[FFmpeg-devel] [PATCH] lavf/adtsenc: Add support for APE tags

[Michael Niedermayer]

On Wed, Oct 23, 2013 at 03:13:23AM -0300, James Almer wrote:
> On 12/07/13 8:27 AM, Paul B Mahol wrote:
> > On 7/12/13, Thierry Foucu <tfoucu at gmail.com> wrote:
> >> On Thu, Jul 11, 2013 at 1:16 PM, Paul B Mahol <onemda at gmail.com> wrote:
> >>
> >>> On 7/11/13, Thierry Foucu <tfoucu at gmail.com> wrote:
> >>>> On Thu, Jul 11, 2013 at 11:08 AM, James Almer <jamrial at gmail.com>
> >>>> wrote:
> >>>>
> >>>>> Should fix ticket #2269
> >>>>>
> >>>>> Signed-off-by: James Almer <jamrial at gmail.com>
> >>>>> ---
> >>>>>  libavformat/adtsenc.c | 9 +++++++++
> >>>>>  1 file changed, 9 insertions(+)
> >>>>>
> >>>>> diff --git a/libavformat/adtsenc.c b/libavformat/adtsenc.c
> >>>>> index 60d7b07..7118105 100644
> >>>>> --- a/libavformat/adtsenc.c
> >>>>> +++ b/libavformat/adtsenc.c
> >>>>> @@ -25,6 +25,7 @@
> >>>>>  #include "libavcodec/avcodec.h"
> >>>>>  #include "libavcodec/mpeg4audio.h"
> >>>>>  #include "avformat.h"
> >>>>> +#include "apetag.h"
> >>>>>
> >>>>>  #define ADTS_HEADER_SIZE 7
> >>>>>
> >>>>> @@ -162,6 +163,13 @@ static int adts_write_packet(AVFormatContext *s,
> >>>>> AVPacket *pkt)
> >>>>>      return 0;
> >>>>>  }
> >>>>>
> >>>>> +static int adts_write_trailer(AVFormatContext *s)
> >>>>> +{
> >>>>> +    ff_ape_write_tag(s);
> >>>>>
> >>>>
> >>>> I think this should be an option of the adts muxer as not all the adts
> >>>> decoder will know what to do with this and some will crash.
> >>>> and if some people use libav as a library and want to create adts chunk
> >>>> (HLS) they may be calling trailer at the end of each chunk and this
> >>>> will
> >>> be
> >>>> adding the ape metadata.
> >>>>
> >>>> What do you all think?
> >>>
> >>> You sure this code is used for HLS?
> >>>
> >>
> >> in HLS, audio only could be ADTS audio. And apple is using the size of the
> >> chunk and the duration of it to compute its bitrate. So, we should not have
> >> extra metadata for each chunk.
> > 
> > Yes, but: is this code used by HLS? and metadata is not set after each
> > chunk, just at
> > end.
> 
> Ping.
> 
> What version is preferred in the end? Forced, or enabled with an AVOption? 

either is fine with me, what is not fine is none thus


> If the latter, I'd prefer the AVOption to be "write_apetag" instead of "apetag" 

applied that then


> as i originally named it in alternative patch i submitted. That way it will be 
> consistent with similar AVOptions from other muxers.


Thanks
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

DNS cache poisoning attacks, popular search engine, Google internet authority
dont be evil, please
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20131023/f03bb5ba/attachment.asc>