[Webkit-unassigned] [Bug 40162] Prevent Geolocation making callbacks to a ScriptExecutionContext that no longer exists

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Jun 9 08:51:53 PDT 2010


https://bugs.webkit.org/show_bug.cgi?id=40162





--- Comment #18 from Jeremy Orlow <jorlow at chromium.org>  2010-06-09 08:51:51 PST ---
(In reply to comment #17)
> Mostly, ease of understanding and refactoring. Not long ago, DOMTimer code was all in DOMWindow, and we had to factor it out to more easily fix bugs, and to support timers in workers. Geolocation design is pretty much identical, so solving the same problems differently isn't good.

I agree that solving the problem in two ways is not good, but it seems as though we do it all over the place in WebKit and historically we haven't put too much emphasis on cleaning up the worse ways of doing things.  When you refactor code, there is the risk of introducing new bugs and there is of course opportunity cost.  That said, I think a strategically placed FIXME (so the next time someone is writing code based on the old code they get a pointer towards the better way of doing something) is the very least you should do in such a case.

Anyway, I was talking to Steve and he was telling me that there is a lot of code in GeoLocation that's kind of a mess.  For example, there are two interfaces.  And it's pretty clear now that the second one is far superior and that the first should be merged into it.  He also seemed to think the code would be a bit cleaner if the code was factored as you suggested.  And I think I have him convinced the callbacks should handle suspend/resume more gracefully.  That on top of the fact that Steve's already waist deep in GeoLocation code, it seems like it might make sense for him to just finish the job.

-- 
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