451518 - in an assertion ComputedWidth() checked twice for being >= 0

Status: RESOLVED FIXED

(Core :: Layout, defect)

Description

[andrzej.p]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; pl; rv:1.9.0.1) Gecko/2008070208 Firefox/3.0.1
Build Identifier: 

in layout/xul/base/src/nsBoxFrame.cpp in the beginning of nsBoxFrame::Reflow there is a NS_ASSERTION, which asserts aReflowState.ComputedWidth() >=0 && aReflowState.ComputedWidth() >= 0 (the same thing twice). It seems a bug. 

Reproducible: Always

Steps to Reproduce:
Review layout/xul/base/src/nsBoxFrame.cpp.
Actual Results:  
Found the suspicious line of code where the same thing is checked twice.

Expected Results:  
Probably should be aReflowState.ComputedWidth() >=0 && aReflowState.ComputedHeight() >= 0.

Comment 1

[Phil Ringnalda (:philor)]

Attachment: Fix v.1

I'm untroubled by any actual understanding of the code, but I'm betting this is what you meant to do in bug 367442, since it's what you did in the copy of the same code in nsLeafBoxFrame.cpp.

Comment 2

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 335010 [details] [diff] [review]
Fix v.1

You want ComputedHeight(), not mComputedHeight().  This patch wouldn't compile as written...

Good catch, though!  and r+sr with the 'm' removed.

Comment 3

[Phil Ringnalda (:philor)]

Attachment: for checkin

Well, the good news is that now my script to build debug actually does, instead of building opt in my opt objdir and then changing to the debug objdir, apparently hoping (and apparently with good reason) that I wouldn't notice what it had done.

Comment 4

[Boris Zbarsky [:bzbarsky]]

Pushed changeset 7615810bd835