[webkit-reviews] review denied: [Bug 65643] Repaint issues with -webkit-svg-shadow used on a container : [Attachment 103714] preliminary patch sans changelog/test

[bugzilla-daemon at webkit.org]

Nikolas Zimmermann <zimmermann at kde.org> has denied  review:
Bug 65643: Repaint issues with -webkit-svg-shadow used on a container
https://bugs.webkit.org/show_bug.cgi?id=65643

Attachment 103714: preliminary patch sans changelog/test
https://bugs.webkit.org/attachment.cgi?id=103714&action=review

------- Additional Comments from Nikolas Zimmermann <zimmermann at kde.org>
Tim, the idea behind this patch is correct. OTOH it slows down every common
case by requiring these additional tree traversal.
This needs to be optimized. We need to know whether a child needs this logic,
that means we have to cache somewhere if the parent of an arbitary child has
shadow or not.

Storing this inside SVGRenderStylle of the child is not an option, as styles
may be shared.
<g style="-webkit-svg-shadow: ...">
<rect/>
</g>
<rect/>

Both rects may share the same style, so we can't store the information whether
the parent is shadowed in the childs style.
The renderers are unique though, so it might be possible to stash this info
into RenderSVGContainer.

I'll bet you'll find an elegant solution :-)