[Webkit-unassigned] [Bug 6751] For unnamed frames, window.name returns a generated name

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=6751


Daniel Bates <dbates at webkit.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #71377|0                           |1
        is obsolete|                            |
  Attachment #71666|                            |review?
               Flag|                            |




--- Comment #5 from Daniel Bates <dbates at webkit.org>  2010-10-23 21:00:54 PST ---
Created an attachment (id=71666)
 --> (https://bugs.webkit.org/attachment.cgi?id=71666&action=review)
Patch with updated test case

Updated change log.

For the the port-specific API changes in this patch, I chose to preserve the current behavior of returning the unique frame name. We may want to consider changing the API behavior so that it returns the actual name of the frame, which may be the empty string.

One concern I had was regressing the fix for bug 12732, which was landed in changeset 24396 <http://trac.webkit.org/changeset/24396>. The test case landed in this changeset <http://trac.webkit.org/browser/trunk/LayoutTests/fast/frames/iframe-set-inner-html.html> used frame name uniqueness for two unnamed frames to verify the fix. Clearly, this approach conflicts with this patch and the HTML5 spec with regards to frame names.

Further investigation is needed to be able to devise an alternative test for bug 12732. Alternatively, we could expose a DRT method that returns the internal unique name for a frame and modify the test case to use this method. Does anyone have any additional insight/suggestions into bug 12732 and its test case?

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.