On 1/16/14 1:17 PM, "David Henry" <[email protected]> wrote:
> >On 1/16/14 12:28 PM, "Carl Eugen Hoyos" <[email protected]> wrote: > >>David Henry <dhenry@...> writes: >> >>> { 0x8a, AVMEDIA_TYPE_AUDIO, CODEC_ID_DTS }, >>> + { 0x86, AVMEDIA_TYPE_DATA, CODEC_ID_SCTE }, >> >>This looks very outdated, are you sure that you are >>developing with current git head? >>Everything else makes no sense. >> >>I may misunderstand your questions but afaict, you >>have to add a codec id, add some logic to the >>demuxer (as above) and add a decoder. The decoder >>will then receive packets from the demuxer. >> >>Carl Eugen > >Sorry, I forgot to mention that I'm using the 0.9 branch. Long story. >When you say add a decoder, do you mean in libavformat? The demuxer I >Am using is mpegts.c, so it will call the decoder I define through the >AVInputFormat interface? Using read_header() and read_packet()? Strike that last question - It seems you mean add a decoder in libavcodec, With init(), decode(), and close() functions defined in the AVCodec interface. What I don't see is where the muxer (mpegts in this case) calls the codec.decode() function. How/where does that take place? _______________________________________________ Libav-user mailing list [email protected] http://ffmpeg.org/mailman/listinfo/libav-user
