Summary: | REGRESSION repro crash: asserction faliure in dom_position.cpp:494 (dev build only) | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | mitz | ||||
Component: | DOM | Assignee: | Maciej Stachowiak <mjs> | ||||
Status: | VERIFIED FIXED | ||||||
Severity: | Normal | ||||||
Priority: | P1 | ||||||
Version: | 412 | ||||||
Hardware: | Mac | ||||||
OS: | OS X 10.4 | ||||||
URL: | http://www.dhl.com/splash.html | ||||||
Attachments: |
|
Description
mitz
2005-06-20 14:22:19 PDT
Now it's OK again. (In reply to comment #1) > Now it's OK again. Actually it's still broken. Can't confirm this one, sorry. Works on ToT. (In reply to comment #3) > Can't confirm this one, sorry. Works on ToT. Works here part of the time, too, but always fails if it's the first site I go to after launching Safari. still can't confirm... Created attachment 2692 [details]
testcase
I can trigger the crash by opening this file in Safari an clicking inside the
yellow rect, somewhere to the left of the red arrow.
OK, a clean build seems to have solved it for real. Sorry. (In reply to comment #7) > OK, a clean build seems to have solved it for real. Sorry. Actually, that was a clean *deployment* build (where assert() does nothing?). A clean development build still crashes. I can reproduce this on ToT dev build -- may not be reproducible on deployment build as it's an assertion failure I can easily reproduce this with a development build of WebKit by clicking once inside the yellow rectangle to the left of the red triangle. As previous comments note the assertion failure is not triggered with a deployment build. well my bad, apologies for that one :). Fixed (by removing asserts) by David Harrison (as fix for <rdar://problem/4119646> <rdar://problem/ 4125949>). |