386690 - Negative width attribute and percentage height on <svg:foreignObject> causes "ASSERTION: reflow state made child wrong size" and more

Status: RESOLVED FIXED

(Core :: SVG, defect)

Description

[Jesse Ruderman]

Attachment: testcase

This is similar to bug 385246 (which is fixed).  The difference seems to be that the height attribute is a percentage height.

###!!! ASSERTION: reflow state made child wrong size: 'reflowState.ComputedWidth() == size.width', file /Users/jruderman/trunk/mozilla/layout/svg/base/src/nsSVGForeignObjectFrame.cpp, line 609

###!!! ASSERTION: unexpected size: 'size.width == desiredSize.width && size.height == desiredSize.height', file /Users/jruderman/trunk/mozilla/layout/svg/base/src/nsSVGForeignObjectFrame.cpp, line 615

Comment 1

[Jonathan Watt [:jwatt]]

I'm getting an insane number of "cannot call on a dirty frame not currently being reflowed" assertions, but that seems to be fairly normal as of late.

Are you still seeing these other assertions Jesse? I'm not.

Comment 2

[Jesse Ruderman]

Yes, I'm still seeing the assertions in comment 0.

Comment 3

[Jesse Ruderman]

Still happens on trunk.

Comment 4

[Jesse Ruderman]

WFM on trunk. No assertion failures.

Comment 5

[Jesse Ruderman]

Seeing this again (testing with mozilla-central rev 8777ed85c5ac).

Comment 6

[Robert Longson [:longsonr]]

Gone for me now. WFM for you? If so I could you check in the testcase before we screw it up again please Jesse?

Comment 7

[Jesse Ruderman]

WFM. Yes, I'll check in the testcase :)

Comment 8

[Daniel Holbert [:dholbert]]

FWIW, the Comment 5 --> Comment 6 change was almost certainly fixed by the addition of a "NS_MAX" call, mentioned in bug 596765 comment 17.

Updating to indicate this being FIXED by that bug.

Comment 9

[Jesse Ruderman]

Crashtest added:
http://hg.mozilla.org/mozilla-central/rev/d24f0712886d