398109 - toString() on XOW is very odd

Status: RESOLVED FIXED

(Core :: XPConnect, defect)

Description

[Boris Zbarsky [:bzbarsky]]

STEPS TO REPRODUCE:

1)  javascript:alert(window)
2)  javascript:alert(window + "")
3)  Write a mochitest in which you have a test like:

   is(null, window, "Should be null, for some reason")

EXPECTED RESULTS:  All three give the same string for the window

ACTUAL RESULTS:
1)  Gives "[object XPCCrossOriginWrapper [object Window]]"
2)  Gives "[object Window]"
3)  Gives "[]"

This last particularly bothers me, since I can't tell why that's happening.

Comment 1

[Blake Kaplan (:mrbkap) (inactive)]

#3 happens due to a bug in MochiKit's repr (specifically, its isArrayLike function returns true for window objects, which is incorrect). I'll file that in a second. I have a patch for cases 1 and 2.

Comment 2

[Blake Kaplan (:mrbkap) (inactive)]

Attachment: Fix

If only hg allowed me to give more context to the diff...

The only tricky bit here is that this code is shared by both XPCNativeWrapper whose toString I wanted to leave alone.

With this patch, to verify that you have a XOW from JS, you must do Object.prototype.toString.call(obj).

Comment 3

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 282965 [details] [diff] [review]
Fix

sr=bzbarsky

Comment 4

[Blake Kaplan (:mrbkap) (inactive)]

I filed bug 398118 on the MochiKit issue.

Comment 5

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 282965 [details] [diff] [review]
Fix

This fixes a XOW inconsistency and makes alert(window) and alert(window + '') do the same thing.

Comment 6

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Not a blocker maybe, but since there's a patch, it'd be great to get it in.

Comment 7

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 282965 [details] [diff] [review]
Fix

This will help ease confusion about XOWs and will make us consistent again.

Comment 8

[Mike Beltzner [:beltzner, not reading bugmail]]

Comment on attachment 282965 [details] [diff] [review]
Fix

a=endgame drivers for after M9 freeze

Comment 10

[Blake Kaplan (:mrbkap) (inactive)]

Fix checked in.