[FFmpeg-devel] [PATCH 1/1] avformat/demux: Add durationprobesize AVOption

[Stefano Sabatini]

On date Tuesday 2024-02-06 11:52:09 +0100, Nicolas Gaullier wrote:
> Yet another probesize used to get the last pts (and thus the duration)
> of mpegts/ps files. It is aimed at users interested in better durations probing
> for itself, or because using avformat_find_stream_info indirectly and requiring
> exact values: for concatdec for exemple, especially if streamcopying above it.

nit: exemple typo

> The current code does not behave well with high bitrates and high video buffering
> (physical gap between the last video packet and the last audio packet).
> 
> Default behaviour is unchanged: 250k up to 250k << 6 (step by step)
> Setting this new option has two effects:
> - override the maximum probesize (currently 250k << 6)
> - reduce the number of steps to 1 instead of 6, this is to avoid detecting
> the audio "too early" and failing to reach a video packet. Here, even if audio
> duration is found but not the video duration, there will be a retry, so at the
> end the full user-overriden probesize will be used.
> 
> Signed-off-by: Nicolas Gaullier <nicolas.gaullier at cji.paris>
> ---
>  libavformat/avformat.h      |  8 ++++++++
>  libavformat/demux.c         | 13 ++++++++-----
>  libavformat/options_table.h |  1 +
>  3 files changed, 17 insertions(+), 5 deletions(-)

please add missing user doc in doc/formats.texi (you might reuse the
hightlight above to explain why this is needed), also the new field
requires an entry in doc/APIChanges

> diff --git a/libavformat/avformat.h b/libavformat/avformat.h
> index 5d0fe82250..533f5a963d 100644
> --- a/libavformat/avformat.h
> +++ b/libavformat/avformat.h
> @@ -1823,6 +1823,14 @@ typedef struct AVFormatContext {
>       * Freed by libavformat in avformat_free_context().
>       */
>      AVStreamGroup **stream_groups;
> +
> +    /**
> +     * Maximum number of bytes read at the end of input in order to determine the

> +     * stream durations. Used by avformat_find_stream_info() for MPEG-TS/PS.

let's clarify this: is there any reason why this should not be used
with other formats? If this the case, probably a private option would
be best. If not, probably we should amend the doxy as it suggests it
is only useful with MPEG TS/PS.

> +     *
> +     * Demuxing only, set by the caller before avformat_open_input().
> +     */
> +    int64_t duration_probesize;
>  } AVFormatContext;
>  
>  /**
> diff --git a/libavformat/demux.c b/libavformat/demux.c
> index 6f640b92b1..798b44c979 100644
> --- a/libavformat/demux.c
> +++ b/libavformat/demux.c
> @@ -1740,8 +1740,9 @@ static void estimate_timings_from_bit_rate(AVFormatContext *ic)
>                 "Estimating duration from bitrate, this may be inaccurate\n");
>  }
>  
> -#define DURATION_MAX_READ_SIZE 250000LL
> -#define DURATION_MAX_RETRY 6
> +#define DURATION_MAX_READ_SIZE_DEFAULT 250000LL
> +#define DURATION_MAX_RETRY_DEFAULT 6
> +#define DURATION_MAX_RETRY_USER 1
>  
>  /* only usable for MPEG-PS streams */
>  static void estimate_timings_from_pts(AVFormatContext *ic, int64_t old_offset)
> @@ -1749,6 +1750,8 @@ static void estimate_timings_from_pts(AVFormatContext *ic, int64_t old_offset)
>      FFFormatContext *const si = ffformatcontext(ic);
>      AVPacket *const pkt = si->pkt;
>      int num, den, read_size, ret;
> +    int64_t duration_max_read_size = ic->duration_probesize ? ic->duration_probesize >> DURATION_MAX_RETRY_USER : DURATION_MAX_READ_SIZE_DEFAULT;
> +    int duration_max_retry = ic->duration_probesize ? DURATION_MAX_RETRY_USER : DURATION_MAX_RETRY_DEFAULT;
>      int found_duration = 0;
>      int is_end;
>      int64_t filesize, offset, duration;
> @@ -1784,7 +1787,7 @@ static void estimate_timings_from_pts(AVFormatContext *ic, int64_t old_offset)
>      filesize = ic->pb ? avio_size(ic->pb) : 0;
>      do {
>          is_end = found_duration;
> -        offset = filesize - (DURATION_MAX_READ_SIZE << retry);
> +        offset = filesize - (duration_max_read_size << retry);
>          if (offset < 0)
>              offset = 0;
>  
> @@ -1793,7 +1796,7 @@ static void estimate_timings_from_pts(AVFormatContext *ic, int64_t old_offset)
>          for (;;) {
>              AVStream *st;
>              FFStream *sti;
> -            if (read_size >= DURATION_MAX_READ_SIZE << (FFMAX(retry - 1, 0)))
> +            if (read_size >= duration_max_read_size << (FFMAX(retry - 1, 0)))
>                  break;
>  
>              do {
> @@ -1847,7 +1850,7 @@ static void estimate_timings_from_pts(AVFormatContext *ic, int64_t old_offset)
>          }
>      } while (!is_end &&
>               offset &&
> -             ++retry <= DURATION_MAX_RETRY);
> +             ++retry <= duration_max_retry);
>  
>      av_opt_set_int(ic, "skip_changes", 0, AV_OPT_SEARCH_CHILDREN);
>  
> diff --git a/libavformat/options_table.h b/libavformat/options_table.h
> index 91708de453..c2bdb484a7 100644
> --- a/libavformat/options_table.h
> +++ b/libavformat/options_table.h
> @@ -108,6 +108,7 @@ static const AVOption avformat_options[] = {
>  {"max_streams", "maximum number of streams", OFFSET(max_streams), AV_OPT_TYPE_INT, { .i64 = 1000 }, 0, INT_MAX, D },
>  {"skip_estimate_duration_from_pts", "skip duration calculation in estimate_timings_from_pts", OFFSET(skip_estimate_duration_from_pts), AV_OPT_TYPE_BOOL, {.i64 = 0}, 0, 1, D},
>  {"max_probe_packets", "Maximum number of packets to probe a codec", OFFSET(max_probe_packets), AV_OPT_TYPE_INT, { .i64 = 2500 }, 0, INT_MAX, D },

> +{"durationprobesize", "maximum number of bytes to probe the stream durations", OFFSET(duration_probesize), AV_OPT_TYPE_INT64, {.i64 = 0 }, 0, INT64_MAX, D},

duration_probesize? ... to probe the stream duration (why the plural?)