It's 0x16, whereas I'd expect 0x10... But maybe things are depending on it including the EventNameType_SVGGraphic bit (likely) and the EventNameType_XUL bit (very unlikely)?
It's just wrong. 0x10 is right. r=longsonr if you change it.
But should SVGAnimationElement support the onfoo attributes other svg elements do? I think it should, for consistency.
They need to support this: http://www.w3.org/TR/SVG/script.html#AnimationEvents
I'm just worried that since we've shipped effectively with EventNameType_SVGGraphic | EventNameType_SMIL limiting to EventNameType_SMIL might break something. What do other browsers do here?
Note that Webkit doesn't support animation events at all.
Created attachment 681548 [details] [diff] [review] patch Only graphic things like path, circle etc should support EventNameType_SVGGraphic
In response to the "might break something" in comment 4: I think we should just fix this. There's a limited amount of content out there that uses SMIL, and only a fraction of that relies on SMIL-specific events (particularly considering comment 5), and only a fraction of that would hypothetically depend on this brokenness. In any (rare) cases where content accidentally does happen to rely on this bug, it's easy enough for the author to fix it.