1108887 - Back out initial SVG iframe implementation

Status: RESOLVED FIXED

(Core :: SVG, defect)

Description

[Brian Birtles (:birtles, Feb 26~ parental leave)]

In bug 949435 we started implemented a native SVG iframe element. After further discussion with the SVG WG and HTML people this no longer seems like a good idea and instead we want to pursue allowing HTML iframe in SVG.

We should backout the parts of bug 949435 that landed since that no longer seems to be the way ahead and adds unnecessary complexity to the code base.

Comment 1

[Robert Longson [:longsonr]]

Attachment: parts3and4.txt

Looks like this originally landed in 4 parts per bug 949435 comment 104

Here's a backout for parts 3 and 4.

Parts 1 and 2 seem rather more difficult to backout as the code has changed.

Comment 2

[Robert Longson [:longsonr]]

Attachment: part2.txt

Comment 3

[Brian Birtles (:birtles, Feb 26~ parental leave)]

Comment on attachment 8560853 [details] [diff] [review]
part2.txt

>+NS_IMETHODIMP
>+nsGenericHTMLFrameElement::SetIsPrerendered()
>+{
>+  MOZ_ASSERT(!mFrameLoader, "Please call SetIsPrerendered before frameLoader is created");
>+  mIsPrerendered = true;
>+  return NS_OK;
>+  return NS_OK;
>+}

Repeated line here.

Comment 4

[Robert Longson [:longsonr]]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=12858957a71a (parts 3 and 4 only)

Comment 5

[Robert Longson [:longsonr]]

Comment on attachment 8560619 [details] [diff] [review]
parts3and4.txt

Needs a DOM peer review as it contains a webidl change (the complete backout of SVGIFrameElement.webidl)

Comment 6

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8560853 [details] [diff] [review]
part2.txt

>+nsGenericHTMLFrameElement::SetIsPrerendered()
>+  return NS_OK;
>+  return NS_OK;

Remove one of those, please.

r=me

Comment 7

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8560619 [details] [diff] [review]
parts3and4.txt

sr=me

Comment 8

[Robert Longson [:longsonr]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/ab8180c889da

Comment 9

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/ab8180c889da

Comment 10

[Robert Longson [:longsonr]]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=98abdce03ba7

https://hg.mozilla.org/integration/mozilla-inbound/rev/cf841de97d18

Comment 11

[Robert Longson [:longsonr]]

Attachment: part1.txt

Comment 12

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8562810 [details] [diff] [review]
part1.txt

Are we very sure that no SVG element will ever grow anything reflected as DOMSettableTokenList?

Comment 13

[Boris Zbarsky [:bzbarsky]]

e.g. are we sure we'll never want .dropzone on SVG elements?

Comment 14

[Robert Longson [:longsonr]]

SVG 2 doesn't seem to have anything like this at the moment (i.e. nothing about drag/drop is in the SVG specification) and we could always yoyo it back again if necessary.

On the other hand if you'd rather leave things as they are, I'm happy to call this bug fixed at this point. The dead code for SVG iframes is all gone.

Comment 15

[Boris Zbarsky [:bzbarsky]]

I think I'd rather leave the code in Element.

Comment 16

[Robert Longson [:longsonr]]

I guess we're done then.

Comment 17

[Ryan VanderMeulen [:RyanVM]]

(In reply to Robert Longson from comment #10)
> https://treeherder.mozilla.org/#/jobs?repo=try&revision=98abdce03ba7
> 
> https://hg.mozilla.org/integration/mozilla-inbound/rev/cf841de97d18

https://hg.mozilla.org/mozilla-central/rev/cf841de97d18