On Thu, Dec 28, 2023 at 11:31:05AM +0100, Tomas Härdin wrote:
> tis 2023-12-26 klockan 23:30 +0100 skrev Michael Niedermayer:
> > Hi
> > 
> > On Tue, Dec 26, 2023 at 04:52:47PM +0100, Tomas Härdin wrote:
> > > 
> > 
> > [...]
> > > +
> > > +static int apc_write_header(AVFormatContext *s)
> > > +{
> > > +    AVIOContext *pb = s->pb;
> > > +    AVCodecParameters *par;
> > > +    AVStream *st;
> > > +
> > > +    if (s->nb_streams != 1) {
> > > +        av_log(s, AV_LOG_ERROR, "Must have exactly one stream\n");
> > > +        return AVERROR(EINVAL);
> > > +    }
> > > +
> > > +    st = s->streams[0];
> > > +    par = st->codecpar;
> > > +
> > > +    if (par->ch_layout.nb_channels <= 0 || par-
> > > >ch_layout.nb_channels > 2) {
> > > +        av_log(s, AV_LOG_ERROR, "Must be mono or stereo\n");
> > > +        return AVERROR(EINVAL);
> > > +    }
> > > +
> > > +    if (par->extradata_size != 0 && par->extradata_size != 8) {
> > > +        av_log(s, AV_LOG_ERROR,
> > > +            "Must have exactly 0 or 8 bytes of extradata, got
> > > %i\n",
> > > +            par->extradata_size);
> > > +        return AVERROR(EINVAL);
> > > +    }
> > > +
> > > +    ffio_wfourcc(pb, "CRYO");
> > > +    ffio_wfourcc(pb, "_APC");
> > > +    ffio_wfourcc(pb, "1.20");
> > 
> > > +    avio_wl32(pb, 0); // number or samples
> > 
> > please add to the comment "updated in apc_write_trailer()"
> 
> Sure
> 
> 
> > > +static int apc_write_packet(AVFormatContext *s, AVPacket *pkt)
> > > +{
> > > +    size_t extradata_size = 0;
> > 
> > > +    const uint8_t *extradata = av_packet_get_side_data(
> > > +            pkt, AV_PKT_DATA_NEW_EXTRADATA, &extradata_size);
> > > +
> > > +    if (extradata_size == 8) {
> > > +        // we got predictors from encoder
> > > +        // try to seek back end write them
> > > +        int64_t pos = avio_tell(s->pb), err;
> > > +        if ((err = avio_seek(s->pb, 20, SEEK_SET)) >= 0) {
> > > +            avio_write(s->pb, extradata, extradata_size);
> > > +            avio_seek(s->pb, pos, SEEK_SET);
> > > +        } else {
> > > +            av_log(s, AV_LOG_ERROR, "Got predictors from encoder
> > > but couldn't seek back to write them\n");
> > > +            // fail since we should always be able to do this
> > > within the avio cache
> > > +            // unless the encoder gave us predictors way too late
> > > for some reason
> > > +            return err;
> > > +        }
> > > +    }
> > 
> > I think the encoder should buffer data or use 2 passes
> > if it needs future data. seeking back in the muxer is a bit odd.
> > 
> > if it becomes available always in teh first packet then maybe the
> > whole header could be written in the first packet
> 
> Good idea. IIRC the issue was that extradata is either given
> immediately (remuxing) or after a single-packet delay (encoding), hence
> the check for *new* metadata
> 
> > I would suggest, you add APC support to nut.
> > Thats a good test to ensure the packet and extradata is set at the
> > right time
> > for a generic muxer
> 
> How would I go about doing that?

nut mainly needs to be able to map the ffmpeg API CODEC_ID to some external id
so the codec id needs to be in one of the table nut scans for it

thx

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

Its not that you shouldnt use gotos but rather that you should write
readable code and code with gotos often but not always is less readable