[webkit-reviews] review requested: [Bug 118052] [WK2][GTK] SIGSEV in webkitWebViewBaseSizeAllocate : [Attachment 205799] Patch

[bugzilla-daemon at webkit.org]

Alberto Garcia <agarcia at igalia.com> has asked  for review:
Bug 118052: [WK2][GTK] SIGSEV in webkitWebViewBaseSizeAllocate
https://bugs.webkit.org/show_bug.cgi?id=118052

Attachment 205799: Patch
https://bugs.webkit.org/attachment.cgi?id=205799&action=review

------- Additional Comments from Alberto Garcia <agarcia at igalia.com>
I actually think that the problem is that
!drawingArea()->size().isEmpty() test. Why do we need to check that at
all?

If the point is to know if the web view must be resized when the
widget is mapped then I don't see the purpose of this.

About why the SIGSEGV is happening: the drawingArea can be null if the
web process crashes. A check was added some time ago (bug 62541) in
order to prevent that but it was later moved to
resizeWebKitWebViewBaseFromAllocation() (bug 77743).