[FFmpeg-devel] [PATCH] RealAudio SIPR @16k decoder (4/4) demuxer

Michael Niedermayer michaelni
Thu Oct 16 18:19:35 CEST 2008


On Wed, Sep 24, 2008 at 10:04:46PM +0200, Benjamin Larsson wrote:
> I'll commit the OK'ed parts tomorrow.
> 
> > 
> > 
> >> @@ -146,6 +157,14 @@
> >>                  ((uint8_t*)st->codec->extradata)[i] = get_byte(pb);
> >>              rm->audio_framesize = st->codec->block_align;
> >>              st->codec->block_align = rm->sub_packet_size;
> >> +            if (st->codec->codec_id == CODEC_ID_SIPR) {
> >> +                if ((flavor>=0) || (flavor<4))
> >> +                    st->codec->block_align = sipr_subpk_size[flavor];
> >> +                else {
> >> +                    av_log(s, AV_LOG_ERROR, "flavor = %d not in [0-3] range, invalid value!\n", flavor);
> >> +                    return -1;
> >> +                }
> >> +            }
> > 
> > is there really no field at the demxuer layer that is the packet size?
> 
> I don't know. I have no idea how .rm works.

doesnt sound convincing as argument to accept the hunk, which doesnt
mean its wrong just that someone has to check and confirm that there
is no cleaner way ...



> The only thing I do know is
> that they stuff several codec frames into 1 packet. I think most of the
> block_align manipulation is because of that. I think Roberto has to
> answer this question if you want a better answer.
> 
> > this also applies a little to the other codecs which also each do
> > different things with block_align ...
> > 

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

There will always be a question for which you do not know the correct awnser.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20081016/4bc38544/attachment.pgp>



More information about the ffmpeg-devel mailing list