[Webkit-unassigned] [Bug 150888] [GStreamer] Do not use GThreadSafeMainLoopSource to send notifications to the main thread in MediaPlayerPrivateGStreamer

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=150888

--- Comment #5 from Zan Dobersek <zan at falconsigh.net> ---
(In reply to comment #4)
> (In reply to comment #3)
> > > Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:615
> > > +    if (isMainThread())
> > > +        player->notifyPlayerOfVideo();
> > > +    else
> > > +        player->m_notifier.notify(MainThreadNotification::VideoChanged, [player] { player->notifyPlayerOfVideo(); });
> > 
> > This isMainThread() check and the conditional immediate execution could be
> > rolled up into the notifier class. Would save some lines.
> > 
> > MainThreadNotifier::notify() would accept a templated functor as the second
> > parameter, invoke it immediately if on the main thread, or wrap it in
> > std::function<> and dispatch it onto the main thread via RunLoop otherwise.
> 
> I thought about that, but decided to do it in callers to avoid the function
> creation just to invoke a member method. But I'm not sure if creating the
> function would have any real impact in the end, what do you think?

You wouldn't create a std::function<> object until you know you're not on the main thread and have to use the RunLoop dispatch instead.

    template<typename F>
    void notify(T notificationType, const F& callbackFunctor)
    {
        if (!addPendingNotitifation(notificationType))
            return;

        if (!isMainThread()) {
            auto weakThis = m_weakPtrFactory.createWeakPtr();
            std::function<void ()> callback(callbackFunctor);
            RunLoop::main().dispatch([weakThis, notificationType, callback] {
                if (weakThis && weakThis->removePendingNotitifation(notificationType))
                    callback();
            });
        } else
            callbackFunctor();
    }

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20151106/98901974/attachment-0001.html>