The call to gtk_border_free() within unpack_border() felt completely
in the wrong place, as the border actually pertains to the GValue
being unpacked. Plus, the GValue itself was also being leaked.
In finalize(), clear all rulesets.
In parse_declaration(), Free the GValue under unhandled error situations.
In gtk_css_provider_load_internal(), Do not leak the file contents.
In rgba_value_parse(), unref the symbolic color once we've resolved it.
In gradient_value_parse(), take the GtkGradient so we leave no dangling
references.
GValues stored in GtkCssRulesets are gslice managed, so don't
g_memdup() GValues from shorthand properties. This fixes
memory corruptions when reloading contents in a GtkCssProvider.
We want the menu realized so we know the size it's allocating to itself.
And we need that size to position the menu properly.
This is best visible on right-to-left.
Fixes sidebar in evince not showing up. This was caused by the
visiblility of the widget changing and the paned not fixing its child
window's visibility accordingly.
If the user pressed Enter to confirm the file chooser while the filename entry
was empty, then gtk_file_chooser_default_should_respond() would go back and forth
between the cases for handling the filename entry and the file list.
Don't rely on priv->resize_grip_visible as the code comment in the
variable declaration indicates.
This fixes warnings with GtkPlug, which can cause resize_grip_visible to
be TRUE but grid_window to be NULL - running tests/teststatusicon
reproduces this.
This broke with 7ef113ce56
Shorthand properties are basically the same a in CSS. For storage in
style properties or the CSS provider, they are unpacked into the real
values, so it is possible to partially override them.
No properties are yet converted to the new world yet, this is just the
code for supporting them.
... instead of duplicating code. This causes an extra g_value_copy().
If that turns out to be a performance issue, we can invent something
that handles this (like passing a gboolean take_value).
The reason for this duplication deletion is that we want to complicate
the setting code to handle shorthands by unpacking them and storing the
separate values.
Instead of initing the default style properties in the class_init
func of the style properties, init them when they are first needed -
when they are queried or when new ones are registered.
That way, they will always be available.
This will be used as a base both for parsing text-shadow and box-shadow
properties. The type is private, as there's no real use in exporting
this in a public API.
https://bugzilla.gnome.org/show_bug.cgi?id=649314
This has mostly two advantages:
- the most obvious one is the theme can render a border around the
sidebar if it wants to.
- we also can avoid hardcoding a container border width for the sidebar,
and just use a padding from the theme. This also allows different
themes to define a different padding, etc.
The drawback is we must draw the background ourselves, but it's easy
enough.
https://bugzilla.gnome.org/show_bug.cgi?id=650530
Lots of code calls gtk_menu_popup() and we don't want to resize the
window needlessly.
In this particular case, keyboard navigation to submenus caused those
submenus to shrink.
Note: I'm not sure this fix doesn't have nasty side effects, as I'm not
a specialist on menu popup code, so if it does, we'll need to revert it.
Until then, let's keep it, it fixes a bug.
If props == NULL in gtk_symbolic_color_resolve(), fail sanely for named
colors. The docs used to say it was not allowed to pass NULL for named
color, but that had problems:
1) You do not know if a color was created that way. This is especially
hard for generic users (like language bindings).
2) It wasn't even true. Colors using other symbolic colors would also
fail when trying to resolve their named colors, but the docs didn't
say so.
And because I want to use the function to resolve static colors early
where possible, I changed things.
This provides a huge speedup as we only need to preprocess style
properties when they are indeed inherited. This roughly doubles the
performance of the CSS matcher and brings the time taken by
gtk_css_provider_get_style() from 19% to 7% in my favorite benchmark.
One for the style properties, one for the widget style properties.
This way we can make one hash table by pspec which means we don't have
to repeat the pspec lookup.
Keep rulesets as an on-stack/heap structure instead of allocating all
instances separately.
Also, pass a ruleset to the ruleset parser, so we can make the ruleset
parser do lots of fancy things that might be useful for performance.
The code used the quarked name before, but when we already have the
pspec we want to have a lookup that does not involve quarking. And
lookup is equally fast if we only have the name.
Previously we got the list of all matching rules and then iterated it to
find the first one that had the property. Now we look while matching
rules, so we don't lookup rules that we don't need.
Instead of relying on GScanner and its idea of syntax, code up a parser
that obeys the CSS spec.
This also has the great side effect of reporting correct line numbers
and positions.
Also included is a reorganization of the returned error values. Instead
of error values describing what type of syntax error was returned, the
code just returns SYNTAX_ERROR. Other messages exist for when actual
values don't work or when errors shouldn't be fatal due to backwards
compatibility.
This is pretty important, because otherwise recursions cause crashes.
And if you accidentally change your theme to one that crashes on load,
all your gonna SEGV and then on reboot, gdm tries to load the theme...
Call gtk_css_provider_load_from_file() instead of the internal function.
This has two advantages:
1) It simplifies the code a lot
2) It gets rid of GMappedFile usage. GMappedFile does not work
everywhere, so this is finally portable.
This way, we achieve two things:
1) We can unify file loading to one location
2) We can emit the error from file loading using the parsing-error
signal. This is very useful for @import handling in particular.
Emits the error without the need for a scanner. Also simplifies
gtk_css_provider_take_error() because we now can assert an available
scanner at all times.
Instead of having an error member in the CSS provider's private struct,
connect a signal handler when an error is passed in. This has two
advantages:
1) It makes the code clearer as we don't have to keep track of an error
member anywhere.
2) It causes a non-emission of the g_warning() when an error was passed
in, because it only triggers when no signal handlers are connected.
So we get identical behavior to GTK 3.0 where warnings where only
emitted when no error was passed in.
Instead of aborting a parse whenever we encounter an error, parse to the
end. But if a GError was passed in, reset the provider completely as if
nothing had been parsed.
Value parsing only sometimes emitted errors. Sometimes it didn't emit
errors but ignored the value, sometimes it took a default, sometimes it
converted it to something it deemed suitable.
While refactoring, I moved the whole GValue <=> char * conversion
routines to a separate file, to make navigating the core css provider
easier.
Previously, we only checked for errors after parsing the full
declaration. Now we detect errors with the property before even
attempting to parse its value.
The benefit here is that the error reporting reports the correct line
and position numbers.
The previous code failed to account for all child visibility and paned
mapedness invariants which could cause stray GDK windows to appear.
Not good.
Credit goes to Xan for triggering it.
This is a special-purpose button that can be used together with
GPermission objects to control the sensitivity of system settings.
Suitable permission objects can e.g. be obtained from PolicyKit.
http://bugzilla.gnome.org/show_bug.cgi?id=626457
Background color was not painted on the whole tagged segment after
line breaking, this was due to the default ->prepare_run() function
wiping the internal PangoColors clean... fixed this by unconditionally
setting local rgba colors from prepare_run instead of comparing them
and bailing out.
This now allows text view to render text with alpha values in
the text foreground and backgrounds, the work is almost complete,
currently the error-underline-color is still a GdkColor style property
and since we use only GdkRGBA for rendering it needs to be converted
and applied, probably a new rgba version of the style property should
also be introduced.
This commit adds tests/testtextview that must be run from the tests/
directory to show translucent text in action.
The progressbar is composed by two different rendered areas: the trough
(i.e. the non-filled part of the bar) and the bar itself.
The bar should be able to fill the whole height/width of the trough
without resorting to nasty hacks in the theme, and we can control the
amount of space between the bar and the trough with the padding already.
https://bugzilla.gnome.org/show_bug.cgi?id=649593
GtkAssistant is widely recognized as a butt-ugly widget.
This commit changes its style to look more modern. We
deprecate the sidebar and header image properties and
don't show them anymore. Instead, page titles are arranged
in a sidebar, with the title of the current page shown
in highlighted style.
This commit simply removes some child->parent == container checks,
to add some flexibility for containers with 'inner structure'.
If these checks are considered useful, we can bring them back
with a is_child vfunc that allows container implementations to
decided who they consider legitimate child.
This is a variant of gtk_widget_child_notify() that takes an
explicit container, instead of relying on widget->parent to
be the correct container to use.
With gtk-auto-mnemonics on, we hide mnemonics on focus out. We should also
check if the modifier is pressed on focus in and if so, show mnemonics again.
https://bugzilla.gnome.org/show_bug.cgi?id=618815
This reverts commit 1c46e04f30.
The change broke too many widgets that relied on the size being
constant. A proper fix would require letting themes override the size.
That would probably also require letting themes specify the size
relative to font size.
I was hunting a memory leak and couldn't find it; at least I'm
pretty sure all of these are OK. But document things better
for the future.
Also use g_hash_table_replace in one more case for consistency.
https://bugzilla.gnome.org/show_bug.cgi?id=649457
And the getter, too: gtk_menu_item_get_right_justified() and the
corresponding property. Also make the only caller use the private
structure (as it did before the recent patches).
Fixes a regression introduced in 07d49ee5.
Libtool by default refuses to link static libraries into shared
libraries. In Windows, libuuid is however a static library and needed
for shared libgtk; as a work around, use "-Wl,-luuid" to pass the option
directly to the linker.
https://bugzilla.gnome.org/show_bug.cgi?id=642214
