[webkit-reviews] review denied: [Bug 12718] REGRESSION(r18184): Segmentation fault when loading abc.go.com : [Attachment 13123] Patch v1 (no changelog; no test)

[bugzilla-request-daemon at macosforge.org]

Maciej Stachowiak <mjs at apple.com> has denied Maciej Stachowiak
<mjs at apple.com>'s request for review:
Bug 12718: REGRESSION(r18184): Segmentation fault when loading abc.go.com
http://bugs.webkit.org/show_bug.cgi?id=12718

Attachment 13123: Patch v1 (no changelog; no test)
http://bugs.webkit.org/attachment.cgi?id=13123&action=edit

------- Additional Comments from Maciej Stachowiak <mjs at apple.com>
It's good to know this solves the crash; on the other hand, using autorelease
to resolve lifetime issues is somewhat poor form, as it may only be masking the
immediate symptom. It's getting put into d->m_delegate, which is a smart
pointer, so it should stay alive as long as the ResourceHandle is alive. So
that must mean the ResourceHandle actually is getting destroyed too early. How
is it that it can get destroyed before connectionDidFinishLoading:?