423436 - Elements embedding SVG by ref. don't resize if they depend on intrinsic percentage width/height

Status: RESOLVED FIXED

(Core :: SVG, defect, P2)

Description

[Jonathan Watt [:jwatt]]

Someone contacted me to complain that the new SVG site banner they're writing for Firefox 3 doesn't work. Their problem is that when the width of an <object> embedding SVG comes from the SVG's intrinsic width, we don't resize the <object> when the CB width increases if the SVG's intrinsic width is a percentage.

Apparently I forgot to implement nsSubDocumentFrame::GetIntrinsicSize despite putting in the checks for nsSubDocumentFrame having a percentage intrinsic width/height (bug 294086):

http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/layout/generic/nsLineLayout.cpp&rev=3.307&mark=732-734#726
http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/layout/generic/nsHTMLReflowState.cpp&rev=1.297&mark=412#404

Comment 1

[Jonathan Watt [:jwatt]]

Attachment: patch

Comment 2

[Jonathan Watt [:jwatt]]

This is a brain dead patch to fix a stupid and unfortunate rendering issue, so let's block on this.

Comment 3

[Damon Sicore (:damons)]

Does this patch require a beta cycle?  Must it block beta 5?

Side note, please, when determining blocker status, erase the knowledge from your brain that there is a patch.  The bug should block on merits of severity and impact alone, not on how fast it would be to fix.  :)

Comment 4

[Jonathan Watt [:jwatt]]

Fair enough, point taken. Still we should definitely take this IMO, but we can probably take it after beta 5.

Comment 5

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Comment on attachment 309971 [details] [diff] [review]
patch

r+sr=dbaron, but could you parenthesize the &&-ed stuff in the nsHTMLReflowState.cpp change so that it's clearer that there's an && in the middle of a long chain of ||s?

Comment 6

[Jonathan Watt [:jwatt]]

Sure. Checked in with that change.

I'm going to check in some reftests for this just as soon as I figure out why it seems to be okay to resize the window once in a reftest, but not a second time (to restore it to its original size for the tests that follow it).