No need to construct a detailed signal name for
every action when we can just look up the signal ID
once and use the quark that the GParamSpec already
has. Also, we don't need to loop over the actions
every time we get a notification.
We were having a problem where property actions were
not getting state updates because prop_actions_connect
was triggered from some instance_init function while
the widget class is not in place yet.
Delay that call until the widget is fully constructed,
so we can guarantee that we are dealing with the
correct class private struct, and see all class actions.
We don't get motion events from the popup (due to grabs),
so just don't hide as long as we're grab shadowed. This
makes the controls stay up until the volume popup is
dismissed.
We already dropped this invariant in gtk_widget_verify_invariants()
because it was not true in all cases. Also, it is not really useful
these days as we extended what it means to be a "child" to also
include widgets in different toplevels.
For example, a popup in a popup button need not be realized just
because the button is in a realized window. The main invariants
we want are:
* Parent is realized before child
* Widget is realized before it is mapped
This sounds like its not a huge deal, but in fact it is a massive win
for things like menus, because when we go between menus in a menubar
each switch between two open menus involves a lot of intermingled
crossing events to different surfaces and for each of these the
tooltip window of the toplevel gets assigned to the new surface. This
shouldn't be a huge deal, as the tooltip window is not even visible,
but due to the realized invariant it get re-realized each time it gets
re-assigned.
The `element-type` annotation is for GList and GSList only, and turns
out adding support for GListModel in gobject-introspection breaks Vala
and the GIR for GIO.
Instead of using `element-type`, we can use the `attributes` annotation,
which is ignored by code generators based on the GIR data.
When we find out that we've been flipped, set
final_position and final_rect before allocating
the popover. This prevents 'smashed-in beak' disease.
Fixes: #2671
Clarify that gtk_file_chooser_add_filter is only taking
ownership of a floating reference, which translates to
"transfer none", not "transfer full".
Fixes: #2016
This is a possible fix for https://gitlab.gnome.org/GNOME/gtk/-/issues/2657
Use a NULL return from g_file_query_info_finish() to detect cancellation
of the query, and avoid derferencing a stale pointer.
Now that the title buttons are encapsulated in a separate widget, use
it in the header bar.
Hide them when empty, so that they don't add extra spacing.
Some machinery (like clicking on expander) relies on prelight state
being up to date, but we don't set prelight on touch events. Do that
transiently, as long as a row is clicked.
Fixes tapping on expanders.
If the popover fails to be shown, it would internally undo visible
and mapped state. If we just proceed as normal, the widget enters
in inconsistent state, and a grab remains issued on the invisible
widget, preventing further input from the input device.
This function to revoke implicit grabs may be called with a NULL
device, which means all devices should be revoked. Fix the check
so this actually happens.
Fixes implicit grabs (maybe) being unset in result to a GTK grab.
Update the layout when any of the following properties changes:
* modal
* transient-for
* resizable
* deletable
Stop updating it from GtkWindow, make that function static.
Only turn on visible focus when a key event actually leads
to a change in focus location (ie, 'keynav').
Make the visible focus disappear after 5 seconds of no
keyboard interaction, to avoid permanent focus ring
distraction.
As an extra bonus, make it so that we make the focus
visible while the Alt key is pressed. This gives us
a 'find my focus!' shortcut, and goes well with the
prexisting use of Alt for finding mnemonics.
Discussed in: #2644
In the n == 2 case, we want to avoid the buttons being
drawn all the way out to the ends of the box, but in the
n >= 3 case, that is exactly what we want for the first
and last button.
When the icon theme changes, we want to both initiate
a css SOURCE change (since css values may depend on the
icon theme), as well as calling gtk_system_setting_changed,
since widgets need to drop cached images, e.g. in GtkIconHelper.
Replace calls to gtk_style_context_reset_widgets by
gtk_system_setting_changed, with the appropriate setting,
and add style invalidation where required.
This gives us a hook to walk the widget tree whenever a global
setting changes and do per-widget invalidations. This will
replace gtk_style_context_reset_widgets().
Add the following actions:
* window.close
* window.toggle-maximize
* window.minimize
Ensure they are disabled when not appropriate, such as for modal windows.
Use them in GtkHeaderBar.
The application menu isn't particular relevant anymore, and the support for
showing fallback appmenu as a headerbar icon is one of the things tying
GtkWindow to GtkHeaderBar.
Remove support for "menu" window decoration element completely, update
GtkHeaderBar docs.
When freeing the display GtkIconTheme and that was the last owner we
ran into a deadlock, because we unref the "next-to-last" ref inside a
gtk_icon_theme_ref_aquire/release() pair, which makes the final unref
to happen in the release(), while the ref lock still was held.
The unref triggers dispose which tries to NULL out the ref, but that then
deadlocks on the mutex being held by the caller already.
We fix this by moving the release unref outside the lock. This is safe
because refcounts are atomic, and we *do* own the ref.
- HC got caught up on a condition made for color (suggested) buttons.
the rule just made the slider a tag lighter on dark probably due to the
overlaid AA making it stronger. Got rid of it for now, if someone cares
enough we can reintroduce an exception (that doesn't break HC).
When toggling caret visibility (with F7) we would hit an assertion if the
cursor is currently blinking. This adjusts things to ensure that we should
be showing the carent when scheduling our blink timeouts.
Fixes#2647
Emit accessible change signals after setting the
new selection limits. This was broken in
commit f6bedd0d5e, and showed up as broken
accessibility tests.
This was only living in gtkcontainer.c for historic
reasons. Move it closer to where it belongs, and
rename it from 'idle' to 'layout', since it is
really about the layout phase of the frame clock,
nowadays.
These popovers handle clicks outside the widget, and expect input
to be routed through it when it is shown. This pretty much matches
the characteristics of GTK grabs, and setting one up will ensure
the cancellation of event controllers that were active at the time
of showing the popover.
Closes: https://gitlab.gnome.org/GNOME/gtk/-/issues/2446
Commit 4a3742979d which removed
GTK_FILE_CHOOSER_ACTION_CREATE_FOLDER slipped
a bad g_assert() refactoring, which is causing
a crash when selecting a folder in a GtkFileChooserButton
dialog.
Fixes#2635
- Apply sizing regardless of style class.
- Fix bottom border not reaching the bottom corners.
- Remove unnecessary style overrides that should have been removed
during the redesign.
- Fix broken visited link styling by avoiding invalid selector
`🔗visited` - CSS :link and :visited are mutually exclusive.
- In link button, move %link from the child label to the parent button,
so that the hover/active effect is applied within the whole button and
the visited style is applied correctly.
- Remove invalid selectors `:backdrop:backdrop`.
This turns out to be necessary for transitioning filters,
and we run into criticals if transitioning numbers
unexpectedly returns NULL. I've observed that with
* {
filter: invert(80%);
}
This was broken in c9e972eecb.
And implement this property by listening for focus
changes, and updating the adjustments. This is a
replacement for setting focus adjustments on containers.
Make GtkAspectFrame not derive from GtkFrame anymore,
since frames now always draw, well, a frame. Also,
add proper setters for the properties of GtkAspectFrame.
Update our sole user.
Fixes: https://gitlab.gnome.org/GNOME/gtk/-/issues/2627
The viewport draws a frame at the same place as
the scrolled window, so there is really no need
to have that ability in both. Just drop the frame
from viewports.
Always track visited state of links. This way all visited links in all
labels look the same. Whether the theme wants to style :visited is the
theme's business.
This makes it more obvious that gtk_label_setup_mnemonic() depends on
the root of the widget and it therefore makes sense to call it in
::root/::unroot.
Don't allocate a new GString if we never need it and therefore don't
create the PangoAttrList if we have no attributes anyway. Update callers
to handle the possible NULL return value.
Don't crash while picking. The event signal broke
when GdkEvent was turned into a type instance,
since the automatic marshallers don't know how to
deal with that. Manually set the right marshaller.
GdkEvent has been a "I-can't-believe-this-is-not-OOP" type for ages,
using a union of sub-types. This has always been problematic when it
comes to implementing accessor functions: either you get generic API
that takes a GdkEvent and uses a massive switch() to determine which
event types have the data you're looking for; or you create namespaced
accessors, but break language bindings horribly, as boxed types cannot
have derived types.
The recent conversion of GskRenderNode (which had similar issues) to
GTypeInstance, and the fact that GdkEvent is now a completely opaque
type, provide us with the chance of moving GdkEvent to GTypeInstance,
and have sub-types for GdkEvent.
The change from boxed type to GTypeInstance is pretty small, all things
considered, but ends up cascading to a larger commit, as we still have
backends and code in GTK trying to access GdkEvent structures directly.
Additionally, the naming of the public getter functions requires
renaming all the data structures to conform to the namespace/type-name
pattern.
Use :focus-within for focus in entries, since the
actual focus is on the text within, and :focus for
notebooks, since we don't want to draw an outline
around the notebook when the focus is in content.
This is used for widgets that contain the focus widget,
reserving the focused state for the focus location itself.
This aligns our focus state handling with
https://www.w3.org/TR/selectors-4/
which could happen after confirming the "file overwrite"
dialog and may result in a different file being overwritten
causing data loss.
The oblivious file selection can be done by a mouse
click or keyboard press sent inadvertently just after
confirming the "file overwrite" dialog (and before the
enclosing GtkfilechooserDialog is closed).
Fixed by adding a flag to ignore any button/key press
events sent to the file list. We set this flag just
after the user accepts the "file overwrite" dialog (which
means the enclosing GtkfilechooserDialog is about to
get closed). And we restablish the flag when the dialog
is shown again (in its map() handler).
Fixes data loss issue #2288
Stop rewriting key and focus events on the GDK side.
Instead deliver them as they are, and propagate them
from the root on the gtk side, in gtkmain.c. And
stop complaining about focus events on popups - we
can just ignore them if we have no use for them.
- People seem to misunderstand the unsharp-mask-like increase
of visual contrast for "fuziness". That is not the reason for
the change. The stylistic change of just going with flat
text label allows to simplify the code and drop complexity.
Fixes https://gitlab.gnome.org/GNOME/gtk/-/issues/2280
- sligtly increase contrast for the treeview borders
- FIXME: High Contrast seems to drop the borders completely,
there might be some trickery for using border-left-color and
border-top-color this way as even forcing the color 'red' seems
to render invisible on HC.
Fixes https://gitlab.gnome.org/GNOME/gtk/-/issues/2238
Without this, it seems impossible to make cross-section
keynav in the Emoji chooser work. I've tried, but got
lost between the focus, grab_focus, move_cursor and
keynav-failed vfuncs and signals, and their competing
implementations GtkFlowBox and GtkEmojiChooser.
Set version and soversion separately for the library.
When we do the 4.0 release, we will set:
gtk_soversion = '1'
gtk_library_version = '1.0.0'
See https://gitlab.gnome.org/GNOME/gtk/-/issues/1963
If we call any functions that may call ensure_layout
themeselves, we risk having the cached layout pulled
out from underneath. Better play it safe and take a
reference.
Make this a full async function, and add a simple wrapper.
Call gtk_show_uri_full() if you need control over the
results, and use gtk_show_uri() if you are fine with
ignoring any errors.
We have event controller apis to replace these.
There is one remaining use of gtk_get_current_event_time
in gtkwindow.c, so we can't drop the implementation yet.
Add a section in the migration guide for this.
This commit handles complicated cases where we selections.
We handle this by adding extend and modify parameters to
the ::move-cursor signals, and adjust the bindings
accordingly.
We don't really have an event anywhere close in most
cases, and we already pass GDK_CURRENT_TIME in half
the cases anyway.
If we want to be serious about this, we need to pass
the event itself, since future focus-stealing protocols
may not rely on just a timestamp.
