[Webkit-unassigned] [Bug 49437] SVGAnimations with IRI references via 'xlink:href' are slow

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Fri Nov 12 06:17:10 PST 2010


https://bugs.webkit.org/show_bug.cgi?id=49437





--- Comment #4 from Dirk Schulze <krit at webkit.org>  2010-11-12 06:17:11 PST ---
(In reply to comment #2)
> (From update of attachment 73731 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=73731&action=review
> 
> Hm, I'm slightly worried that we have no new tests about this.
Yes, asking Sean if we might upload his test. I tried to create a test my self, but couldn't reproduce such a perf issue like on his test.

> What happens when you change 'target' dynamically via JS, while the animation is running? I guess m_targetElement won't be recached from reading the code...
> 
Bug 12065 has a test for it and I plan to upload as a patch on this bug.

> I think you should try that and compare to Opera.
> 
> > WebCore/svg/SVGAnimateMotionElement.cpp:55
> > -bool SVGAnimateMotionElement::hasValidTarget() const
> > +bool SVGAnimateMotionElement::hasValidTarget()
targetElement() is not const anymore, since I save the reference to the target in m_targetElement, this caused these functions to be non-const.

> 
> Why is it a non-const version?
> 
> > WebCore/svg/SVGAnimateTransformElement.cpp:106
> > +    SVGElement* targetElement = this->targetElement();
> 
> I'd rather use const_cast<SVGAnimateTransformElement*>(this)->targetElement() here!
Well, I'm not a friend of const_cast and we would need it nearly everywhere, where we call targetElement(), like in all hasValidTarget functions.

The patch did not apply because of devregion=english changes in XCode :-(

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



More information about the webkit-unassigned mailing list