RE: Implementation Report for mediafragments.js

Hi all,

I have done the test again for Synote Media Fragment Player. It's great that
most parsing problem has been fixed. There is only one problem about
UC0072-UA. I think the trailing dot "t=3." is not parsed correctly by the
parser. I can see a "Media Fragments URI Parsing Warning: Invalid time
dimension" in my firebug's log. In Thomas's test [1], TC0072-UA should
return "start:3". Is it a problem? Many thanks.

BTW, I totally agree that parser is only part of an implementation and the
player still need to do some lower level work, such as checking the duration
and framerate of the video, etc. That's why I am trying to generalise the
interface for the Media Fragment URI implementation from most of major media
players online. The problem for me is that not every one of them provide
same metadata about the video through javascript API and some metadata is
missing in most players (for example, the framerate in TC0021-UA).

[1] http://tomayac.com/mediafragments/implementationtests.html

Regards
Yunjia Li

PhD Student & Research Fellow
Web and Internet Science Group
School of Electronics and Computer Science 
The University of Southampton 
Southampton SO17 1BJ 
http://afterglowlee.blogspot.com


-----Original Message-----
From: Thomas Steiner [mailto:tomac@google.com] 
Sent: 17 November 2011 10:05
To: Davy Van Deursen
Cc: Media Fragment; Erik Mannens; yl2@ecs.soton.ac.uk
Subject: Re: Implementation Report for mediafragments.js

Hi Davy, all,

> First of all, I noticed a mismatch in the spec between [1] and [2]. In 
> [1], npt-mmss is defined, while in [2], it is not. The group decided 
> to make the hours optional (see also ACTION-191 [3]); hence the extension
to RFC2326.
> However, this should be corrected in [2]. Note that in both cases 
> (npt-hhmmss and npt-mmss), the trailing dot is allowed.
Thanks for the clarification. So we extend an RFC :-) Not sure this is a
good thing, but the group decided so in the past, so I will not argue. The
important point is that trailing dots are explicitly allowed, so the parser
should (and will) allow it.

> Knowing this, TC0080-UA and TC0081-UA should be marked as valid media 
> fragments according to the spec (which is at least how I interpret it :)).
Yes. As legal as it gets :-)

> If no one objects, I can perform the changes (in both the spec and the 
> test cases).
+1 for the proposed changes.

Best,
Tom

--
Thomas Steiner, Research Scientist, Google Inc.
http://blog.tomayac.com, http://twitter.com/tomayac

Received on Thursday, 17 November 2011 11:02:51 UTC