Bug 67221 – [chromium] Fix scissor rects on clipped nested iframes

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Adrienne Walker]

[chromium] Fix scissor rects on clipped nested iframes

[Adrienne Walker]

Created attachment 105669 [details]
Patch

[Adrienne Walker]

See: http://code.google.com/p/chromium-os/issues/detail?id=19635

[Kenneth Russell]

Comment on attachment 105669 [details]
Patch

Vangelis, could you unofficially review this?

[Vangelis Kokkevis]

Comment on attachment 105669 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=105669&action=review

Good catch!

> Source/WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:264
>                  scissor.intersect(layer->scissorRect());

Or you could simply remove the if() and if layer->scissorRect().isEmpty() then the intersection will be empty too.

[Vangelis Kokkevis]

(In reply to comment #4)
> (From update of attachment 105669 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=105669&action=review
> 
> Good catch!
> 
> > Source/WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:264
> >                  scissor.intersect(layer->scissorRect());
> 
> Or you could simply remove the if() and if layer->scissorRect().isEmpty() then the intersection will be empty too.

Hmm, unless of course we're worried about scissor rects that due to the transform applied to them end up with negative sizes.  Maybe the code you have is safer.

[Adrienne Walker]

> (In reply to comment #4)
> > (From update of attachment 105669 [details] [details])
> > View in context: https://bugs.webkit.org/attachment.cgi?id=105669&action=review
> > 
> > Good catch!
> > 
> > > Source/WebCore/platform/graphics/chromium/LayerRendererChromium.cpp:264
> > >                  scissor.intersect(layer->scissorRect());
> > 
> > Or you could simply remove the if() and if layer->scissorRect().isEmpty() then the intersection will be empty too.
> 
> Hmm, unless of course we're worried about scissor rects that due to the transform applied to them end up with negative sizes.  Maybe the code you have is safer.

If we have IntRects with negative sizes, I think we'll be in more trouble than just here.  I agree with your original comment and will simplify this.  :)

[Adrienne Walker]

Created attachment 105794 [details]
Patch

[Adrienne Walker]

(In reply to comment #7)
> Created an attachment (id=105794) [details]
> Patch

Can I get an unofficial (or official) review whenever one of y'all has time? I would love to land this before the m15 branch.  :)

[Vangelis Kokkevis]

Comment on attachment 105794 [details]
Patch

unofficial r+ from me!

[Adrienne Walker]

Committed r94315: <http://trac.webkit.org/changeset/94315>