[Webkit-unassigned] [Bug 108960] [GTK] Remove subresource leaks from WebKit1 and WebKit2

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=108960





--- Comment #24 from George McCollister <george.mccollister at gmail.com>  2013-02-13 10:59:33 PST ---
(In reply to comment #23)
> (In reply to comment #22)
> > (In reply to comment #21)
> > > Carlos brought up a good point that the old API returned borrowed references, so we'd need to preserve that behavior to avoid further leaks. Perhaps you can create the resources for get_resources and then deref them in a timeout?
> > 
> > Looks like using a timeout might be the only option for derefing the objects although creating a race condition seems like a bad idea. How long of a timeout do you think would be required to minimize the likely hood of derefing the resources while they are still in use?
> 
> That's a good question. I assumed that anyone wanting to use the resources "later" would have increased their reference count, but perhaps that isn't the case. Do you know if Midori uses them later?

No, looks like it just uses them within the signal handler so I think we could  just specify an interval of 1.

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