[Webkit-unassigned] [Bug 55198] LayoutTests/http/tests/media/video-play-suspend is now hanging on chromium

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=55198





--- Comment #2 from Steve Lacey <sjl at chromium.org>  2011-02-25 09:30:56 PST ---
(In reply to comment #1)
> (In reply to comment #0)
> > This just started hanging for chrome due to a cl of mine (http://codereview.chromium.org/6581012/) which causes the MediaPlayer states to transition from 'loading -> loaded ->idle' instead of 'loading -> idle'. This appears to be the correct spec thing to do as a progress event is now fired when the media has completed loading (which it wasn't before), but a side effect is that the 'suspend' event is now not being fired as the network doesn't actually suspend (indicated by a 'loading -> idle' transition).
> > 
> > I ran the test under safari and the test also hangs there too.
> > 
> > So it seems to me that the test is actually not testing for correct behaviour.
> > 
> > Thoughts?
> 
> 'suspend' is to be fired when the UA stops downloading for some reason:
> 
>     When a media element's download has been suspended, the user agent must set the 
>     networkState to NETWORK_IDLE and queue a task to fire a simple event named suspend 
>     at the element.
> 
> so HTMLMediaElement fires it when the media engine signals a network state change from 'loading' or 'loaded' to MediaPlayer::Idle.  
> 
> I think the problem is in HTMLMediaElement::setNetworkState, where MediaPlayer::Loaded results in m_networkState being set to NETWORK_IDLE without firing a 'suspended' event. I think it should fire a 'suspend' event just before the final 'progress' event, though we will have to see if it causes other problems.

I also see this in the spec for the suspend event:

      The user agent is intentionally not currently fetching media data, but does not have the entire
      media resource downloaded.

Which implies to me that you get a suspend event if download stops but there's more data to download.

The sentences before and after the paragraph you list also imply (to me) that if download is suspended but there's more to download, that's when you get a suspend event. There's no wordage that implies you get a suspend event in any other situation...

Confusion (in my mind at least :-) abounds.

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