837315 - Add an event for SVGs that indicates when the SVG and all its inner resources are done loading

Status: RESOLVED FIXED

(Core :: SVG, defect)

Description

[Seth Fowler [:seth] [:s2h]]

Right now there is no event that indicates that an SVG and all its inner images are done loading. We need something similar to |window.onload| for SVGs. The closest thing we have is the |SVGLoad| event, but it only indicates that the SVG DOM _itself_ is ready - that is, it behaves more like |DOMContentLoaded| than |window.onload|. We can't tell when SVG images should be considered loaded (and thus should unblock onload for the document they're embedded in) without this event. Not having this prevents us from correctly fixing bug 704059, resulting in a number of race conditions and random orange bugs, including bug 703806.

Comment 1

[Daniel Holbert [:dholbert]]

s/images/resources/ in summary -- this includes stylesheets, <use>, and any filters/gradients/etc that happen to be stored in (data-URI or blob-URI encoded) external resources.

Basically, we need something like window.onload. (but there are apparently a number of places where that gets blocked by the fact that we don't have a window object)

Comment 2

[Seth Fowler [:seth] [:s2h]]

Attachment: Add SVG document loaded event.

Proposed patch. Try job here: https://tbpl.mozilla.org/?tree=Try&rev=86372d515c8b

Comment 3

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 709864 [details] [diff] [review]
Add SVG document loaded event.

Hmm.  Is there a reason to not just call this event "load"?

r=me either way, I guess.

Comment 4

[Seth Fowler [:seth] [:s2h]]

(In reply to Boris Zbarsky (:bz) from comment #3)
> Hmm.  Is there a reason to not just call this event "load"?

Well, the thinking was that giving it a name like this would clarify that it was a private event only for our internal consumption.

Comment 5

[Seth Fowler [:seth] [:s2h]]

Thanks for the review btw!

Comment 6

[Boris Zbarsky [:bzbarsky]]

Ah, ok.  I buy that.

Comment 7

[Seth Fowler [:seth] [:s2h]]

OK, I'm happy with the try results. Let's get this in.

Comment 8

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/874fc2730371

Comment 9

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/874fc2730371