[FFmpeg-devel] [PATCH] ffprobe: add support to audio frame information printing

[Clément Bœsch]

On Tue, Jan 10, 2012 at 10:50:13PM +0100, Stefano Sabatini wrote:
> ---
>  ffprobe.c |   75 +++++++++++++++++++++++++++++++++++++++++++-----------------
>  1 files changed, 54 insertions(+), 21 deletions(-)
> 
> diff --git a/ffprobe.c b/ffprobe.c
> index be6ed19..fcb8c4c 100644
> --- a/ffprobe.c
> +++ b/ffprobe.c
> @@ -1290,7 +1290,20 @@ static void show_frame(WriterContext *w, AVFrame *frame, AVStream *stream)
>      const char *s;
>  
>      print_section_header("frame");
> -    print_str("media_type",             "video");
> +
> +    s = av_get_media_type_string(stream->codec->codec_type);
> +    if (s) print_str    ("media_type", s);
> +    else   print_str_opt("media_type", "unknown");
> +    print_int("key_frame",              frame->key_frame);
> +    print_ts  ("pkt_pts",               frame->pkt_pts);
> +    print_time("pkt_pts_time",          frame->pkt_pts, &stream->time_base);
> +    print_ts  ("pkt_dts",               frame->pkt_dts);
> +    print_time("pkt_dts_time",          frame->pkt_dts, &stream->time_base);
> +    if (frame->pkt_pos != -1) print_fmt    ("pkt_pos", "%"PRId64, frame->pkt_pos);
> +    else                      print_str_opt("pkt_pos", "N/A");
> +
> +    switch (stream->codec->codec_type) {
> +    case AVMEDIA_TYPE_VIDEO:
>      print_int("width",                  frame->width);
>      print_int("height",                 frame->height);
>      s = av_get_pix_fmt_name(frame->format);
> @@ -1310,45 +1323,65 @@ static void show_frame(WriterContext *w, AVFrame *frame, AVStream *stream)
>      print_int("top_field_first",        frame->top_field_first);
>      print_int("repeat_pict",            frame->repeat_pict);
>      print_int("reference",              frame->reference);
> -    print_int("key_frame",              frame->key_frame);
> -    print_ts  ("pkt_pts",               frame->pkt_pts);
> -    print_time("pkt_pts_time",          frame->pkt_pts, &stream->time_base);
> -    print_ts  ("pkt_dts",               frame->pkt_dts);
> -    print_time("pkt_dts_time",          frame->pkt_dts, &stream->time_base);
> -    if (frame->pkt_pos != -1) print_fmt    ("pkt_pos", "%"PRId64, frame->pkt_pos);
> -    else                      print_str_opt("pkt_pos", "N/A");
> +    break;
> +
> +    case AVMEDIA_TYPE_AUDIO:
> +        s = av_get_sample_fmt_name(frame->format);
> +        if (s) print_str    ("sample_fmt", s);
> +        else   print_str_opt("sample_fmt", "unknown");
> +        print_int("nb_samples",         frame->nb_samples);
> +        break;
> +    }
> +
>      print_section_footer("frame");
>  
>      av_free(pbuf.s);
>      fflush(stdout);
>  }
>  
> -static av_always_inline int get_video_frame(AVFormatContext *fmt_ctx,
> -                                            AVFrame *frame, AVPacket *pkt)
> +static av_always_inline int get_decoded_frame(AVFormatContext *fmt_ctx,
> +                                              AVFrame *frame, int *got_frame,
> +                                              AVPacket *pkt)
>  {
>      AVCodecContext *dec_ctx = fmt_ctx->streams[pkt->stream_index]->codec;
> -    int got_picture = 0;
> +    int ret = 0;
> +
> +    *got_frame = 0;
> +    if (dec_ctx->codec_id   != CODEC_ID_NONE) {

since you changed the line, maybe you could fix the weird spacing

> +        switch (dec_ctx->codec_type) {
> +        case AVMEDIA_TYPE_VIDEO:
> +            ret = avcodec_decode_video2(dec_ctx, frame, got_frame, pkt);
> +            break;
>  
> -    if (dec_ctx->codec_id   != CODEC_ID_NONE &&
> -        dec_ctx->codec_type == AVMEDIA_TYPE_VIDEO)
> -        avcodec_decode_video2(dec_ctx, frame, &got_picture, pkt);
> -    return got_picture;
> +        case AVMEDIA_TYPE_AUDIO:
> +            ret = avcodec_decode_audio4(dec_ctx, frame, got_frame, pkt);
> +            break;
> +        }
> +    }
> +    return ret;
>  }
>  
>  static void show_packets(WriterContext *w, AVFormatContext *fmt_ctx)
>  {
> -    AVPacket pkt;
> +    AVPacket pkt, pkt1;
>      AVFrame frame;
> -    int i = 0;
> +    int i = 0, ret, got_frame;
>  
>      av_init_packet(&pkt);
>  
>      while (!av_read_frame(fmt_ctx, &pkt)) {
>          if (do_show_packets)
>              show_packet(w, fmt_ctx, &pkt, i++);
> -        if (do_show_frames &&
> -            get_video_frame(fmt_ctx, &frame, &pkt)) {
> -            show_frame(w, &frame, fmt_ctx->streams[pkt.stream_index]);
> +        if (do_show_frames) {
> +            pkt1 = pkt;
> +            while (1) {
> +                ret = get_decoded_frame(fmt_ctx, &frame, &got_frame, &pkt1);
> +                if (ret < 0 || !got_frame)
> +                    break;

I'm not sure, but I think the API needs you to deal with the special case
of ret < 0. See 220481e1d6582277ae4c02b05f87251a8670e6cc for instance. I
can look for a sample to confirm this if you need one.

> +                show_frame(w, &frame, fmt_ctx->streams[pkt.stream_index]);
> +                pkt1.data += ret;
> +                pkt1.size -= ret;
> +            }

IIRC the documentation also mentions only the video frames; maybe you want
to add audio frames there too.

[...]

-- 
Clément B.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 490 bytes
Desc: not available
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20120111/66fe4ce1/attachment.asc>