[Webkit-unassigned] [Bug 38742] [chromium] Implement canEstablishDatabase call for workers.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=38742





--- Comment #11 from jochen at chromium.org  2010-05-07 12:25:05 PST ---
(In reply to comment #7)
> (From update of attachment 55367 [details])
> Looks reasonable, couple of notes:
> 
> WebKit/chromium/src/WebWorkerBase.cpp:157
>  +          return false;
> It is enough to have simple ASSERT(controller) here, since this method is not
> supposed to be called for non-worker contexts and there is no meaningful
> scenario when we actually want to return false from here.
> 
> WebKit/chromium/src/WebWorkerBase.h:150
>  +      class AllowDatabaseMainThreadBridge;
> This class is a detail of implementation and should be declared/defined in cpp
> file.
> 
> WebKit/chromium/src/WebWorkerBase.h:152
>  +      static void allowDatabaseTask(
> This can be just a static method in cpp file, it does not need to be in .h
> 
> WebKit/chromium/src/WebWorkerBase.cpp:198
>  +  void WebWorkerBase::AllowDatabaseMainThreadBridge::cancel()
> Where is this method used?

Hups, I've intended to invoke this when the runloop was terminated before the
call was completed. In that case, the didComplete call should be get scheduled
on the context thread.

> 
> WebKit/chromium/src/WebWorkerBase.cpp:206
>  +      MutexLocker lock(m_synchronousMutex);
> Why this mutex is necessary? It is usually not needed to protect bool values
> like that.

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