823227 - Make the Global Scope Polluter work for non-html scopes

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Boris Zbarsky [:bzbarsky]]

Now that we always have a gsp, we should just do named frame lookups from it.  That's what the spec says to do anyway, and that way we don't have to have weird Window behavior.

Comment 1

[Boris Zbarsky [:bzbarsky]]

Attachment: part 1. Install a global scope polluter even if we don't have an HTML document.

Comment 2

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 724747 [details] [diff] [review]
part 1.  Install a global scope polluter even if we don't have an HTML document.

I'm almost surprised that we already null check everywhere that we need to.

Comment 3

[Boris Zbarsky [:bzbarsky]]

Attachment: Now with a test

Comment 4

[Bobby Holley (:bholley)]

I fixed the bug as-filed in bug 850517, but now we're running into problems with the GSP not existing for non-HTML documents. I'm going to use this bug for that, starting with bz's patches.

Comment 5

[Bobby Holley (:bholley)]

Comment on attachment 726216 [details] [diff] [review]
Now with a test

Given that I think we're in a bit of a hurry on this one, I'm going to do the try push and review requests simultaneously.

Comment 6

[Bobby Holley (:bholley)]

Attachment: Part 1 - Have the global scope polluter find the document via the global. v1

Comment 7

[Bobby Holley (:bholley)]

Attachment: Part 2 - Resolve named frames for non-HTML documents. v1

Comment 8

[Bobby Holley (:bholley)]

Attachment: Part 3 - Stop pref-controlling the GSP. v1

Comment 9

[Bobby Holley (:bholley)]

Attachment: Part 4 - Tests. v1 r=bholley

Comment 10

[Bobby Holley (:bholley)]

https://tbpl.mozilla.org/?tree=Try&rev=ea5e68edd663

Comment 11

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 726351 [details] [diff] [review]
Part 1 - Have the global scope polluter find the document via the global. v1

>+  nullptr

Could just leave it off, too.  Either way.

>+  nsCOMPtr<nsIDOMHTMLDocument> domDoc =

nsIHTMLDocument, please.  Might not be worth worrying about if this code goes away.


r=me

Comment 12

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 726352 [details] [diff] [review]
Part 2 - Resolve named frames for non-HTML documents. v1

Definitely nsIHTMLDocument here, please.  r=me

Comment 13

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 726353 [details] [diff] [review]
Part 3 - Stop pref-controlling the GSP. v1

This is the part mrbkap already reviewed.  So r=, but him not me.  ;)

Comment 15

[Alex Keybl [:akeybl]]

Tracking due to bug 851924, which points to a concerning web regression.

Comment 16

[Bobby Holley (:bholley)]

Arg, I wrote "linu64" rather than "linux64", so no try jobs were triggered. Pushing again:

https://tbpl.mozilla.org/?tree=Try&rev=02b3830e01c3

Comment 17

[Bobby Holley (:bholley)]

Green on try.

remote:   https://hg.mozilla.org/integration/mozilla-inbound/rev/34511e89a1a8
remote:   https://hg.mozilla.org/integration/mozilla-inbound/rev/96fc6637a942
remote:   https://hg.mozilla.org/integration/mozilla-inbound/rev/0b6a4443faa1
remote:   https://hg.mozilla.org/integration/mozilla-inbound/rev/b2942eb4d283

Comment 18

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/34511e89a1a8
https://hg.mozilla.org/mozilla-central/rev/96fc6637a942
https://hg.mozilla.org/mozilla-central/rev/0b6a4443faa1
https://hg.mozilla.org/mozilla-central/rev/b2942eb4d283