Commit Graph

33113 Commits

Author SHA1 Message Date
Benjamin Otte
a23bc88557 css: Fix dependencies for font lookup
'larger' and 'smaller' depend on the parent value. D'oh.
2012-12-01 16:27:28 +01:00
Kalev Lember
1a61e1ea23 gtkliststore: Add explicit braces to avoid dangling else
Otherwise the following else becomes attached to the wrong if.
2012-12-01 14:02:50 +01:00
Benjamin Otte
0d9d576172 css: Add more features to font-size code
We now support the keywords (like xx-small, medium, larger, smaller...)
and I've changed the default value to be "medium".

This required some shuffling of the "get default font size" code. But
all is well now.
2012-12-01 14:00:24 +01:00
Benjamin Otte
1dd3ee6b59 css: Move default font handling
The default font is no longer handled like a custom style sheet that
overrides everything, but as the initial value. This is the same
behavior as in web browsers.

And it allows the theme to actually use the 'font-family' and
'font-size' properties. Of course, a well behaved theme will respect the
setting as much as possible and for example use relative font sizes
(which aren't yet supported, but will be soon).
2012-12-01 01:49:06 +01:00
Benjamin Otte
3ff7f1fd43 style: Add _gtk_style_provider_private_get_settings()
This gives a GtkSettings object for resolving system-dependant things -
like the default font family and font size.
No code does this yet, but we have an API.

Only GtkSettings implements this.
2012-12-01 01:24:12 +01:00
Benjamin Otte
41f8ba3c35 settings: Implement StyleProvider interface without StyleProperties
I want to deprecate GtkStyleProperties.
2012-12-01 00:56:12 +01:00
Benjamin Otte
fe61e6b3fe settings: Never override some font properties
According to the CSS specification, everything but font-size and
font-family are defined by CSS. So we don't override them anymore.
2012-12-01 00:18:01 +01:00
Benjamin Otte
be53513954 API: Deprecate functions to register custom CSS properties
We should slowly transition code to using the GTK-provided CSS. So here
we go.
2012-11-30 22:45:20 +01:00
Benjamin Otte
fc67f0d4f8 stylecontext: Remove unused private functions 2012-11-30 22:41:22 +01:00
Benjamin Otte
0c12a6ffcb cssstyleproperty: Don't mark border width properties as NO_RESIZE
I was to overeager when marking properties in
66d22f101d apparently.
2012-11-30 20:41:58 +01:00
Benjamin Otte
35362337ec widget: Optimize away unndeeded redraws
If only styles were updated that don't require resizes, don't queue one.
2012-11-30 20:40:18 +01:00
Benjamin Otte
8560ff88cd styleproperty: Add _gtk_css_style_property_changes_affect_size()
This gives fast access to the question of wether a style update requires
a resize.
2012-11-30 20:37:46 +01:00
Benjamin Otte
e0586e3680 stylecontext: Expose a function to get at the current changes 2012-11-30 20:36:57 +01:00
Benjamin Otte
66d22f101d styleproperty: Mark properties as not needing resizes 2012-11-30 20:10:23 +01:00
Benjamin Otte
a5770cef36 styleproperty: Add an affects-size property
This property will be used to avoid gtk_widget_queue_resize() calls in
favor of gtk_widget_queue_draw().
2012-11-30 20:10:23 +01:00
Benjamin Otte
4ccb8e5d33 stylecontext: Pass a bitmask to the changed values 2012-11-30 20:10:23 +01:00
Benjamin Otte
cc0b493216 reftests: Fix a reftest
Recent Adwaita changes caused diffs. We don't want that, so we run this
test against the default settings.
2012-11-30 20:10:23 +01:00
Alexander Larsson
3c279b1492 icon cache: Fix double free
The symbolic icon cache code free crashed with a double free
due to a missing ->next in the loop.
2012-11-30 16:19:13 +01:00
Michael Natterer
9eea724e3b gtk: convert keyvals to unicode before committing to the imcontext
(cherry picked from commit 1eb0d98d73)
2012-11-30 15:10:51 +01:00
Michael Natterer
f619f91a8f gtk: add more OSX special casing for the deadacute and deaddoubleacute keys
So " plus foo prduces foo-with-diaereses and ' plus c produces
c-with-cedilla.
(cherry picked from commit c0102b3089)
2012-11-30 15:09:46 +01:00
Alexander Larsson
e36f80f0ef css: Make test pass
The new css tree may change the order of selectors (keeping the
same semantics). This affects how the selectors are printed later,
which causes some css parsing tests to not match the references.

Fortunately the order is consistent between runs given the same
css, so we just have to switch around the order in some of the
.ref.css files.
2012-11-30 14:53:30 +01:00
Alexander Larsson
3c421db473 css: Allocate the css tree in a single chunk
This gives us several advantages:

* Smaller struct on 64bit (32bit indexes vs 64bit pointers)
* Less overhead for allocation
* Less fragmentation
2012-11-30 14:53:30 +01:00
Alexander Larsson
2dbeabdc58 css: Fix leak of lists while building tree 2012-11-30 14:53:30 +01:00
Alexander Larsson
528e6032ae css: Add accessor functions for traversing css tree
This will let us later change how the tree is stored
2012-11-30 14:53:30 +01:00
Alexander Larsson
42716cb60f css: Add const to _gtk_css_selector_tree_match_all arg 2012-11-30 14:53:30 +01:00
Alexander Larsson
d69c6fff7a css: Don't keep around linear selectors
Now we use the selector tree everywhere, so there is no need to
keep around the linear selectors unless we're using them to
verify the tree correctness, so free them.
2012-11-30 14:53:30 +01:00
Alexander Larsson
c10a0962cf css: Remove gtk_css_ruleset_matches
This is only not needed anymore, and only the VERIFY_TREE code
should access ->selector.
2012-11-30 14:53:30 +01:00
Alexander Larsson
fd4712590a css: Use tree for gtk_css_provider_get_style_property 2012-11-30 14:53:30 +01:00
Alexander Larsson
9b989a1427 css: Implement ruleset_get_change() with the tree
We traverse the tree on the matches instead of using
the linear selectors.
2012-11-30 14:53:30 +01:00
Alexander Larsson
146b6f3a93 css: Ensure the tree built is always the same
We add some "artificial" ordering to the otherwise unordered
tree nodes. This means the tree will be the same every time for the
same input. This is good because e.g. tree order affects the
reordering of the simple selectors, which may affect how
css providers are printed, which need to be consistent for
the css tests to work.
2012-11-30 14:53:30 +01:00
Alexander Larsson
3d5e8f98ec css: Use the tree to print css selectors 2012-11-30 14:53:30 +01:00
Alexander Larsson
6bec577771 css: Track the tree selector matches 2012-11-30 14:53:30 +01:00
Alexander Larsson
f678827918 css: Fix type of GtkCssSelectorRuleSetInfo match
The old type was a leftover from a previous version.
2012-11-30 14:53:30 +01:00
Alexander Larsson
6b85db7555 css: Track parent in the css tree nodes
This way we can reconstruct matched css rules
2012-11-30 14:53:30 +01:00
Alexander Larsson
69feabd139 css: Better tree match verification 2012-11-30 14:53:30 +01:00
Alexander Larsson
0ede06d229 css: Fix up position with region tree matching
This was using the wrong result in case of a match (results from
the position, not the region. Also, the descendant checks were
wrong.
2012-11-30 14:53:30 +01:00
Alexander Larsson
d6932dd823 css: Don't reorder some selectors when building selector tree
When building the tree we generally reorder the selectors inside
the same simple selector in order to pick a good first selector
to balance the tree better. However, some kinds of selectors
can't really be reordered, even thought they are simple.

This is since the matching code for some types handle
the existance of a directly preceeding selector differently:

 REGION and ANY selectors look for a DESCENDANT previous
 POSITION selector look for a REGION previous
2012-11-30 14:53:30 +01:00
Alexander Larsson
63cb467e06 css: Fixed typo in PRINT_TREE debug code 2012-11-30 14:53:30 +01:00
Alexander Larsson
1bfa6593c8 css: Create and use a tree for css selector matching 2012-11-30 14:53:30 +01:00
Alexander Larsson
daefb12a23 css: Add GtkCssSelectorTree creation and matching
From a set of GtkCssSelectors and the rulesets they match to
we create a large decision tree that lets us efficitently match
against all the rules and return the set of matched rulesets.

The tree is created such that at each level we pick the initial rule[1]
in all the considered selectors for that level and use put the
one that is in most selectors in the node. All selectors matching that
are put in the previous part of the tree.
2012-11-30 14:53:29 +01:00
Alexander Larsson
16f2b20f96 css: Add _gtk_css_matcher_matches_any()
This returns true if the matcher matches *anything*. We need
to check this later, because such matchers are dangerous in loops
that iterate over all parents/siblings since such loops would not
terminate.
2012-11-30 14:53:29 +01:00
Alexander Larsson
3b4040d619 css: Track which selectors are "simple" 2012-11-30 14:53:29 +01:00
Alexander Larsson
cfdc68db5c Reuse rendered symbolic icons
With the previous commit all loads of the same icon will share a single
GtkIconInfo, which typicallty means the pixbuf is shared via Info->pixbuf.

However, atm we don't share symbolic icons, which causes these to be re-read
and re-parsed every time. This is especially bad if the icon is used many times
in some form of list. So, we cache the pixbufs and reuse them.

https://bugzilla.gnome.org/show_bug.cgi?id=689081
2012-11-30 11:33:26 +01:00
Alexander Larsson
92e904a257 Cache GtkIconInfo
In order to avoid loading and keeping around the same icon multiple times
we keep a cache of all outstanding GtkIconInfo objects for a given theme.

Additionally we return to the app not the normal pixbuf from the info,
but rather a proxy copy of it sharing the same data, but no extra
reference. This allows us to track when the app is no longer using
the pixbuf, and we can thus ensure that the GtkIconInfo in the cache
stays around for at least as long as the pixbuf is alive.

When the app unrefs the pixbuf we put the Info on a short LRU list
to keep it alive a bit longer, in case the app needs it in a short
while.

https://bugzilla.gnome.org/show_bug.cgi?id=689081
2012-11-30 11:33:26 +01:00
Matthias Clasen
e75860e466 Add a missing va_end call
va_copy must be matched with va_end.
2012-11-29 23:18:49 -05:00
Matthias Clasen
5b9c687595 Avoid unreachable code
Reestablish an else that went missing in commit 5ff328d2. Without
it, the return is reached unconditionally, and the code behind
it is dead.
2012-11-29 23:10:40 -05:00
Matthias Clasen
b3b83545c4 Don't return prematurely
This code was clearly meant to return _after_ the loop.
Don't put the return inside the loop body, then.
2012-11-29 23:10:35 -05:00
Matthias Clasen
c5c7323942 GtkAboutDialog: Initialize a variable
Coverity complains about it being uninitialized, otherwise.
2012-11-29 23:10:34 -05:00
Matthias Clasen
52af9aede6 GtkNotebook: Clarify a switch statement
We were intentionally falling through here, this commit
just adds comments to make that explicit.
2012-11-29 23:10:34 -05:00
Matthias Clasen
4557e4d7a5 GtkCellAccessible: Add missing break statements
This switch was entirely breakless. Amazing accessibility code,
found by Coverity.
2012-11-29 23:10:34 -05:00