Closed Bug 63703 Opened 25 years ago Closed 25 years ago

Race condition in table alignment

Categories

(Core :: Layout: Tables, defect)

x86
Windows ME
defect
Not set
normal

Tracking

()

VERIFIED WORKSFORME

People

(Reporter: bugzilla, Assigned: karnaze)

References

()

Details

(Keywords: css2, regression)

Attachments

(1 file)

Build ID: 12/25 trunk (tip) Steps to Reproduce: (1) Go to http://bugzilla.mozilla.org/show_bug.cgi?id=63687 (or any Bugzilla bug). (2) Look at the amount of space between "Description:" and "Opened [date time]". Result: the two cells are now close together, instead of aligned at the left and right ends of the viewport. Relevant code: <table><tr><td align=left>Left</td> <td align=right width=100%>Right</td></tr></table> This works in a 12/13 nightly, so it regressed sometime since then (when I get more time, I'll track it down further). This is, amazingly enough, a race condition, which leads me to believe that this might be a parser bug (dropping the |align| attribute sometimes?). Continuously reload the testcase I'll attach and watch as the cells sometime align correctly and sometimes don't. Strangely, they never seem to align properly if I keep reloading a Bugzilla bug...
Attached file testcase
Keywords: nsbeta1, regression
[note that it's easier to see them align properly if you save the testcase locally and continuously reload it]
This seems like it could be related to bug 63445.
Can't see anything wrong after ten reloads of the testcase in linux 2000122406, however I am seeing wrong alignment on parts of this bug page... Create a new attactment it left-aligned with the input fields leftmost edge above it, and the depends on and blocks and votes texts are even further across to the right. In fact that whole section is aligned to the center of the page.
(not a parser bug, merely causing an incremental reflow can cause the table layout to change.) Note: Based on the number of times it was mentioned on IRC today, this bug is headed for mostfreq status pronto. ;-)
Bug 63945 could be reporting this too. Noting it here because the author of that bug has a couple more testcases, the latter of which he claims has slightly different behaviour.
Isn't this related to bug 63445??
This must have been fixed with that path of karnaze. I can't repro this on build 2001011220 on WinNT.
Don't see anymore either, marking wfm.
Status: NEW → RESOLVED
Closed: 25 years ago
Resolution: --- → WORKSFORME
Keywords: mozilla0.8
QA contact update
QA Contact: chrisd → amar
I dont see the problem anymore.. Marking as verified Build ID: 2001052720
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: