[Webkit-unassigned] [Bug 39027] http://philip.html5.org/tests/canvas/suite/tests/2d.composite.uncovered.fill.source-in.html fails

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=39027





--- Comment #16 from Julien Chaffraix <jchaffraix at webkit.org>  2011-05-13 20:00:15 PST ---
(From update of attachment 92671)
View in context: https://bugs.webkit.org/attachment.cgi?id=92671&action=review

>> Source/WebCore/html/canvas/CanvasRenderingContext2D.cpp:945
>> +        // our test coverage is enough.
> 
> i'm not sure exactly what this FIXME means.  should this reference a webkit bug?

It was more like a gut feeling as philip's test suite does not cover source-atop as it does with source-in or source-out (unless I missed something). I had a quick look and we cover some of source-atop in canvas-composition-alpha.html. I can either squash this change or use a FIXME with a new bug. Do you have a preference?

>> Source/WebCore/html/canvas/CanvasRenderingContext2D.cpp:947
>> +            || state().m_globalComposite == CompositeSourceOut) {
> 
> this should be on one line

Done.

>> Source/WebCore/html/canvas/CanvasRenderingContext2D.cpp:1068
>> +        || state().m_globalComposite == CompositeSourceOut) {
> 
> same comments as above

Done.

>> Source/WebCore/html/canvas/CanvasRenderingContext2D.cpp:1077
>> +        c->restore();
> 
> this smells like bad copy-pasta.  can you refactor this into a helper?

OK, I will have to use some template magic as rect and m_path are different types but it makes sense.

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