Closed Bug 597109 Opened 14 years ago Closed 14 years ago

JSProxyHandler should deal with PropertyStub

Categories

(Core :: JavaScript Engine, defect)

x86_64
Linux
defect
Not set
normal

Tracking

()

RESOLVED FIXED

People

(Reporter: mrbkap, Assigned: mrbkap)

References

Details

(Whiteboard: fixed-in-tracemonkey)

Attachments

(1 file)

Attached patch PatchSplinter Review
Currently, when JSProxyHandler calls into a C++ implementation of getPropertyDescriptor, it doesn't check whether desc.{getter,setter} is PropertyStub. For a C++ implementation of a proxy handler, it doesn't know whether the desc is going to be used as a parameter to JS_DefineProperty, so it has to use PropertyStub. Therefore, JSProxyHandler should deal with both null and PropertyStub as being equivalent.

This patch has r=gal.
Attachment #475924 - Flags: review+
http://hg.mozilla.org/tracemonkey/rev/89535221195d
Assignee: general → mrbkap
Whiteboard: fixed-in-tracemonkey
http://hg.mozilla.org/mozilla-central/rev/89535221195d
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
No testcase. This hides a difference between Waldo's PropDesc, which never returns js::PropertyOp function pointers (see PropDesc::initialize), and the mrbkap-via-luke PropertyDescriptor counterpart, which does. Yuck.

Need that test, and I think there are bugs in jsproxy.cpp still not handling both null and PropertyStub.

/be
Flags: in-testsuite?
Did I mention we have one too many *Prop*Desc* thingies?

/be
(In reply to comment #3)
> Need that test, and I think there are bugs in jsproxy.cpp still not handling
> both null and PropertyStub.

I'm wrong, the code is just (increasingly) confusing, not wrong.

Still want a test here.

/be
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: