[webkit-reviews] review denied: [Bug 23900] Crash if any frame in a frameset is named _top : [Attachment 28786] The right way to fix this

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Fri Mar 20 10:28:27 PDT 2009


Darin Adler <darin at apple.com> has denied Sverrir Á. Berg
<sverrir at chromium.org>'s request for review:
Bug 23900: Crash if any frame in a frameset is named _top
https://bugs.webkit.org/show_bug.cgi?id=23900

Attachment 28786: The right way to fix this
https://bugs.webkit.org/attachment.cgi?id=28786&action=review

------- Additional Comments from Darin Adler <darin at apple.com>
Thanks for tackling this.

I'm excited that you think we can make frames without names work properly. Are
you sure there are no new problems caused by removing the names? I believe we
use those frame names for back/forward housekeeping on Mac OS X at least. I'm
not sure it's safe to just throw that away.

What's your rationale for removing the disconnected frame feature? The fact
that _setIsDisconnected is in WebFramePrivate.h means that it’s SPI on Mac OS
X. So removing it would break Mac OS X applications that are using that API. So
you can’t just remove it. Did you remove it because you didn’t understand it or
because of some specific problem with it?

There’s no ChangeLog entry for WebCore. You need one.

The patch needs to include changes to test results for the tests that depend on
the old frame naming system, not just the new test.

There are tabs and a conflict marker in the LayoutTests/ChangeLog entry.


More information about the webkit-reviews mailing list