343935 - "ASSERTION: no common ancestor at all???" or "ASSERTION: reflow dirty lines failed" involving <xul:textbox> with <html:input> inside

Status: RESOLVED WORKSFORME

(Core :: Layout: Form Controls, defect)

Description

[Jesse Ruderman]

Tested with a Mac trunk debug build.

Steps to reproduce:

1. Load the testcase.

Result: ###!!! ASSERTION: no common ancestor at all???: 'parent', file /Users/admin/trunk/mozilla/layout/base/nsLayoutUtils.cpp, line 344

Also, there are two long green textboxes, which doesn't seem right if you look at the source.

2. Click the top green textbox.
3. Type 'x'.

Result: Contrary to what you might expect, no 'x' appears.

4. Click near the top of the lower green textbox (but not on its border).

Result: Crash dereferencing a null nsCOMPtr<nsIContent> in GetRangeForFrame called from nsIFrame::GetContentOffsetsFromPoint.

Comment 1

[Jesse Ruderman]

Attachment: testcase

Comment 2

[Jesse Ruderman]

Attachment: testcase 2 -- remove nem instead of moving it

This causes different output:

WARNING: Content has no document.: file /Users/admin/trunk/mozilla/layout/generic/nsTextFrame.cpp, line 5956

WARNING: Reflow of frame failed in nsLineLayout: file /Users/admin/trunk/mozilla/layout/generic/nsLineLayout.cpp, line 998

###!!! ASSERTION: reflow dirty lines failed: 'NS_SUCCEEDED(rv)', file /Users/admin/trunk/mozilla/layout/generic/nsBlockFrame.cpp, line 916

Comment 3

[Martin Mokrejs]

Attachment: gdb stack traces of two such issues

I can confirm the same issue here on Linux with current cvs HEAD as of today. I hit it twice in few minutes when testing the newly compiled binary.

Comment 4

[Alex Vincent [:WeirdAl]]

I've been hitting this assertion recently.  Running testcase 1, though, I do not crash with SM trunk.

Comment 5

[Alex Vincent [:WeirdAl]]

(sorry, I should've clarified: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20061026 SeaMonkey/1.5a)

Comment 6

[Jesse Ruderman]

No longer crashes (Mac trunk debug).  The assertions still happen.

Comment 7

[Martijn Wargers (dead)]

With current debug trunk build, I'm only getting a "reflow dirty lines failed" assertion with testcase 2.

Comment 8

[Bob Clary [:bc] (inactive)]

Jesse, I can't reproduce this on trunk mac os x. Can we -> WFM?

Comment 9

[Jesse Ruderman]

WFM here too.  Thanks.