Comments on: SMIL work: performance and integration

by: Brian

Mon, 17 Oct 2005 08:15:43 +0000

No, I’m afraid not. There’s still a long way to go before declarative animation is ready. Hopefully we’ll have it by 2.0.

by: Ian

Mon, 17 Oct 2005 08:00:10 +0000

Will this be in 1.5?

by: Brian

Sun, 16 Oct 2005 21:33:33 +0000

Yes you’re right. This part definitely needs more thought. Relying on events also means that SVG documents added entirely by script won’t animate.

As for breaking scripts that add event listeners to the svg load event, this is already broken as load events are only dispatched if the onload attribute is specified. See:

http://lxr.mozilla.org/seamonkey/source/content/xml/document/src/nsXMLContentSink.cpp#1020

This will have to be reconsidered when bug 277955 is handled (’onload should respect externalResourcesRequired’).

https://bugzilla.mozilla.org/show_bug.cgi?id=277955

But now that you’ve raised it, I’ll look into other ways of starting animation.

Thanks smaug,

Brian.

by: smaug

Sun, 16 Oct 2005 11:25:44 +0000

I was just browsing through the patch…
The following doesn’t look right, someone may use svg.addEventListener(…) to handle
the svg load event.

+ if (aEvent->message == NS_SVG_LOAD) {
+ if (mAnimationRegistry)
+ mAnimationRegistry->Start();
+ if (!HasAttr(kNameSpaceID_None, nsSVGAtoms::onload))
+ return NS_OK;
+ }