden-when='action-missing' doesn't hide items when the
action is merely disabled
https://bugzilla.gnome.org/show_bug.cgi?id=688421
Modify the tracker so that it manages the visibility of
GtkMenuTrackerItem by issuing insert and remove callbacks to the
user of the tracker.
This works by treating the GtkMenuTrackerItem as a virtual section which
contains 1 item when the item is visible and 0 items when it is hidden.
For efficiency reasons, we only employ this trick in the case that the
item has a hidden-when='' attribute set on it.
https://bugzilla.gnome.org/show_bug.cgi?id=688421
Add an internal API for checking if a GtkMenuTrackerItem is visible,
along with a signal for reporting changes in that flag. The item will
become invisible in situations according to the new hidden-when=''
attribute, which can be set to 'action-disabled' or 'action-missing'.
This new flag doesn't actually do anything yet, and none of the
consumers of GtkMenuTracker do anything with it (nor should they). A
followup patch will address the issue.
https://bugzilla.gnome.org/show_bug.cgi?id=688421
Refactor the code in the action observer remove function in order to
make way for the (efficient) handling of hiding of the item in the case
that hidden-when='' is given.
https://bugzilla.gnome.org/show_bug.cgi?id=688421
Strictly speaking, can_activate should always be set back to FALSE when
the action disappears from the muxer (since we can't activate it
anymore) but we forgot to do that.
This 'bug' could never cause a problem because 'can_activate' is never
directly queried for anything at all and the item would get marked
insensitive anyway. As soon as the action was re-added, can_activate
would be recalculated based on the new action before anything else could
happen.
All the same, this should be cleared here.
https://bugzilla.gnome.org/show_bug.cgi?id=688421
Remove a hash lookup from the separator sync logic (which is run every
time we change a menu). Instead, we do the lookup when creating the
section and cache the result.
This refactor will also help us in a future commit to add support for
hiding menu items based on missing actions.
https://bugzilla.gnome.org/show_bug.cgi?id=688421
This adds save/restore calls to the clear-to-transparent call in
the pixel cache, to avoid changing the default color of the
cairo_t. It also removes a call set_operator call that is no longer
necessary (it was trying to manually restore the state).
https://bugzilla.gnome.org/show_bug.cgi?id=721480
The signal callbacks are defined to take pointers as their arguments, but the
callbacks found in testsuite/gtk/builder.c are passing a GParamSpec by value
as the second argument. This confuses and angers the compiler on ppc64el,
resulting in segfaults after return from the function due to stack-smashing
by the (completely-unused) argument.
https://bugzilla.gnome.org/show_bug.cgi?id=721700
GtkApplicationWindow frees its internal action group on dispose for the
usual reasons: to avoid the possibility of reference cycles caused by
actions referring back to the window again.
Unfortunately, if it happens to be inside of a GtkActionMuxer at the
time that it is disposed, it will (eventually) be removed from the muxer
after it has been disposed. Removing an action group from a muxer
involves a call to g_action_group_list_actions() which will crash
because the internal action group to which we normally delegate the call
has been freed.
A future patch that reworks the quartz menu code will introduce a use of
GtkActionMuxer in a way that causes exactly this problem.
We can guard against the problem in a number of ways.
First, we can avoid the entire situation by ensuring that we are removed
from the muxer before we destroy the action group. To this end, we
delay destruction of the action group until after the chain-up to the
dispose of GtkWindow (which is where the window is removed from the
GtkApplication).
Secondly, we can add checks to each of our GActionGroup and GActionMap
implementation functions to check that the internal action group is
still alive before we attempt to delegate to it.
We have to be careful, though: because our _list_actions() call will
suddenly be returning an empty list, people watching the group from
outside will have expected to see "action-removed" calls for the
now-missing items. Make sure we send those. but only if someone is
watching.
https://bugzilla.gnome.org/show_bug.cgi?id=710351
A widget intended to offer contextual actions for a given view.
It allows packing children into the start or end as well as offering
a single centered child box.
https://bugzilla.gnome.org/show_bug.cgi?id=721665
Notifications can only be associated with application actions,
but clear is a window action. Introduce a "clear-all" action
that forwards to clear on all windows.
https://bugzilla.gnome.org/show_bug.cgi?id=721633
Add gtk_tree_path_new_from_indicesv which takes an array of
integers with a length. Use "Rename to" annotation to rename the
method as gtk_tree_path_new_from_indices. This is needed because
the original method takes variadic arguments which is not supported
by introspection.
https://bugzilla.gnome.org/show_bug.cgi?id=706119
GtkMenuItemAccessible was assuming that an accel label is
always the immediate child of a menu item. It also did not
deal with manually set accels. Fix both of these.
https://bugzilla.gnome.org/show_bug.cgi?id=721086
Improve the color swatch accessible to set a proper role
depending on whether the swatch is selectable or not, and
set the checked atk state when appropriate.
https://bugzilla.gnome.org/show_bug.cgi?id=721048
The initial state of GtkModelMenuItem is that of a normal menu
item, but the accessible we are using, GtkCheckMenuItemAccessible,
starts out with a role of 'check menu item'. Fix that up by
explicitly setting the initial accessible role.
GtkModelMenuItem does not emit the ::toggled signal when a radio
item is activated, so listen for property notification for that
property. We still keep the ::toggled signal handler, in order
to not break other uses of check and radio menu items.
https://bugzilla.gnome.org/show_bug.cgi?id=720983
GtkModelMenuItem emits no property notification, since none of its
properties are readable. But the toggled property is just a proxy
for GtkCheckMenuItem::active, so we should ensure that property
notification is emitted for the ::active property.
Ensure the hscrollbar & vscrollbar at gtk_scrolled_window_add() time,
this allows one to subclass GtkScrolledWindow with templates and add
children, as this will happen at instance initialization time before
the construct adjustment properties take effect.
And update the colorchooser a11y test to a) use GtkColorChooser
instead of the deprecated GtkColorSelection and b) match this
change. Pointed out in
https://bugzilla.gnome.org/show_bug.cgi?id=721045
fvwm seems to have problems keeping _NET_WORKAREA in sync with
the number of desktops. Instead of reading garbage, silently use
the full screen as workarea for desktops that are not covered
by the _NET_WORKAREA property.
https://bugzilla.gnome.org/show_bug.cgi?id=698248
This commit makes the label accessible implement AtkHypertext,
which returns a AtkHyperlink object for each link in the text.
At the same time, add AtkHyperlinkImpl objects as children
to the label accessible.
Also some private API to indicate that links have changed, and
call that from GtkLabel when needed.
Adjust expected output of the affected a11y tests.
https://bugzilla.gnome.org/show_bug.cgi?id=721410https://bugzilla.gnome.org/show_bug.cgi?id=721421
The get_end_index implementation was casting the accessible
to AtkText even though GtkLinkButtonAccessible does not
implement this interface. This did not show up in the a11y
tests because the they were not dumping the affected AtkHyperlinkImpl
properties. Oops.
Ignacio Casal Quinteiro reported a problem whereby an empty section at
the start of a menu has a separator placed after it. This was caused by
the implementation of the logic that separators should be inserted at
the top of all non-empty sections that are not the first section. This
logic is obviously incorrect in the case that the first section is empty
(in which case we would not expect to see a separator at the top of the
second section).
Change the logic so that we only insert separators when we see a
non-zero number of actual items in the menu before us.
https://bugzilla.gnome.org/show_bug.cgi?id=721119
When not doing cross-builds, use the values of CFLAGS, CPPFLAGS and
LDFLAGS as the default value for CFLAGS_FOR_BUILD, CPPFLAGS_FOR_BUILD
and LDFLAGS_FOR_BUILD, respectively.
This avoids having to manually specify these variables in order to get
extract-strings to build properly.
This should really be handled by ax_prog_cc_for_build.m4. That has been
reported upstream. This is a workaround for now.
https://bugzilla.gnome.org/show_bug.cgi?id=721346
If the platform data passed with actions and activations includes
a startup notify ID, we should read it and pass it down to GDK.
This ensures that the right startup notify is completed after the
signal emission, and that the user time of the GdkDisplay is properly
updated (which in turn makes sure the windows are not subjected
to focus-stealing-prevention)
https://bugzilla.gnome.org/show_bug.cgi?id=721304
Add a utility project to get config.h and gdkconfig.h from their *.h.win32
(or win32_broadway, if applicable) counterparts, using custom build rules,
so that these "generated" files can also be removed on clean and
"regenerated" upon update. This also enables the removal of configs in
certain projects that isn't really needed as a result.
Also update and merge the projects and property sheets to include a single
property sheet that it needs, which will then in turn include the other
property sheets that is needed, so that things are cleaner.
Add a utility project to get config.h and gdkconfig.h from their *.h.win32
(or win32_broadway, if applicable) counterparts, using custom build rules,
so that these "generated" files can also be removed on clean and
"regenerated" upon update. This also enables the removal of configs in
certain projects that isn't really needed as a result.
Also update and merge the projects and property sheets to include a single
property sheet that it needs, which will then in turn include the other
property sheets that is needed, so that things are cleaner.
Updates to the Visual Studio 2010 projects will follow later.
This header, which is not universally available, is accidently made to be
included unconditionally during the refactoring of gtkapplication.c,
so restore the #ifdef check.
Applications need a way to fix or adapt the decoration layout,
for situations like split header bars. Setting the layout from
the theme with a style property did not offer a good way to do
this, and the ::show-close-button property does not provide
fine-grained control.
To improve the situation, move the layout string to a property of
GtkHeaderBar which is backed by a setting. This allows platforms to
set a default button layout independent of the theme, while applications
can override the default.
The style GtkWindow style property is now deprecated and ignored.
The button now claims its menu as a child for a11y purposes,
which makes it possible for ATs to see it when the navigate
the tree top-down.
Update the a11y test to match.
We go to extra length to set the desktop_uri to NULL when
desktop == home, but then we were adding the (non-functional)
place item anyway. Don't do that.
A number of C99 math.h functions, along with inttypes.h were included for
Visual C++ 2013, along with much improved C99 capabilities, so update
config.h.win32.in to reflect this.
Fixes a tiny typo in commit f51c9d4154
which manifested itself in GtkSpinButton's panels being drawn with an
incorrect, not updated state.
This patch took me more hours than you might think! :P
https://bugzilla.gnome.org/show_bug.cgi?id=709491
The reason why some of the strings in gtkprintunixdialog.ui
were missing is that we did not extract translatable string
from <item> elements. Fix that.
gtkapplication.c has turned into a bit of an #ifdef mess over time, and
many of the current checks are incorrect. As an example, if you build
Gtk for wayland, and exclude the X11 backend, much of the functionality
required by wayland (such as exporting menu models) will be disabled.
Solve that by introducing a backend mechanism to GtkApplication (named
GtkApplicationImpl) similar to the one in GApplication. Add backends
for Wayland, X11 and Quartz, with X11 and Wayland sharing a common
'DBus' superclass.
GtkApplicationImpl
|
/--------------+-------------------\
| |
GtkApplicationImplDBus GtkApplicationImplQuartz
|
/-----------+-----------------\
| |
GtkApplicationImplX11 GtkApplicationImplWayland
GtkApplicationImpl itself is essentially a bunch of vfuncs that serve as
hooks for various things that the platform-specific backends may be
interested in doing (startup, shutdown, managing windows, inhibit, etc.)
With this change, all platform specific code has been removed from
gtkapplication.c and gtkapplicationwindow.c (both of which are now free
of #ifdefs, except for a UNIX-specific use of GDesktopAppInfo in
gtkapplicationwindow.c).
Additionally, because of the movement of the property-setting code out
of GtkApplicationWindow, the _GTK_APPLICATION_ID properties (and
friends) will be set on non-GtkApplicationWindows, such as dialogs.
https://bugzilla.gnome.org/show_bug.cgi?id=720550
Since update_windows list is a static variable in GdkWindow.c which
contains pointers to windows which needs to be updated, it can happen
that it contains a pointer to a window even after quit from a gtk_main().
If another gtk_main() is called in the same process it tries to process
windows in the list which leads to a crash.
Correct reference count handling of added windows prevents such applications
from crash.
https://bugzilla.gnome.org/show_bug.cgi?id=711552
We don't want the maximum size to be smaller than the minimum size. Not
just because it's wrong but also because when this happens the rest of
GTK gets mighty confused and infloops resizing to min-size and
max-size in turns causing a flickering window. Well, at least if you
run X without a window manager. Or your window manager hasn't finished
starting up.
Private RHEL bug finding this issue:
https://bugzilla.redhat.com/show_bug.cgi?id=1035409
This could happen if data was requested from a separate screen now that
multi-screen is no longer supported.
Ideally, we'd want to support copying to other screens, but that
requires solving in GDK as that's X-specific so cannot be well
abstracted by GDK (without the reintroduction of multiple screens).
https://bugzilla.gnome.org/show_bug.cgi?id=719314
Both GtkApplicationWindow and GtkHeaderBar listen for changes
of the gtk-shell-shows-app-menu setting, so they need to somehow
coordinate who is going to take action and show a fallback.
We prefer the menu button in the title over the menubar, so
let GtkApplicationWindow opt out if it finds that the header bar
has been configured to show window controls.
If we don't have a window icon, we hide the titlebar_icon,
we still add it, so we can't simply go by the number of
children when deciding whether to show the separator or
now. Instead, update the separator visibility as we create
the various buttons.
And deprecate the X11-specific version of it.
We call this new API _set_shadow_width() and not _set_frame_extents()
because we already have a gdk_window_get_frame_extents() with a
different meaning and different type of value.
https://bugzilla.gnome.org/show_bug.cgi?id=720374
Turn this into a GtkApplication with an app menu.
Allow to override the gtk-shell-shows-app-menu setting
and the decoration-button-layout style property.
This is one of the few cases where it makes some sense to blur
the line between and empty string and NULL: without this, it is
hard to reset the subtitle e.g. from a builder file. And we
have the has-subtitle property now to enforce subtitle size
allocation independently.
Only fill the location entry with the file name of the tree view's
selected file when the selection was done by the user.
When the file chooser's action is GTK_FILE_CHOOSER_ACTION_OPEN, it
selects the first file in the tree view once loading has finished. For
this case we don't want it to insert the file name in the location
entry, as it hinders efficient navigation using the location entry. To
achieve this, use a priv flag to keep track of whether the
selection-changed signal was caused by the file chooser itself.
https://bugzilla.gnome.org/show_bug.cgi?id=386569
Now that the nice titlebar example is in gtk3-demo, we can
use testtitlebar as an actual testbed for headerbar stuff.
This immediately reveals size allocation issues when titlebar
widgets change size.
It is a fairly common case to just want a title, and not
reserve extra space for a subtitle. This is much easier
to get right by setting a boolean property than by
constructing a custom title widget.
When setting a custom titlebar that happens to be a GtkHeaderBar,
we connect to notify::title to pick up title changes on the headerbar,
but we forgot to sync the title initially. Fix that.
You can still hover a mouse on insensitive elements; it's up to the
theme to disable that.
This is in line with the HTML/CSS interpretation of :hover.
Insensitive elements still cannot be clicked.
https://bugzilla.gnome.org/show_bug.cgi?id=719486
Failing to load a thumbnail returns a NULL pixbuf. Since the hidpi
patches this wasn't checked when creating the surface. Result: assertion
failure.
https://bugzilla.gnome.org/show_bug.cgi?id=719977
This demo condenses the essentials of advanced management of
input events. Depending on the information available in input events,
this demo will try to represent as much information as possible for
those.
https://bugzilla.gnome.org/show_bug.cgi?id=719987
The focus handling code is shared between core and XI2 implementations,
so just handle the extra XI2 types for passive grabs. Those must be dealt
with in the same way than active grabs. Focus events with this crossing
mode could happen currently through the XIGrabFocusIn passive grab.
https://bugzilla.gnome.org/show_bug.cgi?id=719762
This fixes potential assertions if a GTK+ app gets to receive
a XINotifyPassiveGrab/Ungrab pointer crossing event, currently
triggerable by XIGrabEnter passive grabs.
http://bugzilla.gnome.org/show_bug.cgi?id=719762
The bin window's background would have to be drawn in the bin window's
size and inside the pixel cache draw function to not cause transparency
issues.
But because it's unnecessary as the view window draws the same
background, we just skip it.
https://bugzilla.gnome.org/show_bug.cgi?id=709027
...for all files except the README.txt and the .sln files, which have to
have DOS/Windows line endings. This makes application of patches, when
applicable, easier.
Make sure that the tests don't access the host's session bus or
installed gsettings schemas.
Also disable tests for some classes that leak a connection to the
session bus.
https://bugzilla.gnome.org/show_bug.cgi?id=711715
This allows themes do whatever they want as separators, with
paddings, borders and backgrounds.
If "wide-separators" property is true, then, instead of just draw
a frame, also render its background, and take into account the
padding property for its limits.
https://bugzilla.gnome.org/show_bug.cgi?id=719713
g_file_new_for_uri() is guaranteed to return a non-NULL value, so this
check was redundant, and was confusing the static analyser into
returning a false positive, where it thought the file could be NULL.
https://bugzilla.gnome.org/show_bug.cgi?id=712760
At this point, segments[1] is always uninitialised, and is used to
initialise itself. Looking at the code in the branch above, this appears
to have been a typo from segments[0], as segments[1] seems to typically
be 2 * segments[0].
Found by scan-build.
https://bugzilla.gnome.org/show_bug.cgi?id=712760
When animation is disabled, we use 'none' as the effective
transition type. So far, this transition type failed to change
the size request, causing the revealer to always take up the
space of the child, even when the child is not shown.
https://bugzilla.gnome.org/show_bug.cgi?id=719568
Use (out caller-allocates) annotation for the "value" argument to
GtkTreeModelFilterModifyFunc. This is needed because GI based language
bindings coerce GValue input args into native types and there is no
opportunity to set the value within the GValue itself.
https://bugzilla.gnome.org/show_bug.cgi?id=719460
The call to gtk_button_set_relief() in gtk_toolbar_init() indirectly
used the style context of the half-created widget, before we had a
chance to add the "toolbar" style class to it.
Reorder gtk_toolbar_init() to ensure that the proper style class is
set first.
https://bugzilla.gnome.org/show_bug.cgi?id=719595
When enable-animations is false, the revealer's child-revealed property is
notified immediately, so make sure to connect to it before toggling the
revealer.
https://bugzilla.gnome.org/show_bug.cgi?id=719510
Win32 does not have alpha channel currently ; fix the check
for this, so trying to enable CSDs on this platform will
not "succeed" and crash the app anymore.
Partially fixes gtk3-widget-factory.
Since commit 7c2a5072 the gtkdbusgenerated.[c|h] are not included in the
dist tarball and thus have to be generated, which broke the Visual C++
builds.
This patch adds property sheets and custom build rules for the Visual C++
projects so that gtkdbusgenerated.[c|h] will be generated upon building the
GTK+ DLL sources.
This also tells people building GTK+ from the projects that they need to
have Python 2/3 installed and the Python interpretor needs to be in their
PATH before building GTK+ from the projects.
-Improve optimization a bit for broadwayd, by enabling link time code
generation
-Add PlatformToolset tag for the Visual C++ 2010 projects, to ease
transition to Visual C++ 2012/2013
Ignore the "show-desktop" property on GtkPlacesSidebar for the
defaultvalue test.
Currently, "make check" is passing because it runs the test under a xvfb
with no XSETTINGS provider, so we see the Gtk default value. No matter
what we set the default value to in Gtk, however, there will be some
desktop environment in which someone running the installed test outside
of an xvfb will get the wrong result. Best to ignore it.
https://bugzilla.gnome.org/show_bug.cgi?id=712302
Change the GtkSettings default for "shell-shows-desktop" back to TRUE
and also change the default value of the "show-desktop" property on
GtkPlacesSidebar so that the defaultvalue test passes.
https://bugzilla.gnome.org/show_bug.cgi?id=712302
This is so we always have the latest information given by XRandR (or other), and not
rely on Core protocol information that might not have been updated yet.
This is specially visible when a monitor is connected (less frequent) or disconnected
(much more frequent), callbacks on GdkScreen::monitors-changed that call
gdk_screen_get_width/height() could get the screen size previous to the monitor
rearrangement.
So in order to fix this, keep track of the latest monitors information, and calculate
the bounding box in order to know the screen size.
https://bugzilla.gnome.org/show_bug.cgi?id=715029
Return values of g_variant_get_child_value() were not unreffed
correctly together with one value returned by g_variant_get().
Use g_variant_get_data() instead of copying each byte separately.
https://bugzilla.gnome.org/show_bug.cgi?id=712799
Instead, use the monitor's work area.
This might have unforseen side effects that warrant a later revert, such
as:
- Apparently some WMs assume maximizing when a window is maximum screen
size.
- WMs might not shrink the window by the decorations' size when it tries
to be fullscreen.
- Applications might have buggy size request code that causes weirdly
sized windows.
Scroll valuators were being just appended again and again, leading
to 1) a growing memory issue anytime a device changed 2) the first
scroll valuators to stay permanent on the application lifetime, as
the first stored valuators would always match.
https://bugzilla.gnome.org/show_bug.cgi?id=705203
Use info available in Avahi TXT records for creation of gtk printer
and request details when needed (through gtk_printer_request_details()).
If there is a printer advertised on Avahi by a remote CUPS server
try to get its PPD file at first or get its capabilities through an IPP
request if it fails.
https://bugzilla.gnome.org/show_bug.cgi?id=712751
When downscaling images, Cairo apparently uses algorithms different
enough to make this test trip over. So add the downscaled image as the
reference instead of downscaling the previous reference image.
Fixes the border-image reftest. For real now.
The new downscaling code in Cairo doesn't allow this test, so we remove
the CSS that made the border downscaled.
So the test does test less now, but it still tests the repeat modes of
border images.
Passive grabs may take pointer focus out of the application, even though
the pointer didn't leave the window, but those events still trigger resetting
of the scroll axes. This is most visible with compiz, and possibly other
reparenting WMs, where passive grabs happen on the WM-managed window that
is a parent of the application toplevel.
As it is not possible to have scrolling happening on the timespan a passive
grab takes action, it is entirely safe for GTK+ to assume none happened if
it gets a crossing event of that nature.
https://bugzilla.gnome.org/show_bug.cgi?id=699574#c33
One requirement of .ui files is that each object must have an ID,
even if it is never referred to or directly loaded from the code.
This makes editing .ui files much more onerous than it has to be,
due to the frequent need to invent new IDs, while avoiding
clashes.
This commit makes IDs optional in the XML. They only need to
be provided for objects which are referred to or explictly loaded
from the code. Since GtkBuilder needs IDs for its own internal
accounting, we create IDs of the form ___object_N___ if not
specified in the XML.
https://bugzilla.gnome.org/show_bug.cgi?id=712553
This information will be useful in case someone stumbles on a situation
similar to https://bugzilla.gnome.org/show_bug.cgi?id=699574, so we can
figure out where do the crossing events come from or go to easily.
It's been reported in several applications that scrolling feels jerky
since commit cc7b3985b3.
Investigation reported that the combination of passive 4-7 button grabs
on the toplevel and the presence of native subwindows might trigger
too often crossing events from the child window to the toplevel and
back as scroll "buttons" trigger the passive grab. Those crossing events
would reset the scroll valuators rendering scrolling from jerky on
touchpads (where there's intermediate smooth events between the emulated
button ones) to ineffective on regular mouse wheels (where the crossing
event would reset the valuators right before the single smooth scroll
event we get is delivered)
So, only reset scroll valuators when the pointer enters the toplevel
(we only care about this when the pointer is on the window after it's
been possibly scrolling somewhere else), and it doesn't come from an
inferior.
The situations where this happened varied though, the native subwindow
could be one created explicitly by the application, or created indirectly
through gdk_window_ensure_native(). The latter was mainly the case for
evolution (through gtk_selection_set_owner()) and any GtkScrolledWindow
under the oxygen-gtk3 theme (through gdk_window_set_composited())
https://bugzilla.gnome.org/show_bug.cgi?id=699574
When the menubar inserted by GtkApplicationWindow is the widest
widget in a csd window, its allocation gets cut short. Fix this
by taking the decoration size into account while calculating
the size request (it is implicitly taken into account in the
size allocation phase by _gtk_window_set_allocation).
Do the menubutton for app menu fallback ourselves in GtkWindow
for the csd, non-custom titlebar case. This fits better with
the way we handle other title buttons. Themes have control
over the placement of this button by placing menu in the
decoration-button-layout style property.
Make the sunny example useful by giving it a header bar
with app menu fallback. To test this under gnome-shell,
set APP_MENU_FALLBACK=1 in the environment.
Allow showing the fallback app menu with a menu button
in the header bar. Applications have to explicitly enable
this by calling gtk_header_bar_set_show_fallback_app_menu.
GtkAboutDialog highlights emails written as <...> and
urls written as http://... . gnome-terminal manages to
put <http://...> into its license text, which sadly
confuses the parser into running evolution on http://...
Fix things up far enough that <http://...> is now
recognized as url, and only the part inside the <> is
underlined (for email addresses, we include the <> in
the underline).
Commit 719dd636a9 replaces
margin-left/right with margin-start/end. CSS does not have
margin-start/margin-end properties, the sed script was a bit overeager.
Fwiw, CSS implements RTL margin styling via :dir(rtl) selectors.
Add margin-{start,end} and gtk_widget_{get,set}_margin_{start,end}
and drop margin-{left,right} and gtk_widget_{get,set}_margin_{left,right}.
margin-{start,end} handle right also in RTL.
https://bugzilla.gnome.org/show_bug.cgi?id=710238
Add a GtkSetting for whether the desktop shell is showing the desktop
folder icons.
This is on by default because most desktop shells do show the icons on
the desktop. We already have a patch in gnome-settings-daemon to bind
this to the org.gnome.desktop.background show-desktop-icons GSettings
key which is off by default on GNOME.
https://bugzilla.gnome.org/show_bug.cgi?id=712302
Printing a file with to PDF/etc. with the virtual printer option doesn't add
the file to the list of recently used files. It should be there, so I can
easily access it.
https://bugzilla.gnome.org/show_bug.cgi?id=668598
We add a custom im module for broadway that calls some broadway
specific APIs to show/hide the keyboard on focus in/out. We then forward this
to the browser, and on the ipad we focus an input field to activate
the keyboard.
gtk_menu_tracker_add_items() fetched the action-namespace from the menu
item, but didn't pass it into gtk_menu_tracker_section_new() when its
internal namespace was still NULL.
https://bugzilla.gnome.org/show_bug.cgi?id=712164
The broadway backend would move the focus from one window to another based on
where the mouse was (i.e. 'focus-follows-mouse' approach). Handling the focus
this wait didn't play well with widgets which rely on focus-in-event and
focus-out-event, like the GtkEntry when using a completion popup window, see
e.g:
https://bugzilla.gnome.org/show_bug.cgi?id=708984
So instead, setup broadway to require a click in a window to move the focus
(i.e. 'click-to-focus' approach):
* The implicit GDK_FOCUS_CHANGE events that were generated upon reception of
BROADWAY_EVENT_ENTER or BROADWAY_EVENT_LEAVE are removed.
* The broadway daemon will now keep track of which is the focused window
* Whenever the daemon detects an incoming BROADWAY_EVENT_BUTTON_PRESS, it will
trigger the focused window switch, which sends a new BROADWAY_EVENT_FOCUS to
the client, specifying which windows holds the focus.
* Upon reception of a BROADWAY_EVENT_FOCUS, the client will generate a new
GDK_FOCUS_CHANGE.
* gdk_broadway_window_focus() was also implemented, which now requests the
focus to the broadway server using a new BROADWAY_REQUEST_FOCUS_WINDOW.
This is based on an initial patch from Aleksander Morgado <aleksander@lanedo.com>.
If a motion event handler (or other handler running from the flush-events
phase of the frame clock) recursed the main loop then flushing wouldn't
complete until after the recursed main loop returned, and various aspects
of the state would get out of sync.
To fix this, change flushing of the event queue to simply mark events as
ready to flush, and let normal event delivery handle the rest.
https://bugzilla.gnome.org/show_bug.cgi?id=705176
Don't recurse the mainloop in _gtk_tree_view_column_start_drag().
It doesn't serve any discernible purpose, and recursing the
mainloop from the flush-events phas of the frame clock breaks
frame synchronization with mutter.
https://bugzilla.gnome.org/show_bug.cgi?id=705176
If a queue_redraw() (invalidating a region, or the whole widget) was
called from the draw() call, it could get ignored if surface_dirty
existed, as it would then be updated, but destroyed right at the end of
the _gtk_pixel_cache_repaint(), leading the next call to
_gtk_pixel_cache_draw() have its call to repaint() be a no-op
(since there's no surface_dirty) and then simply draw from (non
updated) surface.
Signed-off-by: Olivier Brunel <jjk@jjacky.com>
https://bugzilla.gnome.org/show_bug.cgi?id=711545
Arrange for the scales to control the speed of the activity
mode progress bar in gtk3-widget-factory. This will be useful
in demonstrating smoother progress bar animation in the future.
Add an application shortcut for the Music folder in testfilechooser.
The file chooser used to filter duplicates out, but this broke
in the sidebar merge.
currently it's using the same sizes for natural and minimum, but it
happens that, when it's allowed to use the arrow, the minimum size
can be smaller than natural.
https://bugzilla.gnome.org/show_bug.cgi?id=693227
Although I can't find explicit documentation for clipboard pointer, it
seems to be possible to modify clibpoard memory without side-effects.
According to MSDN,
http://msdn.microsoft.com/en-us/library/windows/desktop/aa366596%28v=vs.85%29.aspx
"The global and local functions are supported for porting from 16-bit
code, or for maintaining source code compatibility with 16-bit
Windows. Starting with 32-bit Windows, the global and local functions
are implemented as wrapper functions that call the corresponding heap
functions using a handle to the process's default heap."
"Memory objects allocated by GlobalAlloc and LocalAlloc are in private,
committed pages with read/write access that cannot be accessed by other
processes. Memory allocated by using GlobalAlloc with GMEM_DDESHARE is
not actually shared globally as it is in 16-bit Windows. This value has
no effect and is available only for compatibility. "
https://bugzilla.gnome.org/show_bug.cgi?id=711553
Setting event compression to false will allow inter-frame
mouse motion events to be delivered, which are necessary
for painting applications to produce smooth strokes.
https://bugzilla.gnome.org/show_bug.cgi?id=702392
This is based on the rolling hashes code from
http://cgit.freedesktop.org/~krh/weston/log/?h=remote
It works by incrementally calculating hashes for every 32x32 block
in each frame sent, and then refering back to such blocks when
encoding the next frame. This means we detect when a block matches
an existing block in the previous frame in a different position.
This is great for detecting scrolling, which we need now that
the gdk level scrolling is neutered.
* gtk/gtkprintunixdialog.c (printer_status_cb): Do not reset the
waiting_for_printer on status change as the default printer might
get added later.
https://bugzilla.gnome.org/show_bug.cgi?id=577642
Call gtk_cell_layout_clear() on the area instead of the completion in
gtk_entry_completion_clear_text_column_renderer(), because it is also
called from within gtk_entry_completion_clear().
gtk_entry_completion_set_text_column() always added a cell renderer,
regardless of whether there was an existing one already installed. This
patch reuses an old renderer if it exists, but only if it was added by a
previous call to this function.
To avoid conflicts, all renderers that were added manually are removed
when calling this function. Also, the renderer added by this function is
removed when manually adding new renderers. This effectively gives
GtkEntryCompletion two modes (managed and manual cell renderers) and
allows seamless switching between the two.
This is a minor API break. However, this shouldn't be an issue in
practice as applications couldn't call set_text_column() more than once
because of this bug. Also, it is unlikely that many applications mix
set_text_column() and custom cell renderers. The interaction between the
two modes was erratic and not documented well.
https://bugzilla.gnome.org/show_bug.cgi?id=635499
Call gtk_entry_completion_set_text_column() when setting the
"text-column" property directly.
The completion appeared empty when setting "text-column" directly (for
example from a GtkBuilder file), because the setter creates and adds the
GtkCellRendererText.
https://bugzilla.gnome.org/show_bug.cgi?id=710533
The MINMAXINFO struct was being populated based upon geometry hints when
GDK_HINT_MAX_SIZE flag was enabled, then promptly having its values blown
away with default values.
https://bugzilla.gnome.org/show_bug.cgi?id=711110
_gtk_widget_draw_internal() was clipping by passing the subwindow
sizes as a path to cairo_clip(). This was breaking for windows
larger than 23 bits in width/height, due to cairo using fixed point
(24.8) for the path coordinates.
We fix this by pre-clipping the subwindow region to the existing
cairo clip region in the full 32bit gdkwindow precision. This fixes
the GooCanvas Large Items test.
https://bugzilla.gnome.org/show_bug.cgi?id=710958
I have been convinced that it is a bad idea to change the behaviour
at the same time as deprecating it, so go back to respecting the
Gtk/ButtonImages xsetting in buttons created with
gtk_button_new_from_stock() when it is set.
The setting as well as the function are still deprecated, and the
default value of the setting will remain FALSE.
I have been convinced that it is a bad idea to change the behaviour
at the same time as deprecating it, so go back to respecting the
Gtk/MenuImages xsetting in GtkImageMenuItem when it is set.
The setting as well as the widget are still deprecated, and the
default value of the setting will remain FALSE.
Since 49c4ad2f the order between "Page_Down" & "Next" had been changed (due to
sorting) and this resulted in the Page Down key being identified as "Next"
instead of "Page_Down"
"Page_Up" wasn't affected, since it comes before "Prior"
Signed-off-by: Olivier Brunel <jjk@jjacky.com>
https://bugzilla.gnome.org/show_bug.cgi?id=710411
For some widgets, like GtkTreeView, which setup a clock frame
update during realize, it was possible to call
gdk_frame_clock_begin_updating() twice, but only ever disconnecting
from it once. This happens because the realized flag is set at an
unpredictable time by the GtkWidget's realize implementation.
Keep the signal handler ID from us connecting to the "update" signal
to avoid connecting to it twice.
This fixes high wake-up count from any application using GtkTreeView,
even idle ones.
https://bugzilla.gnome.org/show_bug.cgi?id=710666
This is what we used to get through the Net/FallbackIcontheme
setting. Nobody has ever set this setting to a different value,
and people have come to rely on GTK+ applications getting their
icons this way.
The objective is simplify the semantics of the code so that we don't
need to check for "(iter != NULL && !g_sequence_iter_is_end (iter))"
in the callers.
When doing fallback for symbolic icons, we first shorten
the name at dashes while preserving the -symbolic suffix.
But after exhausting that, we should also try stripping
the suffix.
https://bugzilla.gnome.org/show_bug.cgi?id=708163
When showing a tooltip on the edge of a monitor, the tooltip could be wrongly
placed and be shown going from one monitor to the next.
This happened because the current_window wasn't set visible, and when it wasn't
the returned allocated size would be 1, hence wrong calculations.
Signed-off-by: Olivier Brunel <jjk@jjacky.com>
https://bugzilla.gnome.org/show_bug.cgi?id=698730
Some symbols in the generated Wayland code were getting
decorated with WL_EXPORT, causing them to show up in the
libgdk exports. We don't want that.
https://bugzilla.gnome.org/show_bug.cgi?id=710141
We may get a NULL region passed to the backend, which means
'nothing is opaque'. In that case, don't crash, but pass
the information on to the compositor.
http://bugzilla.gnome.org/show_bug.cgi?id=709854
Rework how accels are handled on GtkApplicationWindow.
Instead of having GtkApplication fill the GtkAccelMap which is then used
by GtkApplicationWindow to create a GtkAccelGroup filled with closures
that is then associated with the window, do it directly.
GtkApplication now keeps a list of accels and their actions.
Accelerators on a GtkApplicationWindow ask GtkApplication to execute the
appropriate action.
This saves a fair bit of complexity and memory use (due to not having to
create all those closures and accelmap entries). The new approach also
supports multiple accels per action (although there is not yet a public
API for it).
This patch (and the ones before) Reviewed and ACK'd by Matthias Clasen.
Add support for pulling the primary accel out of the GtkActionMuxer.
With this change, it is no longer necessary to have the accel=''
attribute hardcoded onto each menu item (and, in fact, it should be left
off if you intend to have support for dynamic accelerator changing).
Specifying accel='' is a good way to force an accelerator not to be
displayed on a menu item.
Store "action and target" format inside each GtkMenuTrackerItem. This
makes action invocation more efficient (no hash table lookups or
allocations) and slightly simplifies handling of action namespace.
More importantly, this will be used when we start to get accels from
GtkActionMuxer.
Reuse the existing infrastructure in GtkActionMuxer for propagation of
accelerator information: in particular, what accel label ought to appear
on menu items for a particular action and target.
This is a good idea because we want accels to travel along with the
actions that they're tied to and reusing GtkActionMuxer will allow us to
do that without creating another hierarchy of a different class for the
sole purpose of filling in accel labels on menu items.
Doing it this ways also allows those who copy/paste GtkActionMuxer to
insert the accels for themselves.
Add a new method on the GtkActionObserver interface to report changes.
This patch introduces a new concept: "action and target" notation for
actions. This format looks like so:
"'target'|app.action"
or for non-targeted actions:
"|app.action"
and it is used over a number of possible alternative formats for some
good reasons:
- it's very easy to get a nul-terminated action name out of this format
when we need it, by using strrchr('|') + 1
- we can also get the target out of it using g_variant_parse() because
this function takes a pointer to a 'limit' character that is not
parsed past: we use the '|' for this
- it's extremely easy to hash on this format (just use a normal string
hash) vs. attempting to hash on a string plus a GVariant
A close contender was to use detailed action strings here, but these are
not used for two reasons:
- it's not possible to easily get the action name or target out of the
strings without more work than the "action and target" format
requires
- we still intend to use detailed action strings on API (since they are
a lot nicer to look at) but detailed action strings can be given in
non-canonical forms (eg: 'foo::bar' and 'foo("bar")' are equivalent)
so we'd have to go through a normalisation step anyway. Since we're
doing that already, we may as well convert to a more convenient
internal format.
This new "action and target" format is going to start appearing in a lot
more places as action descriptions are introduced.
I suspect that nobody is using '|' in their action names, but in case I
am proven wrong, we can always switch to using something more exotic as
a separator character (such as '\x01' or '\xff' or the like).
Previously, GtkWindow would add the "app" action group to its own
toplevel muxer.
Change the setup so that GtkApplication creates the toplevel muxer and
adds itself to it as "app". Use this muxer as the parent muxer of any
GtkWindow associated with the application.
This saves a small amount of memory and will allow for accels to be
propagated from the application through to all of the windows.
With the stock system being deprecated now, we should provide
meaningful accessible names for buttons that are constructed
from icon names or GIcons. This commit reuses the existing
translations.
It is possible that some common icon names are not covered
here because they were not present as stock items. These can
be added to the table later.
Discovered via a crash because b's (dest's) toplevel was NULL;
ensuring that the dest is actually a GdkWindow or setting b to NULL
prevents that path from being taken.
Spinners are essentially animated symbolic icons. We don't really
want them being rendered to arbitrary sizes. Also the current
technique for rendering the activity is not suitable for larger
sizes. Until we have a better technique we should limit the size
to the MENU icon size or exactly twice that size.
The 'direction' parameter to gtk_widget_keynav_failed() is based on
gtk_list_box_move_cursor()'s 'count' parameter. However if the passed
in movement is GTK_MOVEMENT_DISPLAY_LINES, 'count' is modified by
the keynav handling and will always be 0. To avoid messing up the
'direction' parameter, use a local variable for keynav handling and
leave 'count' untouched.
https://bugzilla.gnome.org/show_bug.cgi?id=709687
Previous commit wrongly identified "active" as containing the new
value. Instead, applications must determine the new value to
update the model with manually based on the value currently in the
model.
Clarify that applications are expected to handle updating the model
from the 'active' property, like GtkCellRendererText does for the
'edited' property.
This test demonstrates that various deprecated ways
to construct buttons and menuitems still work as they
are supposed to, including always-show-image functionality.
If a range goes all the way to the edge of the screen then we don't
have any way to activate autoscrolling. By adding a small region
at the ends of the range we can handle this case. This is the same
approach used in treeviews.
If we're going to run off the end due to an invalid message,
we're going to run off the end. We'll protect this by doing
proper bounds checking in the future, but the malloc gives
us nothing for now.
The newly-added gtk_flow_box_child_is_selected() needed to return a
gboolean, so use g_return_val_if_fail() to return FALSE when an invalid
GtkFlowBoxChild* is passed in.
Make flowbox selection more orthodox. Control and Shift now
modify the selection behaviour pretty much in the same way
they do in a tree view, and clicking without modifiers will
clear the selection.
When dealing with touch devices, we treat modifier-less events
as modifying the selection.
This commit also adds a few other selection necessities, such
as catching Escape key presses to cancel rubberband selection,
and handling grab notify.
GtkFlowBox is a container that its children in a reflowing
grid, which can be oriented horizontally or vertically.
It is similar to GtkListBox in that the children can
be sorted and filtered, and by requiring a dedicated child
widget type, GtkFlowBoxChild. It is similar to GtkTreeView
in that is supports a full set of selection modes, including
rubberband selection.
This is the culmination of work that has happened in the
egg-list-box module, and earlier in libegg. The origins of
this code are the EggSpreadTable in libegg, which was written
by Tristan van Berkom. It was moved to egg-list-box and
renamed EggFlowBox by Jon McCann, and I gave it some finishing
touched in the flowbox-improvements branch of that module.
Previously, the "Places" sidebar was populated by the update_places()
call from within gtk_places_sidebar_style_set(). After
742a2f11a9, update_places() is never called
and the sidebar is never populated unless gtk_places_sidebar_add_shortcut()
happens to be called. This commit fixes this by calling update_places()
at the end of gtk_places_sidebar_init().
https://bugzilla.gnome.org/show_bug.cgi?id=709522
The docs for GtkPlug/GtkSocket were not generated if any
of the win32, quartz, wayland backends were enabled. What
we really mean though, is that we want the docs to be generated
whenever GtkPlug/GtkSocket are included in the library, which
is when the x11 backend is enabled.
As long as we are not ready to switch over the default backend,
arrange ./configure without explicit backend options makes the
x11 backend mandatory and the wayland backend optional (depending
on whether we find Wayland dependencies).
http://bugzilla.gnome.org/show_bug.cgi?id=709212
GtkWidget had a hack where if opacity is 0.999 we set up an opacity group when
rendering the widget. This is no longer needed in 3.10, and GtkStack doesn't
use it anymore.
GdStack is using it, so applications should be ported from GdStack to GtkStack
in 3.12.
https://bugzilla.gnome.org/show_bug.cgi?id=703603
Resize modes don't work anymore, both because nobody ever uses them and
because the frame clock changed the way things work quite a bit. So we
don't want to advertise them as a good idea.
https://bugzilla.gnome.org/show_bug.cgi?id=708787
Add a has-tab-gap style property to GtkNotebook so that we can
disable drawing the gap between tabs and the page in the Adwaita
theme without breaking existing themes.
https://bugzilla.gnome.org/show_bug.cgi?id=707920
GtkSourceView draws before chaining upo to GtkTextView and assumes
that this will be visible, but the pixelcache will just overdraw
that with background.
So, we stop drawing the background to the pixel cache and instead
make it an CAIRO_CONTENT_COLOR_ALPHA surface to make the previously
drawn content see through.
This is slower, but more backwards compatible.
https://bugzilla.gnome.org/show_bug.cgi?id=708423
Previously, we were showing and hiding the custom titlebar
widget in response to state changes such as maximization.
Instead, use gtk_widget_set_child_visible() and leave
show/hide to applications. This makes it possible to set
a custom titlebar and hide it, for a titlebar-less appearance.
https://bugzilla.gnome.org/show_bug.cgi?id=707132
When loading a symbolic icon, g_file_get_contents() is currently used
with the icon pathname, to load its SVG data. This won't work when the
icon is not a local file, for instance when a symbolic icon is loaded
from a GFileIcon with a GResource path.
Fortunately GtkIconInfo already holds a GFile, so we can just use
g_file_load_contents() to load the data instead.
https://bugzilla.gnome.org/show_bug.cgi?id=709056
Updated documentation to specify that '0' should be specified if
one does not need to automatically assign a bound child to a public
or private instance member (now that negative values are private
structure offsets).
If we start with a padding of -1 then it can leak out of the size
allocation request for the column when the treeview is empty. The
GtkTreeView will then collect these -1 values and add them together,
returning -n where 'n' is the number of columns.
This is usually not a problem because treeviews tend to be used with a
scrollbar and the width of the scrollbar will be added to this number
bringing it into positive territory again. On Ubuntu, with overlay
scrollbars, this is not the case, however.
https://bugzilla.gnome.org/show_bug.cgi?id=703062
The size of the shadow and invisible borders can (and usually
will) change between backdrop and focused windows, while the
overall window size remains unchanged. This causes the visible
window to visually 'jump'. We can avoid this by always reserving
the maximum of the focused and unfocused border sizes. The code
for positioning the input-only windows making up the invisible
border is adjusted to deal with this. We now always place the
invisible border right outside the visible content, even if the
shadow extends out much farther.
https://bugzilla.gnome.org/show_bug.cgi?id=707524
And use the "header" style class to do that.
This allows themes to set e.g. the background of the tab header
differently.
Themes will need slight adjustment to make things appear
as before.
https://bugzilla.gnome.org/show_bug.cgi?id=643914
This allows subclasses of GtkTextView that require a corresponding
subclass of GtkTextBuffer to automatically do the right thing when
constructed with a NULL buffer. An example of this is GtkSourceView
which requires a GtkSourceBuffer.
https://bugzilla.gnome.org/show_bug.cgi?id=708584
Improve optimization, by re-enabling WholeProgramOptimization but changing
the linker optimization to not drop items that are not referenced in code
(such as compiled gresource sources that are not directly referenced in
code, as they are still needed for the demos to run properly).
The GtkStack and GtkStackSwitcher code did not really
follow GTK+ conventions for includes. Fix that, and also
fix up a case of gpointer vs gpointer* confusion
in gtkstack.c.
This adds new 'over' and 'under' transitions which work by moving
the new page over the previous one, or moving the previous page off
to reveal the new one. We also add an over/under combination that
is going to be used in GtkAboutDialog.
https://bugzilla.gnome.org/show_bug.cgi?id=707187
Some of the labels were not marked as no-show-all. But the
code clearly tries to manage their visibility, so gtk_widget_show_all()
should not affect them.
https://bugzilla.gnome.org/show_bug.cgi?id=681484
The child property is watched by the StackSwicther which in turns sets a
needs-attention css class on the corresponding button, so that the theme
can for instance show a throbbing animation if one of the hidden pages
needs the user attention.
https://bugzilla.gnome.org/show_bug.cgi?id=707153
So far, this is just supposed to be gdk_cairo_set_source_pixbuf().
Note that this is usually not an API guarantee but courtesy to
applications that used these APIs without a gtk_init() call.
https://bugzilla.gnome.org/show_bug.cgi?id=708547
If an icon is in a Fixed or Threshold directory we normally don't
scale it. However, in the case of HiDPI scaling we *do* want to
scale it, to avoid different layouts in Lo/HiDPI. We look up whatever
the size of the icon would have been in LoDPI and scale to that
in the no-scaling case, thus getting the same layout as the
unscaled case.
https://bugzilla.gnome.org/show_bug.cgi?id=708384
GdkPixbuf will fail returning %NULL if we try to scale a pixbuf to (0, 0),
which will then trigger an assertion in gtk_icon_info_load_icon_finish();
we never want a scale of 0, so ensure it is at least 1.
https://bugzilla.gnome.org/show_bug.cgi?id=708384
We need to subtract border_width from the size we're passing to the
children hfw functions as those are added by ourselves.
Fixes the window-border-width.ui reftest.
The headerbar test and the buttons tests changed because we've
recently changed accessible names, descriptions and roles of
GtkHeaderBar and GtkSwitch.
Previously, when you clicked and held the button down on a
GtkExpander's label or disclosure triangle, then moved the mouse
away and released the button, the expander would still activate.
This brings the behavior in line with the more generally expected
behavior, as exhibited by GtkButton for example.
https://bugzilla.gnome.org/show_bug.cgi?id=706345
The surface is destroyed when we hide a window, but
gdk_window_set_opaque_region can be called before the window is
shown again, so we need to ensure the surface exits.
https://bugzilla.gnome.org/show_bug.cgi?id=707328
This allows GTK+ applications with headerbars to fit in
better in platforms that have window controls on the left.
To use this, set -GtkWindow-decoration-button-layout: 'close:'
in the theme.
https://bugzilla.gnome.org/show_bug.cgi?id=706708
At least for header bars, there's often application controls
in this area, which should be included in the focus chain.
We make it so that the initial focus avoids the titlebar,
but tabbing around will eventually get there.
https://bugzilla.gnome.org/show_bug.cgi?id=708067
This reverts commit 7cecc8e524.
It is impossible to use the selection object of the menu while it is
collapsed (collapsed menus deselect everything for a start), so even
though the original patch was correct, the followup issues are too big
to solve this quickly to a release.
https://bugzilla.gnome.org/show_bug.cgi?id=707786
* gtk/gtkwidget.c: drag-leave signal: Document that it is called before
drag-drop.
drag-data-received signal: Document that it is up to the application
to know why the data was requested (e.g. drag motion or drop).
* demos/gtk-demo/toolpalette.c: interactive_canvas_drag_drop():
Do not transform the drop_item created in the drag-motion handler.
Instead caused drag-data-received to be called, remembering why,
and create a new item there.
interactive_canvas_drag_leave(): Remove the idle-handler hack,
now that we do not need to keep the drag-motion drop_item alive until
the drop.
I noticed that this patch was sitting in bug #605611 from 2009
though it had been approved. I do not remember much about why I
created it.
Pass the master device instead if the last slave is NULL. This is
unlikely to happen in most of the cases, but can happen when running
unit tests where there's no pointer interaction to update the last
slave.
https://bugzilla.gnome.org/show_bug.cgi?id=696756
This is the same behaviour as gtk_widget_get_valign, except
we have no gtk_wiget_get_halign_with_baseline, as baselines make
no sense for halign.
Without this some widgets (like e.g. GtkOverlay) crash if you accidentally
set a BASELINE halign.
It tests gtk_text_buffer_paste_clipboard(),
gtk_text_buffer_copy_clipboard() and gtk_text_buffer_cut_clipboard() in
various situations, including when GtkTextTags are applied to the
selection.
The last test didn't pass.
https://bugzilla.gnome.org/show_bug.cgi?id=339539
It is more logical to first delete the selection and then pasting the
text. When the selection and the text contain tags, the new behavior is
more natural.
A segfault in paste_from_buffer() is also avoided. The segfault occurs when
the text to paste is deleted because it is the selection.
https://bugzilla.gnome.org/show_bug.cgi?id=339539
Move GtkWrapMode from GtkTextTag to the GtkTextView section. The wrap
mode property is in the text view.
Links to the "mark-set" and "mark-deleted" signals.
Add a precision about gtk_text_buffer_get_iter_at_line().
Fix typo in gtk_text_tag_set_priority().
https://bugzilla.gnome.org/show_bug.cgi?id=708076
A proper name or label is required. In the same way, if the
widget is labelled, ATs uses to expose both the label and the
name, making the final output not really user-friendly.
https://bugzilla.gnome.org/show_bug.cgi?id=707872
keyboard navigation didn't support activation since we moved
the keynav to the child row widgets. We fix this by adding a
activate signal handler for the row and setting
widget_class->activate_signal to it.
https://bugzilla.gnome.org/show_bug.cgi?id=707778
Like the install projects that were fixed few days ago, the gengir projects
did not have info on the intermediate and output directories as a result of
the split up of the property sheets. Fix this by including the appropriate
property sheet in the gtk-gengir property sheet so that we can avoid
confusing messages from Visual Studio on whether to reload the gengir
project as it was modified, at least on 2008.
This reverts commit 70ac2b24c3.
It turns out the correct fix is to make pango_layout_get_text() not
return NULL. This has been done, so we can drop this patch.
I won't bump the Pango dependency in configure.ac for this as I don't
consider the crash critical enough.
Use the pixelcache rendered area to inform what part of the cache should
be invalidated upon changes to the underlying textlayout.
By rendering the background to the pixelcache, we can avoid the need to
use RGBA content.
Also, we're using the pixel cache on the text windows bin_window (see
gtk_text_view_get_window) so we need to register the invalidation handler
on that, otherwise the region passed to the invalidate handler will get
clipped to the visible region.
https://bugzilla.gnome.org/show_bug.cgi?id=707244
nth-child() is 1-indexed, not 0-indexed.
It doesn't matter for this test really, but better do it right to not confuse
poor developers who wonder why the first image is highlighted when nth-child(0)
clearly states "none".
We added code to look for settings.ini in system config dirs,
and then proceeded to move it to /usr/share/gtk-3.0 :-(. So,
look in that location as well.
We now pass NULL when the current color should be the default value of
the "color" property and we haven't looked up any value yet. This way we
don't need to look it up all the time and more importantly we can
resolve the default color, which is required because it's a
GtkCssColorValue and not a GtkCssRgbaValue.
Fixes assertions triggering at Polari startup.
The version of device scale that landed in upstream cairo
master already inherits the device scale in cairo_create_similar,
so no need to do that in gtk anymore.
The reftest is a bit flaky because it compares cell renderers with
GtkImage and therefor an icon view with a GtkBox, but it's the best I
can come up with.
https://bugzilla.gnome.org/show_bug.cgi?id=702423
Even when not following states, blacklist the states we don't follow
instead of whitelisting the ones we do. This way, we don't accidentally
eat new ones like the text direction flags.
Fixes a bug where text direction wasn't available when rendering stock
items.
https://bugzilla.gnome.org/show_bug.cgi?id=702423
Due to the split up of the property sheets, the install projects did not
have info on the Intermediate and Output Paths, which caused confusing
messages from Visual Studio to show up upon completing build+"install" and
closing Visual Studio on whether to reload the install project, at least on
Visual Studio 2008.
Also clean up the Visual Studio 2008 install project a bit.
Include the property sheet which defines these properties to fix this.
In the gnome-ostree model builddir contains all generated files not in
git (unless the build system explicitly overrides that). Here the
wayland-client-protocol.h was in $(builddir)/wayland, so we need to
find it using our already extant -I$(top_builddir)/gdk, rather than
relying on same-directory lookup.
Add the concept of shell capabilities, which allow the compositor
to advertise support for the app menu and the global menubar,
which are then propagated as GdkSettings.
https://bugzilla.gnome.org/show_bug.cgi?id=707129
If the compositor supports the gtk-shell interface, use it to
export the application ID, dbus name and paths that can be used
for the application menu.
https://bugzilla.gnome.org/show_bug.cgi?id=707129
Call gtk_widget_get_mapped() in a couple of places before looking at the
widget's parent, since it might be set to a widget that has been
finalized, causing an invalid read.
GTK+ tries to automatically assign the best input module based on the
'system locale'. In the specific case of the IME input method, it will
be the default for the whole GTK+ application if the system locale is
either Japanese (ja), Korean (ko) or Chinese (zh). Other defaults are
equally applicable, e.g. if system locale is Catalan (ca), the special
'Cedilla' input module is chosen.
System locale can be changed (e.g. Win7) through the following sequence
(reboot required):
Control Panel
Region and Language
Administrative
Language for non-Unicode Programs
Change system locale...
The problem with this behaviour is that changing the 'default input
language' (e.g. from English to Japanese+IME) doesn't affect the GTK+
application. Therefore, I can have an English system locale (where GTK+
will choose Simple IM by default) but then have Japanese+IME as input
language.
Default input language can be changed (e.g. Win7) through the following
sequence (no reboot required):
System locale can be changed (e.g. Win7) through:
Control Panel
Region and Language
Keyboards and Languages
Keyboards and other input languages
Change keyboards...
Default input language can also be changed using the language bar directly.
So, instead of using the system-wide default locale to decide which input
method to use as default, better use the input language specified by the
user, which may be the same as the system-wide default locale, or different.
Following the previous example, with an English system locale and a
Japanese+IME input language, the default input method will now be IME
instead of Simple, which is closer to what's expected by the user.
This change only affects the application during startup; i.e. if the user
changes the input language while the application is running, we wouldn't be
changing the default input method to use. We could do this processing the
WM_INPUTLANGCHANGE messages, though.
https://bugzilla.gnome.org/show_bug.cgi?id=700428
The fix for child properties made the a11y dump for the assistant
case change. After close inspection, the new output is what the
code intends to produce, so update the expected result.
Call gtk_widget_get_mapped() in a couple of places before looking at the
widget's parent, since it might be set to a widget that has been
finalized, causing an invalid read.
GtkAssistant is bending the rules about child properties
of non-direct children, and the recent fix to accomodate
GtkInfoBar changes broke things. The effect was that child
properties of assistant pages in ui files were just not
applied, so all pages ended up without titles and with
the normal page type, leading to broken assistants all
over the place.
https://bugzilla.gnome.org/show_bug.cgi?id=706756
We need to copy the GDK .lib/.dll from Release_Broadway\<Platform>\bin
or Debug_Broadway\<Platform>\bin to Release\<Platform>\bin or
Debug\<Platform>\bin respectively during the build of Broadway flavors of
GDK, as the MSVC introspection builds expects the GDK .lib/.dll to be
in Release\<Platform>\bin or Debug\<Platform>\bin.
Use a new property sheet to do so for Broadway builds of GDK-during the
builds of Win32-only GDK, the broadway builds of the GDK .lib/.dll would
be cleared out prior to the build of the Win32-only GDK.
GtkAssistant supports not showing the sidebar with the page
titles (if the page have no titles). Unfortunately, we were
hiding the sidebar in this case, but still rendering the frame
behind it, leading to a broken appearance.
Horizontal scrolling is unusual, but specifying some extra offscreen
space for it in free in the normal case where the viewport is the
same width as the canvas anyway, so lets do it.
If the new requested surface size is enough larger than the previous
one (but the old is still larger than the absolute minimum),
reallocate it anyway.
This fixes an issue where the text view initially requested a really
small extra size which was then increased but that didn't "take".
This patch uses GtkPixelCache to render the contents of the widget,
and typically a bit more, to an offscreen surface. The pixel cache in
turn manages rendering to the actual surface for the widget.
The current strategy for the size to render is the size of the widget
plus half the height.
https://bugzilla.gnome.org/show_bug.cgi?id=701125
There were some code added to this file that is meant for the X11 backend,
but they are being unconditionally built. Add build-time checks for the
X11 backend for these to fix the build on non-X11 platforms.
...for the gdk_cursor_new_from_surface work (commit b2113b73) where the
types of some parameters were changed, and also to silence a critical
GDK_IS_DEVICE when a menu item is selected (courtesy of LE GARREC Vincent
from bug 696756).
https://bugzilla.gnome.org/show_bug.cgi?id=705980
Due to the work on gdk_cursor_new_from_surface (commit b2113b73),
get_cursor_for_pixbuf() in GdkDisplayClass was converted to
get_cursor_for_surface(), which means the GDK Win32 backend needs to be
updated for the code to build and run on Windows, plus some function
prototypes and declarations/calls need to be updated as well.
https://bugzilla.gnome.org/show_bug.cgi?id=705980
We'll use a style class to be able to give this a different appearance,
but for the time being we don't really need to give this such different
margin.
https://bugzilla.gnome.org/show_bug.cgi?id=706592
When setting the lines property, the label will be ellipsized
to that many lines, with the ellipsis only appearing in the
last line. This is different from how ellipsization of multi-line
labels normally works in GTK+.
"title_box" is used for both a custom header bar and for a titlebar.
Since we want to help differentiate these cases in the code, rename
everything titlebar-internal to use "titlebar_".
https://bugzilla.gnome.org/show_bug.cgi?id=706529
This used to point to the GtkPathBar, which doesn't accept mnemonic activation, anyway.
This whole thing was a leftover from when we had a combo box to select a folder, but
this is no longer the case.
https://bugzilla.gnome.org/show_bug.cgi?id=706448
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
This way, the Wayland and the regular clipboard implementation can both
be compiled in and selected based on the display in use.
One thing potentially broken now is text mime type handling as Wayland
seemed to use different mime types in some places.
This lets you force a specific window scale, this is needed
for mutter to be able to disable the scaling as it needs access
to unmangled X window/screen sizes. It can also be useful to
force a specific scale in e.g. tests.
The state of the widget is not enough now to cache the pixbuf - we also
have to take into consideration the image effect itself, since the state
on the actual GtkStyleContext we use might not change, e.g. because the
change was on a parent context.
https://bugzilla.gnome.org/show_bug.cgi?id=705443
We should set the appropriate style classes when we have
constructed the content and know if it is a label, an image,
or both. Doing this in the convenience constructors is
problematic for language bindings, and misses out when the
content is changed after construction.
I'm currently working on porting view::FieldEntry (from libview) to C for use in
upstream GTK+. FieldEntry is a widget which allows users to enter structured
text such as IPv4 addresses or serial numbers. The way that FieldEntry
delineates the fields within the entry is with tabstops, using PangoTabArray
entries to precisely position the fields and delimiters. Because GtkEntry
rebuilds its internal PangoLayout fairly frequently, this requires a property in
the entry that will set the tabs on the layout whenever that happens. This API
looks very similar to one in GtkTextView.
Patch by David Trowbridge <trowbrds@gmail.com>. Updated for Gtk+ 3.10.
https://bugzilla.gnome.org/show_bug.cgi?id=697399
GtkEntry currently draws exactly the same no matter what the state of the
'editable' property. This is pretty confusing for users because there's
no visual feedback at all, it just seems like their keyboard is broken.
This change adds a "read-only" class to the StyleContext, which will
continue to allow the user to select/copy the text, but will draw the
entry as if it were insensitive, providing some indication that the
contents can't be changed.
Signed-off-by: David Trowbridge <trowbrds@gmail.com>
http://bugzilla.gnome.org/show_bug.cgi?id=694831
For backwards compat support we don't want old implementations not
supporting scaling to see the new scaled directories, so move these
to a separate list.
Keyboard activation relies on the menu not being visible,
so ensure that it isn't when the menu is attached.
Problem tracked down by Vincent Le Garrec,
https://bugzilla.gnome.org/show_bug.cgi?id=688738
Attached widgets inherit from the style of the widget they are
attached to. This can sometimes have unintended consequences,
like a context menu in the main view of gedit inheriting the font
that is configured for documents, or the context menu of the preview
in the font chooser coming up with humongous font size.
To fix this problem, we introduce a context menu style class
and use it for all menus that are used like that. The theme
can then set a font for this style class.
https://bugzilla.gnome.org/show_bug.cgi?id=697127
This reverts commit b2e666bf8f.
We need to keep cursor blinking configurable for accessibility
reasons.
https://bugzilla.gnome.org/show_bug.cgi?id=704134
Conflicts:
gdk/win32/gdkproperty-win32.c
gdk/x11/gdksettings.c
gtk/gtksettings.c
gtk/gtktextview.c
The rsvg loader now restricts what external files it will
allow to load from an svg. Thus our xinclude trick doesn't work
anymore. To work around that, embed the payload in a data: uri.
This is somewhat ugly, but the best we could come up with.
The current theme just makes all text bold that appears anywhere
in a headerbar, which is not great. We add 'title' and 'subtitle'
style classes to allow more targeted overriding of the font.
Commits the pre-edit string on receipt of focus_out and reset
commands.
Patch refinements by Cody Russell <bratsche@gnome.org> and
Ek Kato <ek.kato@gmail.com>
-For the binary "installation", look for the DLL files with their file
names consistent with the ones that are generated with the respective
Visual Studio projects.
-Remove any stray GDK DLLs that were left over from a Broadway-enabled
GDK when building a non-Broadway-enabled GTK+ binary set.
When trying to drag, we currently the position of the first motion
event to determine where the drag came from. This might be alright
in the case of the old animation, but the data will be inaccurate
if the user has moved the pointer quite a bit since pressing the
cursor to start dragging. While we could monkey patch the GdkEvent
at the widget layer, this is unintuitive and strange.
Add a new API that takes a set of pointer coordinates describing
the origin of the drag. Additionally, adapt most widgets to use
it and use it with correct coordinates.
https://bugzilla.gnome.org/show_bug.cgi?id=705605
Fixes:
(rhythmbox:22802): Gtk-CRITICAL **:
_gtk_widget_get_preferred_size_for_size: assertion 'size >= -1' failed
that is gtk_paned_get_preferred_size_for_opposite_orientation calls
_gtk_widget_get_preferred_size_for_size on child2 with a negative
size.
As gtkpaned size was (32), child1 minimum size was (55) then
for_child1 became (32) for an handle_size of (5). Thus for_child2
ended up as (-5).
https://bugzilla.gnome.org/show_bug.cgi?id=705624
Just as for GtkRangeAccessible, we were not even trying to disconnect
the signal handler from the adjustment. The same fix works here:
override the widget_set and widget_unset vfuncs.
https://bugzilla.gnome.org/show_bug.cgi?id=705692
We were only disconnecting the signal in finalize, when
the widget was already unset. Instead, override the widget_set
and widget_unset vfuncs of GtkAccessible, and keep a reference
to the adjustment.
https://bugzilla.gnome.org/show_bug.cgi?id=705692
Move all the system includes, defines and function prototypes into a
separate header gdkwin32misc.h, so that we could keep gdkwin32.h as simple
as possible.
Replace the deprecated API calls with the updated APIs, and fix the build
of modules/input/gtkimcontextime.c, as we really needed
gdk/gdkkeysyms-compat.h (gdk/gdkkeysyms.h was already included)
https://bugzilla.gnome.org/show_bug.cgi?id=705068
...this was split into two commits as this source file has different
line endings (for some reason) from the other GDK-Win32 source files that
were updated in the quest to refrain from using deprecated APIs
Make sure we always deal with the same screen when
connecting / disconnecting the theme-variant changed handler.
Pointed out by Morten Welinder in
https://bugzilla.gnome.org/show_bug.cgi?id=705640
When an icon is requested as symbolic, our generic fallback algorithm
uses fullcolor icons when the specified icon name is not found, treating
the "-symbolic" suffix as another component of the icon name.
Change the algorithm to check beforehand if the icon is symbolic, remove
the suffix if so, and re-add it at the end for all the generated icon
names.
https://bugzilla.gnome.org/show_bug.cgi?id=680926
With multiple GDK backends in the process, we run into problems where
we try to use the Wayland im module on X, which crashes. This commit
adds a quick backend filter that removes the wayland, xim and ime
input methods from consideration unless the corresponding GDK backend
is in use.
This code is called early on, without a window, and then later on
with a window. Currently, it returns different results for these
cases when the setting contains a value. That leads to pointless
construction and destruction of im contexts. Instead, just look
at the settings of the default screen. In practice, there is only
one screen, ever.
... to make it possible to insert rows in the middle of the list without having
to fiddle with the sort functions. One of the first users is going to be Glade.
https://bugzilla.gnome.org/show_bug.cgi?id=705558
We want a surface so we can properly represent the scale factor for it.
All backends are converted to use surfaces and we reimplement the
backwards compat code in the generic code.
To handle hidpi support we need to make sure we don't
downscale scaled css images.
Note: If cairo_surface_create_similar starts doing this
by itself we need to back this out.
A problem with the zoom scroll mode is that you have to restart
if you hit the bottom of the screen before you hit the bottom
of your document.
This commit adds an autoscroll feature to the zoom scroll: if
you move outside the window while in zoom scroll mode, we keep
scrolling in the direction you were going until you let go
of the mouse button.
https://bugzilla.gnome.org/show_bug.cgi?id=704703
Triggering zoom scroll mode by Shift click was too much
of an easter egg. It also requires using keyboard and
mouse together, which is hard to do for many users.
Instead, we now trigger zoom scroll mode by click-and-hold
(or touch-and-hold).
https://bugzilla.gnome.org/show_bug.cgi?id=704703
The internal class GtkPressAndHold was so far only
reacting to touch events. But in most cases where
a touch-and-hold or 'long press' pattern is useful,
click-and-hold can also be used.
This patch makes GtkPressAndHold react to mouse
clicks as well.
https://bugzilla.gnome.org/show_bug.cgi?id=704703
Add a boolean property that controls whether a window close button
will be shown in the header bar or not. Doing this in the toolkit
will ensure consistency of the visual apperance.
https://bugzilla.gnome.org/show_bug.cgi?id=702971
If we don't dispatch the pending events then we can enter poll with events
still requiring to be processed and which can then lead to us deadlocking
there.
These files are generated, so adding them to git is somewhat
icky, but it helps translators who currently can't use intltool-update
on a fresh git checkout.
Don't specify packing properties for internal children.
This doesn't work unless the internal children are actually
direct children of the container (which we are about to
change for GtkInfoBar). Also, it is redundant, since we
just set the properties to the values they already have
anyway.
Packing properties in ui files are parsed and handled by
the container, which assumes that the child is a direct
descendant. For internal children, this is inconvenient,
because we don't want to reconstruct the entire internal
structure between the container and the child in the ui
file.
It would be best to not specify packing properties
in that case, but since existing ui files do this for
GtkInfoBar and we are about to change the internal
structure of GtkInfoBar, be more forgiving here.
This reverts commit fbbcb5c01b.
We will be doing this in gnome-settings-daemon itself instead,
as some X11 based platforms using GTK+ will want to override this.
Xsun is no longer shipped to customers, and Oracle/Sun's Xorg distribution
uses "Sun Microsystems" as the vendor name, so this hack is incorrect in
the more common recent cases.
We checked for G_IS_LOADABLE_ICON() before GDK_IS_PIXBUF().
Since we made GdkPixbuf implement GLoadableIcon, the special case for
pixbufs is never used, and the much much slower GLoadableIcon path is
taken instead. Move the GdkPixbuf one to be first to fix that.
https://bugzilla.gnome.org/show_bug.cgi?id=705320
Update the gtk-install-bin property sheets so that it does not "install"
the wrong GDK DLL/LIB when building a broadway-enabled GDK
when the non-Broadway GDK had been previously built.
Add the Visual Studio 2010 projects to build the GDK Broadway backend, just
like the Visual Studio 2008 project files in the last commit. Similarly,
split up the property sheets so that they are easier to maintain and can
be made more flexible for different build types. Also remove some unneeded
stuff from some of these items.
Also, fix the filter file completion for GTK, as a source file was excluded
for that and this was overlooked as it seemingly did not cause any trouble.
-Add Visual Studio 2008 projects and pre-configured gdkconfig.h for
Broadway builds
-Decouple the Visual Studio property sheets, to simplify maintenance and
enhance flexibility for different builds
Visual Studio 2010 projects updates will follow later.
if exist ..\..\..\MSVC_$(Configuration) goto DONE_GDKCONFIG_H
if exist ..\..\..\gdk\gdkconfig.h del ..\..\..\gdk\gdkconfig.h
if exist ..\..\..\GDK_BROADWAY_BUILD del ..\..\..\GDK_BROADWAY_BUILD
if exist ..\..\..\MSVC_$(Configuration)_Broadway del ..\..\..\MSVC_$(Configuration)_Broadway
if exist $(Configuration)\$(Platform)\bin\$(GtkDllPrefix)gdk$(GtkDllSuffix).dll del $(Configuration)\$(Platform)\bin\$(GtkDllPrefix)gdk$(GtkDllSuffix).dll
if exist $(Configuration)\$(Platform)\bin\gdk-$(ApiVersion).lib del $(Configuration)\$(Platform)\bin\gdk-$(ApiVersion).lib
if "$(Configuration)" == "Release" del ..\..\..\MSVC_Debug
if "$(Configuration)" == "Debug" del ..\..\..\MSVC_Release
if exist ..\..\..\MSVC_$(ConfigurationName) goto DONE_GDKCONFIG_H

if exist ..\..\..\gdk\gdkconfig.h del ..\..\..\gdk\gdkconfig.h

if exist ..\..\..\GDK_BROADWAY_BUILD del ..\..\..\GDK_BROADWAY_BUILD

if exist ..\..\..\MSVC_$(ConfigurationName)_Broadway del ..\..\..\MSVC_$(ConfigurationName)_Broadway

if exist $(ConfigurationName)\$(PlatformName)\bin\$(GtkDllPrefix)gdk$(GtkDllSuffix).dll del $(ConfigurationName)\$(PlatformName)\bin\$(GtkDllPrefix)gdk$(GtkDllSuffix).dll

if exist $(ConfigurationName)\$(PlatformName)\bin\gdk-$(ApiVersion).lib del $(ConfigurationName)\$(PlatformName)\bin\gdk-$(ApiVersion).lib

if "$(ConfigurationName)" == "Release" del ..\..\..\MSVC_Debug

if "$(ConfigurationName)" == "Debug" del ..\..\..\MSVC_Release

Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
