Bug 9180

Summary: Case insensitivity flag in RegExps doesn't work on Intel-based Macs
Product: WebKit Reporter: Nick Piasecki <runningnick531>
Component: JavaScriptCoreAssignee: Nobody <webkit-unassigned>
Status: RESOLVED WORKSFORME    
Severity: Normal CC: ap, ddkilzer, ggaren
Priority: P2 Keywords: InRadar
Version: 417.x   
Hardware: Mac   
OS: OS X 10.4   
URL: javascript:alert( "UL".match(/ul/i ) )

Nick Piasecki
Reported 2006-05-30 21:38:21 PDT
Paste javascript:alert( "UL".match(/ul/i ) ) into your address bar. Opera 9 Beta 2, IE 6, and Firefox 1.5 return "UL". Safari 417.9.3 returns null.
Attachments
Add attachment proposed patch, testcase, etc.
Geoffrey Garen
Comment 1 2006-05-31 14:33:43 PDT
Works for me in Version 2.0.3 (418.3).
Nick Piasecki
Comment 2 2006-05-31 20:27:06 PDT
Great news. Should be noted, however, that this bug made it into the released 2.0.3 version (417.9.3). The bug, however, did not exist in previous builds. Is there a regression test to prevent this error from happening again?
David Kilzer (:ddkilzer)
Comment 3 2006-06-01 03:38:59 PDT
(In reply to comment #2) > Great news. Should be noted, however, that this bug made it into the released > 2.0.3 version (417.9.3). This works for me in Safari 2.0.3 (417.9.3) as well.
Nick Piasecki
Comment 4 2006-06-01 06:10:15 PDT
(In reply to comment #3) > (In reply to comment #2) > > Great news. Should be noted, however, that this bug made it into the released > > 2.0.3 version (417.9.3). > > This works for me in Safari 2.0.3 (417.9.3) as well. > Interesting. Could this be an Intel / PPC issue? I'm on an Intel Mac. (If it's not, I'll let it go and wait for the next release. Thanks for all of your help!)
David Kilzer (:ddkilzer)
Comment 5 2006-06-01 07:31:56 PDT
(In reply to comment #4) > Interesting. Could this be an Intel / PPC issue? I'm on an Intel Mac. (If it's > not, I'll let it go and wait for the next release. Thanks for all of your > help!) Ugh! I tested on a PowerPC PowerBook G4. This is very interesting (in a bad way). Reopening bug for now. I wonder what type of Mac Geoffrey used to test?
Nick Piasecki
Comment 6 2006-06-01 14:47:06 PDT
All right, did some more testing: Coworker's PPC iBook: returns "UL" on 2.0.3 (417.9.2) Coworker's PPC G5: returns "UL" on 2.0.3 (417.9.3) My Intel Macbook: returns null on 2.0.3 (417.9.3) Coworker's Intel Macbook: returns null on 2.0.3 (417.9.3) I think this definitely means that something's up with the 417.9.3 build on the newly released Intels. Can anyone give insight or more confirmation?
Geoffrey Garen
Comment 7 2006-06-01 15:53:12 PDT
Confirmed bug on MacBook Pro using Safari 417.9.3. Running under Rosetta fixes the issue. Seems to be an intel-specific bug.
Alice Liu
Comment 8 2006-06-01 16:03:04 PDT
<rdar://problem/4570881>
Alexey Proskuryakov
Comment 9 2006-06-10 03:34:30 PDT
I can reproduce this bug with stock 10.4.6 Safari/WebKit, but not with TOT (debug build) on a MacBook Pro. Any reason not to close as WORKSFORME?
Geoffrey Garen
Comment 10 2006-06-11 10:56:56 PDT
Why don't you try with a release build, just to be sure. Then I think we can close this bug.
Alexey Proskuryakov
Comment 11 2006-06-18 02:47:22 PDT
WFM with release build, too.
Note You need to log in before you can comment on or make changes to this bug.