Commit Graph

22 Commits

Author SHA1 Message Date
Alexander Larsson
e60c9219f1 Make _gtk_css_lookup_get_missing inline
This is called a lot in the loop in gtk_css_style_provider_lookup which
actually showed up on profiles.
2012-11-09 10:16:23 +01:00
Benjamin Otte
1454ba15ba css: Huge refactoring to avoid computing wrong values
Here's the shortest description of the bug I can come up with:
When computing values, we have 3 kinds of dependencies:
(1) other properties ("currentColor" or em values)
(2) inherited properties ("inherit")
(3) generic things from the theme (@keyframes or @define-color)
Previously, we passed the GtkStyleContext as an argument, because it
provided these 3 things using:
(1) _gtk_style_context_peek_property()
(2) _gtk_style_context_peek_property(gtk_style_context_get_parent())
(3) context->priv->cascade

However, this makes it impossible to lookup values other than the ones
accessible via _gtk_style_context_peek_property(). And this is exactly
what we are doing in gtk_style_context_update_cache(). So when the cache
updates encountered case (1), they were looking up the values from the
wrong style data.

So this large patch essentially does nothing but replace the
context argument in all compute functions with new arguments for the 3
cases above:
(1) values
(2) parent_values
(3) provider

We apparently have a lot of computing code.
2012-09-28 18:27:49 +02:00
Benjamin Otte
1d7b23845f css: Pass relevant properties to css lookup code
This is not used yet in this patch.
2012-08-28 15:42:24 +02:00
Benjamin Otte
6fbd8b0003 computedvalues: Pass dependencies to set_value()
This simplifies the code.
2012-08-28 15:42:24 +02:00
Alexander Larsson
7603e6e473 css: Use GtkCssValues instead of GValue in the css machinery
Also, in places where we're computing a new CssValue based on an
old one, make sure that if nothing changes we're returning a reference
to the old one, rather than creating a new identical instance.
2012-03-08 11:03:57 +01:00
Benjamin Otte
64425a28de bitmask: Make setters return a new value
(Actually, it's not a real 'new' value yet, but will be soon.

This is the first step to make bitmasks immutable.
2012-03-01 15:10:36 +01:00
Javier Jardón
9d0febc9a6 Change FSF Address 2012-02-27 17:06:11 +00:00
Benjamin Otte
6ba33e7af5 css: Allow bypassing the compute_value() stage
This is useful when overriding values.
2012-02-02 03:13:38 +01:00
Benjamin Otte
e4c2d9b259 csslookup: Query the bitmask
Checking if the value is NULL is the wrong thing to do - the bitmask is
usd to keep track of that.

The reason for that will become apparent in the next patch.
2012-02-02 03:13:38 +01:00
Benjamin Otte
6962b49a99 css: Introduce GtkCssComputedValues
To be used for storing computed values. Is the replacement for
GtkStyleProperties, which is now legacy code.
2012-01-11 15:48:54 +01:00
Benjamin Otte
84d294ee7f stylecontext: NMake safe to call into during lookup
When resolving a lookup, we may want to query the current style context,
as in the next patch. This works now.
2012-01-09 18:37:57 +01:00
Benjamin Otte
60b8b7a398 csslookup: Use the new gtk_style_context_peek_property() 2012-01-09 18:37:57 +01:00
Benjamin Otte
e87cf5d789 css: Redo value resolving
Instead of on-demand resolvage, we now resolve during lookup. The step
is done via
  _gtk_css_style_property_compute_value()
which currently calls into
  _gtk_css_style_compute_value()
That function has all the old resolving machinery.

The only part missing for now is the handling of win32 code. It will be
added back later.
2012-01-09 18:37:55 +01:00
Benjamin Otte
f2dc63cc6a css: Pass the real context to the CSS lookup
This will be necessary soon.
2012-01-09 18:37:55 +01:00
Benjamin Otte
30eb26087c css: Feed sections to CSS lookup code 2012-01-09 18:37:53 +01:00
Benjamin Otte
544d210a61 css: Move a bunch of functions
... from GtkStyleProperty to GtkCssStyleProperty.
2012-01-09 18:37:52 +01:00
Benjamin Otte
7a76694780 styleproperty: Use _gtk_style_property_get_name() 2012-01-09 18:37:52 +01:00
Benjamin Otte
4383701e25 styleproperty: unconstify
GtkStyleProperty is a real GObject now, so treat it like one and don't
use const.
2012-01-09 18:37:52 +01:00
Benjamin Otte
fbedcdaf7d css: Make CSS resolving work according to spec
See inline code comments taken from
  http://dev.w3.org/csswg/css3-cascade/#cascade

This now respects the special values "inherit" and "initial" properly.
Note that those values cannot be parsed yet. This will be added in a
future commit.
2012-01-09 18:37:50 +01:00
Benjamin Otte
242893a031 css: Pass the parent context to _gtk_css_lookup_resolve()
This way, we can resolve inherit properties.
2012-01-09 18:37:50 +01:00
Benjamin Otte
6dd50c6cca css: Add some docs to GtkCssLookup
In particular, document which parts of the CSS value querying we're
doing here.
2012-01-09 18:37:50 +01:00
Benjamin Otte
ba5e7012b1 css: Redo StyleProvider interface
We now use the GtkStleProviderPrivate interface, which hopefully is
faster and more conformant to CSS. Long term, it definitely should be
both.

I would have liked to split this up into multiple commits, but couldn't
find a way.
2012-01-09 18:37:50 +01:00