[FFmpeg-devel] [PATCH 2/5] avutil/mpegts_audio_desc_metadata: add helper function for AC3 descriptor 0x6a

lance.lmwang at gmail.com lance.lmwang at gmail.com
Thu Aug 6 05:20:38 EEST 2020


On Wed, Aug 05, 2020 at 05:50:56PM +0100, Kieran Kunhya wrote:
> On Wed, 5 Aug 2020 at 12:54, <lance.lmwang at gmail.com> wrote:
> 
> > On Wed, Aug 05, 2020 at 11:14:39AM +0100, Kieran Kunhya wrote:
> > > On Tue, 28 Jul 2020 at 16:31, <lance.lmwang at gmail.com> wrote:
> > >
> > > > From: Limin Wang <lance.lmwang at gmail.com>
> > > >
> > > > Signed-off-by: Limin Wang <lance.lmwang at gmail.com>
> > > > ---
> > > >
> > >
> > > Hi,
> > >
> > > There is already libdvbpsi or bitstream for these kind of applications,
> > it
> > > should not go in FFmpeg, nor in public API.
> >
> > The major goal of the patchset is to support the audio descriptor(AC3)
> > pass through
> > as some fields are needed by some IRDs.
> >
> > I have no clues how to use libdvbpsi or bitstream library to achieve that,
> > please
> > give comments.
> >
> 
> Why does it need to be public api?

Marton Balint give review comment to use the packet side data type like DOVI video 
descriptor when I try to use meta data of stream for the first patchset.

In fact, if it's not input copy instead of encode, we should create audio descriptor
for our AC-3 audio encoder also. If it's not public api, how to let encoder add it?

> The flaw with this method is that PSI is a separate data stream that
> doesn't fit within AVPacket or AVFrame etc, so it's a hack to just add it
> to AVPacket.

It's audio descriptor information although it's stored in PSI table, but it'll map to
specific audio packet in the end, so I don't think it's hacky. I have tested
with mpts with multiple audio stream, it's pass-by if copy specifc audio stream.



> 
> Kieran
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".

-- 
Thanks,
Limin Wang


More information about the ffmpeg-devel mailing list