[Webkit-unassigned] [Bug 39879] Geolocation activity started after frame has been disconnected can cause crash

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Jun 9 02:09:08 PDT 2010


https://bugs.webkit.org/show_bug.cgi?id=39879





--- Comment #34 from Jeremy Orlow <jorlow at chromium.org>  2010-06-09 02:09:06 PST ---
(In reply to comment #33)
> +    else if (!m_frame)
> +        notifier->setFatalError(PositionError::create(PositionError::POSITION_UNAVAILABLE, serviceNoLongerAvailableErrorMessage));
> 
> Will the fatal error stay set if a document returns from b/f cache? I'm worried about the following scenario:
> 1. a request is attempted on a document in b/f cache;
> 2. user navigates back, so the document is in frame again;
> 3. Geolocation still doesn't work.
> 
> Note that even the fact that Geolocation prevents documents from going into b/f cache doesn't necessarily prevent this scenario. Step 1 occurs when a document is already in b/f cache. And of course, that limitation might be lifted one day.
> 
> I'm still not very familiar with Geolocation code, so I'm not sure what GeoNotifiers are. But even if there is no problem, it seems that adding a test would be useful. The two "fatal errors" are different in that "permission denied" stays forever, but "no frame" can be a temporary condition.

Is it possible for this to happen while it's in the bfcache?  If so (and now that you mention it, it does seem plausible) then I'd have to agree.  (And it might be a good reason to go with the design you mentioned in the other bug.)

On the other hand, my intuition is that properly handling suspend/resume events might just clear all of this up as well.  Steve, let's sit down in front of a whiteboard and think about this a bit.

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



More information about the webkit-unassigned mailing list