[Webkit-unassigned] [Bug 21837] critical error in Gtk FrameLoaderClient::frameLoaderDestroyed()

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=21837





------- Comment #13 from zecke at selfish.org  2008-12-05 05:33 PDT -------
(In reply to comment #11)
> (In reply to comment #9)

> 
> I agree that WebKitWebFrame should determine WebCore::Frame's lifetime, and
> FrameLoaderClient should determine WebKitWebFrame's lifetime? Should we follow
> Mac in this regard?

WebCore::Frame and FrameLoaderClient have the same lifetime. So with the above
we end up with a circle... The question is what to do. And I start to believe
the best way to deal with it, is to not have a circle.

We would end up with:
  WebKitWebView owns the Page. The Page ref's the WebKitWebFrame..
  WebKitWebFrame has a ref on the WebCore::Frame
  WebKitWebFrame has a pointer to the WebKitWebView and might be zero.
  WebKitWebPage has a pointer to the main frame, it will never be zero.

Truth:
  WebKitWebView has a WebCore::Frame... as the WebCore::Page is/should keep a
ref on the main frame. This means the WebKitWebFrame mainFrame is always
present.


This means:
  We will have to add null checks in WebKitWebFrame to see if the page/view is
gone. Change the references accordingly...



Do you agree on this? If so we can go and start to look at the implementation..


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