The intention of this patch is to make the code clearer, shorter and
most of all to avoid recreating the widget path and setting it path
twice on the style context when the style context was recreated.
This patch optimizes window resizes by assuming that if a widget
has the same height at a width of 50 as with a width of 150, the
height for width 100 will also be the same.
The patch also further optimizes the cache allocator, now there
are 2 pointer arrays of up to a maximum of 5 requests, the arrays
will only be allocated if a request is ever made in that orientation
and the array will be sparse until each request is made (i.e. if a
label can only wrap to 3 lines, there will only be 3 out of a
possible 5 SizeRequest structures allocated to cache it).
This patch makes contextual height-for-width request caching
optional (the contextual cache is not allocated for widgets that
report GTK_SIZE_REQUEST_CONSTANT_SIZE).
The constant size request mode defines a request mode where
height-for-width geometry is unneeded, thus optimizing GTK+
by reducing the overall amount of requests that need to be
performed and cached while resizing an interface.
Fixes Bug 639584 - initial emission of GtkWidget:style-set is
not happening.
GtkWidget was filtering out ::style-updated (and ::style-set)
emissions until the widget was realized in order to avoid often
useless updates during widget construction and placing.
This is now done instead until the widget has a parent/screen,
which ensures the initial emission happen prior to the first
size negociation, while still filtering out all early emissions
during construction.
Writing onto the passed-in GtkStateData could cause changes that would
propagate to siblings, as the data was not reset again.
By copying the data structure, this is avoided and the proper values are
passed to sibling widgets.
GtkWidget now parses custom attributes like
<style>
<class name="dark-label"/>
<class name="big-heading"/>
</style>
to add style classes to widgets.
https://bugzilla.gnome.org/show_bug.cgi?id=643347
Sensitivity changes were not properly propagated down the
hierarchy. There were two issues here:
a) correctly identifying when a state change request affects
sensitivity
b) not filtering out sensitivity in gtk_widget_propagate_state(),
since gtk_widget_set_sensitivity() uses that to do its work
https://bugzilla.gnome.org/show_bug.cgi?id=641431
Optimized GtkSizeGroup code that is invoked for every queued resize
and every request that is not previously cached by trading qdata on
widgets for 3 extra bitfields on the GtkWidgetPrivate structure.
It is used to get the default providers, without them
the style context can't do much. A check for NULL screen
is done before any sensitive call to
gtk_style_context_set_screen(), in the hope that any widget
will open the display before doing anything related to
styling. Fixes bug #641429, reported by Bastien Nocera.
This management is better done per-widget rather than per-screen,
as windows being destroyed won't trigger a leave notify for the
devices on top of it, and this information is too transitive
to keep weak refs and such.
This fixes the critical warning seen in gtk/tests/testing.
Instead of checking if klass->get_request_mode is != NULL from
the gtk_widget_get_request_mode() api, this allows classes to
trust that there is a default implementation and chain up (specifically
added this for gtkmm wrapper objects).
GtkStateType was generally used as an index in GtkStyle color arrays, so
bigger values will cause invalid memory accesses in widgets that are still
doing that. this was seen in focused GtkIconViews for example
together with commit 8903615a34, this finally fixes bug #640282.
Insensitivity is handled separatedly in _gtk_widget_update_state_flags(),
but the insensitive flag is mistakenly unset afterwards if clear is TRUE
in gtk_widget_set_state_flags().
There is only one widget supposed to have the focused flag at a given time,
so avoid propagating the state down the hierarchy, the focused flag is now
also set in _gtk_widget_set_has_focus().
This function is a more convenient variant than
gtk_widget_set_device_events(), as it will
1) perform changes down a widget hierarchy, to
all windows.
1) use the same event mask than gdk_window_get_events()
This function takes a region ID and cancels all animations
on or beneath that region (as in push/pop_animatable_region).
First user of this is GtkWidget itself, so unmapped widgets
have looping animations cancelled. Fixes bug #638119, reported
by Jesse van den Kieboom.
Now GtkWindow takes some measures when setting toplevelness:
- When a window becomes toplevel after being embedded it saves
the visibility state and reshow's itself so that the window
re-realizes and presents itself again automatically
- When emitting hierarchy-changed, synthetically mark the toplevel
as not anchored, this allows the hierarchy changed propagation to
recurse properly.
GtkWidget also takes care to unset the parent window *after* unparenting
the widget and after emitting the heirarhcy changed that leaves a NULL
toplevel.
That means there are now 2 cycles of "hierarchy-changed" when removing
an embedded toplevel from a parent, first one that makes the new toplevel
a NULL one (since the toplevel flag is not yet restored), the second cycle
makes the removed window toplevel again when setting the parent window
to NULL.
GtkFileChooserDefault watches the toplevel and montitors "set-focus"
signal on it... however the connection needs to be remade when the
GtkFileChooserDialog is in an embedded toplevel.
Measure's taken: GtkWindow propagates hierarchy changes when
_gtk_window_set_is_toplevel() is called, gtk_widget_unparent()
unsets the widget's parent window earlier in the function so that
the possible hierarchy change is still able to properly access the hierarchy.
GtkFileChooserDefault checks if the "new" toplevel is indeed
gtk_widget_is_toplevel() but not the old one, GtkRange has been
updated to use gtk_widget_is_toplevel() inside it's hierarhcy_changed
vfunc, other classes already do this properly.
This patch makes gtk_widget_set_parent_window() undo the toplevelness
of a GtkWindow, GtkWindow then realizes itself as a normal child widget
and behaves like a normal GtkBin by checking gtk_widget_is_toplevel() in
several places (show/hide/map/unmap/draw/size_allocate/check_resize/configure_event).
For normal toplevels, visible is tightly bound to mapped, but for
something like a toplevel that exists within a Clutter stage we
may want to make mapping dependenton external factors, so we shouldn't
actually checked that !mapped toplevels are !visible.
Pointed out by Owen Taylor,
https://bugzilla.gnome.org/show_bug.cgi?id=637834
Previously, for performance reasons we would sometimes
skip invoking the unmap signal (and associated vfunc)
in favor of simply unrealizing. However, widgets then
had no way to clean stuff up when they were hidden
(but still inside a parent which was shown).
This patch also removes _gtk_tooltip_hide() which
was done in both unmap and unrealize in gtkwidget.c,
now can only be in unmap.
There are probably lots of things cleaned up in
unrealize that would now be better to move to unmap.
https://bugzilla.gnome.org/show_bug.cgi?id=629923
Requires fixes to GtkContainer and GtkWindow to unmap their
children, rather than just withdrawing or hiding the container
window.
Requires fix to GtkHandleBox to chain up to GtkContainer unmap.
Historically we avoided these unmaps for efficiency reasons,
but these days it's a bigger problem that there's no way
for child widgets to know that one of their ancestors has
become unmapped.
These checks are a bit expensive so require --enable-debug=yes.
gtk_widget_verify_invariants() checks invariants mentioned
in docs/widget_system.txt in particular, and can verify
others in the future.
Some of the invariants in docs/widget_system.txt don't
in fact hold right now, so those are #if 0'd in this
patch pending someone fixing either the docs or the code.
Most code in gtkrc.c has been turned into a no-op, but that one
reverting in public API (gtk_rc_scanner_new() and such). GtStyle
is also more shallow, now fully relies in the backing
GtkStyleContext and all connection to gtkrc.c has been removed.
GtkBinding has been also affected, there is no replacement yet
for custom keybindings in style files, so that piece of code that
hooked into gtkrc has been replaced by a FIXME so in the future
it may be added back.
In the process of removing all sealed members from headers.
At the same time, add a gtkwindowprivate.h header and move
all internal functions from gtkwindow.h there.
The old functions to get core pointer and devices list are gone as
well. This slice is entirely replaced internally by multidevice
handling and may just go.
This allows us to add a check before executing
->get_preferred_height_for_width() to ensure we always
request for at least the minimum required size (and lets
us remove the warning in gtkcontainer.c telling implementors
to do this check manually from thier container implementations).
This object backs up gtk_widget_override_* operations. This object
is not meant to be public because any intention to modify widgets'
style in a themeable way should involve using regions/classes, so
they're modifiable through CSS. As such, the API is really
short-scoped.
Things like font settings depend on the screen, and widgets
like GtkTextView trigger queries on widgets without screen
when the parent window is being destroyed.
gtk_widget_override_*() deprecates gtk_widget_modify_*(). There are
only functions to modify fg/bg/font/symbolic color, If anything more
fancy/complex is needed. There is the possibility of adding a
GtkStyleProvider yourself.
gtk_widget_(set|unset|get)_state_flags() has been added, using GtkStateFlags
to represent the widget state. GtkStateType API has been implemented on top
of the new one.
This is now used throughout in order to have the full path for a given widget,
including intermediate named regions, the default implementation just returns
the GtkContainer's path copy, no intermediate regions between.
Some unparented widgets like to ask style details, so now the style is
constructed regardless of the parent being present or not, and then
reconstructed if the parent changes.
There will be one GtkStyleContext per widget, at the moment its
lifetime is tied to the widget's, but it could be narrowed down
to GTK_WIDGET_REALIZED.
Now gtk_widget_size_allocate() unsets the resize_needed flags
before returning, essentially this means that any widget that
has a queued resize and is allocated before resize time, including
queued resizes from inside a size_allocate() method will be
cancelled.
alignment/margin vfuncs adjust_size_request/allocation
Now get_height_for_width() will internally update the for_width
before passing it to the real height_for_width() vfunc, allowing
margins and extra space for alignments to be stripped, thus requesting
sufficient height for greater than natural widths (and also accounting
for margins properly). Test case adjusted in testadjustsize to ensure
proper behavior.
