[Webkit-unassigned] [Bug 23147] Introduce Skia to WebKit

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=23147


eric at webkit.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #26472|review?(eric at webkit.org)    |review-
               Flag|                            |




------- Comment #3 from eric at webkit.org  2009-01-06 14:18 PDT -------
(From update of attachment 26472)
We don't use SVGPaintServerGradientSkia anymore.  GradientSkia.cpp should be
taking care of this.  There is a x-platform implementation of
SVGPaintServerGradient which uses Gradient from platform instead.

SVGPaintServerPatternSkia is slated to die (the same death as
SVGPaintServerGradient*, but it looks like that patch hasn't landed yet).

I'm surprised we don't have a nicer way than this:
+        IntRect textBoundary =
const_cast<RenderObject*>(object)->absoluteBoundingBoxRect();
+        targetRect =
object->absoluteTransform().inverse().mapRect(textBoundary);
it seems CG does the same thing though. :(

We use 0 instead of NULL for c++ code according to the style guidelines:
+    context->platformContext()->setGradient(NULL);

We don't generally commit ifdef'd out code, so this should be removed:
+  // TODO(jhaas): implement me
+#if 0
+    cairo_t* cr = context->platformContext();
+    cairo_surface_t* surface = mask()->surface();
+    if (!surface)
+        return;
+    cairo_pattern_t* mask = cairo_pattern_create_for_surface(surface);
+    cairo_mask(cr, mask);
+    cairo_pattern_destroy(mask);
+#endif

Otherwise it looks fine.  I think you can delete the
SVGPaintServerGradientSkia* files from the build and it should still work.


-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.