[Webkit-unassigned] [Bug 37218] Implement FileStreamProxy that calls FileStream methods on FileThread for FileAPI

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Fri Apr 9 17:36:01 PDT 2010


--- Comment #19 from Kinuko Yasuda <kinuko at chromium.org>  2010-04-09 17:36:00 PST ---

(In reply to comment #17)
> (From update of attachment 53009 [details])
> > +static void derefProxyOnContext(ScriptExecutionContext*, FileStreamProxy* proxy)
> > +{
> > +    proxy->deref();
> Can we add an ASSERT before this line, like:
>        ASSERT(proxy->hasOneRef());
> This way, we can ensure that the caller, i.e., FileReader/FileWriter, has
> already
> released the reference and now it should only has ref count 1 now.

Asserting on ref count makes me a bit nervous, so I added another status flag
to allow recycling even after stop() (just in case).

> > diff --git a/WebCore/html/FileStreamProxy.h b/WebCore/html/FileStreamProxy.h
> > +// A proxy module that calls corresponding FileStream methods on the file thread.  Note: you must call stop() before destroying a FileStreamProxy instance, e.g. in the desctructor of an object that holds a reference to the proxy.
> Probably we could say the following for Note part:
>    // ... Note: you must call stop() first and then release the reference to
> the FileStreamProxy instance.


> > +    // Stop() stops any pending tasks, close the file and derefs itself; the caller should not recycle the instance after calling stop().
> It is not needed to repeat stop() in the comment. Also we should use "closes"
> to be consistent with "stops".
> Probably it is better to say "destruct".
>       // Stops the proxy and schedules it to be destructed. All the pending
> tasks will be aborted and the file stream will be closed.
>       // Note: the caller should not recycle the instance after calling stop().

Done.  Also removed the notion about recycling as now it doesn't apply.

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