Bug 11884

Summary: FunctionProtoFunc::callAsFunction uses the wrong global object as 'this' in the apply case
Product: WebKit Reporter: Geoffrey Garen <ggaren>
Component: JavaScriptCoreAssignee: Nobody <webkit-unassigned>
Status: RESOLVED DUPLICATE    
Severity: Normal CC: ap, barraclough, ddkilzer, jwalden+bwo, oliver, sam
Priority: P2 Keywords: InRadar
Version: 420+   
Hardware: Mac   
OS: OS X 10.4   
Attachments:
Description Flags
reduction none

Geoffrey Garen
Reported 2006-12-19 16:05:19 PST
See attached test case.
Attachments
reduction (280 bytes, text/html)
2006-12-19 16:06 PST, Geoffrey Garen 		no flags
Details
View All Add attachment proposed patch, testcase, etc.
Geoffrey Garen
Comment 1 2006-12-19 16:06:12 PST
Created attachment 11925 [details] reduction
Geoffrey Garen
Comment 2 2006-12-19 16:06:46 PST
We should probably analyze what we use for the global object in other cases, as well.
Geoffrey Garen
Comment 3 2006-12-19 16:09:25 PST
You have to open the reduction locally, to avoid a JavaScript security exception. See bug 11885.
Alexey Proskuryakov
Comment 4 2006-12-19 21:22:23 PST
Is this the same as bug 7831?
Geoffrey Garen
Comment 5 2006-12-20 00:48:47 PST
I think bug 7831 is specifically about eval.apply, and not about apply in general, or multiple global objects. It's possible that fixing one may fix the other, though.
David Kilzer (:ddkilzer)
Comment 6 2008-02-20 06:33:31 PST
<rdar://problem/5754363>
Oliver Hunt
Comment 7 2008-06-28 03:47:29 PDT
Sam, this page succeeds in ffx3, but fails with a security check failure in ToT
Sam Weinig
Comment 8 2008-06-28 09:01:56 PDT
We have different rules than Firefox regarding data: urls (we are much more strict). It doesn't seem like that is what this test is really testing though. To do the real test, just put the content from the data: url into its own file.
Gavin Barraclough
Comment 9 2011-12-25 15:49:26 PST
*** This bug has been marked as a duplicate of bug 64250 ***
Note You need to log in before you can comment on or make changes to this bug.