358871 - "ASSERTION: The cellposition should have been cleared" and "ASSERTION: replacing a non dead cell is a memory leak" with colspan=0 and other spans

Status: RESOLVED FIXED

(Core :: Layout: Tables, defect)

Description

[Jesse Ruderman]

###!!! ASSERTION: The cellposition should have been cleared: 'origData->IsDead()', file /Users/admin/trunk/mozilla/layout/tables/nsCellMap.cpp, line 1373

###!!! ASSERTION: replacing a non dead cell is a memory leak: 'origData->IsDead()', file /Users/admin/trunk/mozilla/layout/tables/nsCellMap.cpp, line 1410

Comment 1

[Jesse Ruderman]

Attachment: testcase

Comment 2

[Ryan Jones-Ward [:sciguyryan]]

This actually crashes for me on Windows XP.

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20061030 Minefield/3.0a1 - Build ID: 2006103004

Anyone else seeing thuis?

Comment 3

[Bernd]

probably your build is too old ;-), the checkin to see the assert instead of the crash was at 2006-10-30 11:21

Comment 4

[Bernd]

Attachment: patch

Comment 5

[Bernd]

Comment on attachment 244240 [details] [diff] [review]
patch

handle new cells identical as the code section above does.

Comment 6

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 244240 [details] [diff] [review]
patch

I suppose, but is it documented somewhere what those two booleans mean?  If so, do those comments match what we actually do? That is, do we document that if the cell is due to being spanned by a colspan but _not_ a rowspan we should not set ZeroRowSpan to true?

Comment 7

[Bernd]

this was fixed on trunk 2006-11-01 10:45

Comment 8

[Jesse Ruderman]

Crashtest checked in.