Created attachment 65029 [details] HTML with inlineSVG and SMIL animation If you add SMIL animations in inlineSVG's for HTML5, they don't start animating.
Created attachment 65030 [details] standalone SVG with SMIL animation (animation runs)
Maybe this is all on SVGDocument instead of on an SVGTimeContainer (aka SVGSVGElement) like it should be. There is another bug about abstracting more of this code into a SVGTimeContainer class.
(In reply to comment #2) > Maybe this is all on SVGDocument instead of on an SVGTimeContainer (aka SVGSVGElement) like it should be. There is another bug about abstracting more of this code into a SVGTimeContainer class. I'm not very familiar with the SMIL code. But we have a SMILTimeContainer and it's all in SVGSVGElement. I guess you mean bug 9192. I would say SMILTimeContainer is your SVGTimeContainer class with the functionality you suggested. Need to debug it.
Hi Eric, (In reply to comment #2) > Maybe this is all on SVGDocument instead of on an SVGTimeContainer (aka SVGSVGElement) like it should be. There is another bug about abstracting more of this code into a SVGTimeContainer class. Dirk and I debugged this a bit, the old fashioned way using printfs :) Anyway my suspicion is that SVGSMILElement::finishParsingChildren() is not called, and therefore resolveFirstInterval is never called. Is it possible the new HTML5 parsing and tree building code does not call finishParsingChildren for non html/foreign elements? Cheers, Rob.
HTMLElementStack is responsible for calling finishParsingChildren. It's possible we missed one case.
I think we also fail to run SVG scripts. That might be related to this bug. Thanks for filing it.
(In reply to comment #6) > I think we also fail to run SVG scripts. That might be related to this bug. Thanks for filing it. Yes, looks like this. SVGElement::finishParsingChildren is also used for event handling (sendSVGLoadEventIfPossible).
Any updates here?
The bug is fixed in trunk, but we should add Dirks testcase first.
Attached HTML inline SVG.. test case works in Safari Technology Preview 160, Chrome Canary 111 and Firefox Nightly 110. Can we close this now?