495219 - Iframe window keeps old innerWidth/innerHeight values after iframe has style display: none

Status: VERIFIED FIXED

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

Description

[Martijn Wargers (dead)]

Attachment: testcase

See testcase, when clicking on the buttons, I get 300 and 150 as the innerWidth and innerHeight values for the iframe window. But it should be 0 and 0, because the iframe has style display: none.

This regressed between 2009-02-09 and 2009-02-11:
http://hg.mozilla.org/mozilla-central/pushloghtml?startdate=2009-02-09+12%3A00%3A00&enddate=2009-02-11+07%3A00%3A00
I think a regression from bug 469933.

This is also basically causing a repeat of bug 493658.

Comment 1

[Boris Zbarsky [:bzbarsky]]

Er, yes.  Sorry for missing that...  For web-facing stuff, the cases when we punted on the conversions due to lack of a prescontext probably still need to be punted on, since the docshell state can be stale.

Zack, can you take this?

Comment 2

[Zack Weinberg (:zwol)]

yah, i got it.

Comment 3

[Zack Weinberg (:zwol)]

Attachment: patch: report zero from nsGlobalWindow::GetInner{Width,Height} when no prescontext

This reverts nsGlobalWindow::GetInner{Width,Height} to their old behavior of silently returning zero if there's no prescontext.  That is sufficient to make the test case in attachment 380100 [details] happy; I've munged it into a mochitest and included it in the patch.

I did *not* revert nsGlobalWindow::SetInner{Width,Height}, which used to silently drop the change on the floor if there was no prescontext - do you think I should do that too?  (Can we come up with a test case that exhibits changed behavior for that?)

Comment 4

[Zack Weinberg (:zwol)]

How the heck did I become the QA contact?  Fixing.

Comment 5

[Boris Zbarsky [:bzbarsky]]

> (Can we come up with a test case that exhibits changed behavior for that?)

I don't believe so.

Comment 6

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 380202 [details] [diff] [review]
patch: report zero from nsGlobalWindow::GetInner{Width,Height} when no prescontext

Looks good.

Comment 7

[Mike Shaver (:shaver emeritus)]

Blocking -- please land _ASAP_ on m-c.

Comment 8

[Boris Zbarsky [:bzbarsky]]

Pushed http://hg.mozilla.org/mozilla-central/rev/701acdb5c460

Comment 9

[Boris Zbarsky [:bzbarsky]]

Pushed http://hg.mozilla.org/releases/mozilla-1.9.1/rev/34083fc75f1a

Comment 10

[Aakash Desai [:aakashd]]

verified FIXED via the attached testcase on builds:

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2a1pre) Gecko/20090727 Minefield/3.6a1pre (.NET CLR 3.5.30729) ID:20090727044731

and

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.2pre) Gecko/20090727 Shiretoko/3.5.2pre (.NET CLR 3.5.30729) ID:20090727044703