[FFmpeg-devel] [PATCH] lavf/concatdec: allow to match streams by id.

[Michael Niedermayer]

On Wed, Feb 26, 2014 at 08:33:04PM +0100, Nicolas George wrote:
> That makes the concat demuxer usable with MPEG-PS streams,
> even when the streams in the different parts are detected
> in different order.

Do i understand correctly that in such case the user has to
specify the stream maping one by one ?

I think it would be better if (additionally) the code tried to match
streams automatically based on codec id, resolution, language, channel
layout
(possibly program name if its TS)


[...]
> +static int match_streams(AVFormatContext *avf)
> +{
> +    ConcatContext *cat = avf->priv_data;
> +    AVStream *st;
> +    int *map, i, j, ret;
> +
> +    if (!cat->match_streams ||
> +        cat->cur_file->stream_map_size >= cat->avf->nb_streams)
> +        return 0;
> +    map = av_realloc(cat->cur_file->stream_map,
> +                     cat->avf->nb_streams * sizeof(*map));
> +    if (!map)
> +        return AVERROR(ENOMEM);
> +
> +    for (i = cat->cur_file->stream_map_size; i < cat->avf->nb_streams; i++) {
> +        st = cat->avf->streams[i];
> +        map[i] = -1;
> +        for (j = 0; j < avf->nb_streams; j++) {
> +            if (avf->streams[j]->id == st->id) {
> +                av_log(avf, AV_LOG_VERBOSE,
> +                       "Match slave stream #%d with stream #%d id 0x%x\n",
> +                       i, j, st->id);
> +                map[i] = j;
> +                if (!avf->streams[j]->codec->codec_id && st->codec->codec_id)
> +                    if ((ret = copy_stream_props(avf->streams[j], st)) < 0)

> +                        return ret;

this leaves a stale pointer in cat->cur_file->stream_map

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

You can kill me, but you cannot change the truth.
-------------- 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/20140302/19e02ed2/attachment.asc>