[FFmpeg-cvslog] r19018 - trunk/libavformat/mov.c

[Mike Melanson]

Reimar D?ffinger wrote:
> On Sun, May 31, 2009 at 01:49:29AM -0700, Baptiste Coudurier wrote:
>> Reimar D?ffinger wrote:
>>> On Sun, May 31, 2009 at 01:22:44AM -0700, Baptiste Coudurier wrote:
>>>> Reimar D?ffinger wrote:
>>>>> On Sun, May 31, 2009 at 06:36:26AM +0200, bcoudurier wrote:
>>>>>> Author: bcoudurier
>>>>>> Date: Sun May 31 06:36:26 2009
>>>>>> New Revision: 19018
>>>>>>
>>>>>> Log:
>>>>>> split audio chunks in mov demuxer
>>>>> Please verify that the changes in the FATE test are correct and
>>>>> expected, so Mike can update them.
>>>>> Changed tests are:
>>>>> http://fate.multimedia.cx/index.php?test_spec=120
>>>>> http://fate.multimedia.cx/index.php?test_spec=267
>>>>> http://fate.multimedia.cx/index.php?test_spec=268
>>>> That's expected output, right ? What's the link to actual output ?
>>> Just choose any fate config, e.g.
>>> http://fate.multimedia.cx/index.php?build_record=86174
>>> But actually I thought you just need to test if those commands still give
>>> reasonable output for you after this change (the new FATE results look
>>> reasonable to me, but you can't see e.g. audio corruption just from the
>>> packet checksums. The audio packets are a bit unevenly distributed
>>> though, sometimes there's only one, sometimes there are three in a row
>>> between video packets).
>> This depends on the interleaving of the file. New output is good for me.
>> You can see that smaller audio packets are demuxed.
> 
> Yes, I thought that it looked ok, but you should know better after all
> :-).
> Now only Mike has to find time to update the reference values.

Done.

Now, Baptiste: Any chance of addressing the PVA demuxer problem that 
arose in 18959?

-- 
     -Mike Melanson