If you think a bug might affect users in the 57 release, please set the correct tracking and status flags for Release Management.

Crash if you try to do an empty global regexp

RESOLVED INVALID

Status

()

Firefox
General
--
critical
RESOLVED INVALID
9 years ago
9 years ago

People

(Reporter: Dani, Unassigned)

Tracking

Firefox Tracking Flags

(Not tracked)

Details

(URL)

(Reporter)

Description

9 years ago
User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.0.1) Gecko/2008070208 Firefox/3.0.1
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.0.1) Gecko/2008070208 Firefox/3.0.1

I some entered text in the "Sample Text" text box, checked "Global" check box and firefox crashed

Reproducible: Always

Steps to Reproduce:
1.Write some text in the "Sample text" text box
2.Check "Global" check box
Actual Results:  
Firefox stopped working and crashed

Expected Results:  
"Global" check box checked
The bug appears to be in the web app itself. It implements the "Global" check box using a loop of its own instead of the JS global regexp feature. That loop runs one iteration per match found. It increments its start position each time through by the length of the match. Thus, if the regexp has a zero-length match, it iloops.

Following the STR on my trunk build (both Vista and OSX), I get the "stop script" dialog after the usual 10 seconds. I would think that would happen in 3.0.1 as well. It's hard to see how this could crash, unless your machine is running JS fast enough to create enough DOM nodes to go OOM before 10 seconds is up. Do you have a link to the crash report?

Otherwise, we should also contact the web site author to report the bug. Do you (Dani) want to do that or would you like me to?
(Reporter)

Comment 2

9 years ago
Hi,
I don't have the crash report because after about 30 sec my vista shows me the close program dialog.
My machine is pretty fast and i think it really goes OOM before 10 sec.
I still think its firefox bug because firefox shouldn't allow a script to go OOM.
I also think there is no point of telling this bug to the site author because the site is inactive 4 years.

Thanks for your help.
(In reply to comment #2)
> Hi,
> I don't have the crash report because after about 30 sec my vista shows me the
> close program dialog.

If the browser is crashing, I think it should try to record/submit a crash report, which you may be able to find by opening 'about:crashes'.

> My machine is pretty fast and i think it really goes OOM before 10 sec.
> I still think its firefox bug because firefox shouldn't allow a script to go
> OOM.

Yes, we would want to fix a browser hang or crash. But I can't reproduce it, including with Firefox 3.0.7 on my Vista Home Basic machine with 2.67 GHz Core 2 Duo and 4 GB RAM. Vista did put the "not responding" note in the title bar for a few seconds before the FF "stop scipt" dialog came up. 

Is the browser crashing or hanging all on its own, or are is Vista giving you an "unresponsive program" dialog and you are clicking the button that tells Vista to stop the program?
FYI, the site has been fixed. Can we set WORKSFORME?
How about INVALID...

/be
Status: UNCONFIRMED → RESOLVED
Last Resolved: 9 years ago
Resolution: --- → INVALID
You need to log in before you can comment on or make changes to this bug.