[FFmpeg-devel] [PATCH 2/2] avformat: add mapping for AC-4 in MP4

[Paul B Mahol]

On 2/17/19, James Almer <jamrial at gmail.com> wrote:
> On 2/17/2019 6:14 PM, Paul B Mahol wrote:
>> On 2/17/19, Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:
>>> 2019-02-17 21:50 GMT+01:00, Paul B Mahol <onemda at gmail.com>:
>>>> On 2/17/19, Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:
>>>>> 2019-02-17 21:25 GMT+01:00, Paul B Mahol <onemda at gmail.com>:
>>>>>> Signed-off-by: Paul B Mahol <onemda at gmail.com>
>>>>>> ---
>>>>>>  libavformat/isom.c | 1 +
>>>>>>  1 file changed, 1 insertion(+)
>>>>>>
>>>>>> diff --git a/libavformat/isom.c b/libavformat/isom.c
>>>>>> index 0a4d901be5..0052db539e 100644
>>>>>> --- a/libavformat/isom.c
>>>>>> +++ b/libavformat/isom.c
>>>>>
>>>>>> +    { AV_CODEC_ID_AC4,             MKTAG('a', 'c', '-', '4') },
>>>>>> /* ETSI TS 103 190 */
>>>>>
>>>>> Are there no codec-specific atoms needed when remuxing ac-4?
>>>>
>>>> They are completely optional and not needed for decoding or remuxing.
>>>
>>> Thank you for the explanation!
>>> Do we already have samples?
>>>
>>
>> http://testassets.dashif.org/#feature/details/586fb3879ae9045678eab593
>
> There's a dac4 atom inside the Audio Sample Entry, much like AC-3 has a
> dac3 and EAC-3 a dec3 atom.
> Are you sure it's optional? Codec specific atoms inside the stream's
> Video/Audio Sample Entry tend to be obligatory.

>From specification:
The ac4_dsi shall not be used to configure the AC-4 decoder. The AC-4
decoder shall obtain its configuration only
from ac4_toc , which is part of every sample.