Closed Bug 400349 Opened 16 years ago Closed 16 years ago

Touching node in removed iframe triggers "ASSERTION: XPConnect is being called on a scope without a 'Components' property!"

Categories

(Core :: XPConnect, defect)

defect
Not set
major

Tracking

()

RESOLVED WORKSFORME

People

(Reporter: jruderman, Assigned: mrbkap)

References

Details

(Keywords: assertion, testcase)

Attachments

(1 file, 1 obsolete file)

Attached file testcase
This is a re-filing of bug 323978 without the irrelevant and confidential parts.

Loading the testcase triggers:

###!!! ASSERTION: XPConnect is being called on a scope without a 'Components' property!: 'Error', file /Users/jruderman/trunk/mozilla/js/src/xpconnect/src/xpcwrappednativescope.cpp, line 659

How bad is this?
Flags: blocking1.9?
Attached patch mrbkap's "incomplete patch" (obsolete) — Splinter Review
When mrbkap attached this patch to bug 323978, he said "This fixes a reduced testcase that I have, but not the problem in general, as I see crashes...".

Some crash bugs (including bug 321299) were fixed since then.  Does this patch fix all the remaining issues?
Comment on attachment 285441 [details] [diff] [review]
mrbkap's "incomplete patch"

A completed version of this patch was checked in for bug 321299.
Attachment #285441 - Attachment is obsolete: true
(In reply to comment #1)
> Some crash bugs (including bug 321299) were fixed since then.  Does this patch
> fix all the remaining issues?

The remaining issue here is that we call JS_ClearScope on windows when the window is closed (or the iframe is removed, in this case). While this is an "inconsistent" state, it's not a security problem and the effects are mainly the assertion, and (as indicated by the assertion) things like Components and other properties won't be available.
Flags: blocking1.9? → blocking1.9-
WFM.

Blake, is the issue you mentioned in comment 4 still around?  If it is, can you spin it off into a new bug?
Status: NEW → RESOLVED
Closed: 16 years ago
Resolution: --- → WORKSFORME
Flags: in-testsuite+
(In reply to comment #5)
> Blake, is the issue you mentioned in comment 4 still around?  If it is, can you
> spin it off into a new bug?

It's been filed as bug 470510.
You need to log in before you can comment on or make changes to this bug.