Re: Implementation Report for mediafragments.js

Philip, Thomas, all,

I think we should discuss what we understand as 'an implementation'. To 
me, a parser is only part of a an implementation: you cannot check 
whether the user agent correctly displays the media fragment. In other 
words, we cannot fully test the Media Fragment Semantics [1] with a 
parser only. Any other opinions?

Note that we already have two implementations that make use of Thomas' 
parser: the NinSuna Media Fragments Player and the Synote Media Fragment 
Player. The fact that your parser passes all the nasty syntactical test 
cases is/will be directly reflected in the implementation reports of 
these players of course.

Best regards,

Davy

[1] 
http://www.w3.org/2008/WebVideo/Fragments/WD-media-fragments-spec/#media-fragment-semantics

On 16/11/2011 19:23, Thomas Steiner wrote:
> Hi Philip,
>
> Yepp, it tests for the correct parsing. This was mentioned in the
> asterisk remark (indirectly at least). This makes the chapter, track,
> id tests trivial of course.
>
> Cheers,
> Tom
>
> Thank God not sent from a BlackBerry, but from my iPhone
>
> On 16.11.2011, at 18:17, Philip Jägenstedt<philipj@opera.com>  wrote:
>
>> On Wed, 16 Nov 2011 17:59:52 +0100, Thomas Steiner<tomac@google.com>  wrote:
>>
>>> Dear Erik, Davy, Yunjia, all,
>>>
>>> I am happy to announce that mediafragments.js passes all test cases as
>>> specified in [1]* (with TC0026-UA in the corrected version). You can
>>> run the tests on your own [2], additional unit tests are available,
>>> too [3]. You can interactively play with the library [4], where the
>>> raw JSON data as well as a pretty-printed version of the parsed
>>> results can be seen. The source code is on GitHub [5]. This
>>> (hopefully) closes my action to deliver an implementation report to
>>> the Working Group.
>>>
>>> Yunjia, you might consider upgrading to the latest version.
>>>
>>> Please let me know if you have additional questions. I consider
>>> releasing a user script [6] based on the library that - at least for
>>> temporal and spatial fragments - should add support to all browsers
>>> that support user scripts.
>>>
>>> Best,
>>> Tom
>>>
>>> [1] http://www.w3.org/2008/WebVideo/Fragments/TC/ua-test-cases
>>> *) The library can of course not check for things that require
>>> low-level access to a media item (like resolution, length, chapters).
>>> [2] http://tomayac.com/mediafragments/implementationtests.html
>>> [3] http://tomayac.com/mediafragments/unittests.html
>>> [4] http://tomayac.com/mediafragments/mediafragments.html
>>> [5] https://github.com/tomayac/Media-Fragments-URI
>>> [6] http://greasemonkey.mozdev.org/authoring.html
>>>
>>
>> I'm not sure I understand what the pass criteria are here. TC0056-UA tests the track dimension and passes even when there's no DOM API that allows enabling/disabling tracks. (AFAIK no browser has shipped support for the AudioTrack and VideoTrack APIs yet.) It seems to me that it's only testing the parsing of mediafragments.js, is that correct?
>>
>> --
>> Philip Jägenstedt
>> Core Developer
>> Opera Software
>


-- 
Davy Van Deursen

Ghent University - IBBT
Department of Electronics and Information Systems - Multimedia Lab
URL: http://multimedialab.elis.ugent.be/dvdeurse

Received on Thursday, 17 November 2011 08:46:47 UTC