[FFmpeg-devel] [PATCH v4] avformat/hls: Implement support for using AVSEEK_FLAG_BACKWARD when seeking
Gustav Grusell
gustav.grusell at gmail.com
Wed Feb 2 18:31:20 EET 2022
Any comments on this?
cheers,
Gustav
On Fri, Jan 28, 2022 at 9:24 PM Gustav Grusell <gustav.grusell at gmail.com>
wrote:
> Before, seeking in hls streams would always seek to the next keyframe
> after the given timestamp. With this fix, if seeking in videostream and
> AVSEEK_FLAG_BACKWARD is set, seeking will be to the first keyframe of
> the segment containing the given timestamp. This fixes #7485.
>
> Signed-off-by: Gustav Grusell <gustav.grusell at gmail.com>
> ---
> libavformat/hls.c | 24 +++++++++++++++++-------
> 1 file changed, 17 insertions(+), 7 deletions(-)
>
> diff --git a/libavformat/hls.c b/libavformat/hls.c
> index 4568e72cb2..44afdaab42 100644
> --- a/libavformat/hls.c
> +++ b/libavformat/hls.c
> @@ -1653,7 +1653,8 @@ static void
> add_metadata_from_renditions(AVFormatContext *s, struct playlist *pl
> /* if timestamp was in valid range: returns 1 and sets seq_no
> * if not: returns 0 and sets seq_no to closest segment */
> static int find_timestamp_in_playlist(HLSContext *c, struct playlist *pls,
> - int64_t timestamp, int64_t *seq_no)
> + int64_t timestamp, int64_t *seq_no,
> + int64_t *seg_start_ts)
> {
> int i;
> int64_t pos = c->first_timestamp == AV_NOPTS_VALUE ?
> @@ -1668,6 +1669,9 @@ static int find_timestamp_in_playlist(HLSContext *c,
> struct playlist *pls,
> int64_t diff = pos + pls->segments[i]->duration - timestamp;
> if (diff > 0) {
> *seq_no = pls->start_seq_no + i;
> + if (seg_start_ts) {
> + *seg_start_ts = pos;
> + }
> return 1;
> }
> pos += pls->segments[i]->duration;
> @@ -1691,7 +1695,7 @@ static int64_t select_cur_seq_no(HLSContext *c,
> struct playlist *pls)
> * playlist) and this is a complete file, find the matching segment
> * by counting durations. */
> if (pls->finished && c->cur_timestamp != AV_NOPTS_VALUE) {
> - find_timestamp_in_playlist(c, pls, c->cur_timestamp, &seq_no);
> + find_timestamp_in_playlist(c, pls, c->cur_timestamp, &seq_no,
> NULL);
> return seq_no;
> }
>
> @@ -2362,7 +2366,7 @@ static int hls_read_seek(AVFormatContext *s, int
> stream_index,
> int i, j;
> int stream_subdemuxer_index;
> int64_t first_timestamp, seek_timestamp, duration;
> - int64_t seq_no;
> + int64_t seq_no, seg_start_ts;
>
> if ((flags & AVSEEK_FLAG_BYTE) || (c->ctx->ctx_flags &
> AVFMTCTX_UNSEEKABLE))
> return AVERROR(ENOSYS);
> @@ -2372,8 +2376,7 @@ static int hls_read_seek(AVFormatContext *s, int
> stream_index,
>
> seek_timestamp = av_rescale_rnd(timestamp, AV_TIME_BASE,
>
> s->streams[stream_index]->time_base.den,
> - flags & AVSEEK_FLAG_BACKWARD ?
> - AV_ROUND_DOWN : AV_ROUND_UP);
> + AV_ROUND_DOWN);
>
> duration = s->duration == AV_NOPTS_VALUE ?
> 0 : s->duration;
> @@ -2394,9 +2397,16 @@ static int hls_read_seek(AVFormatContext *s, int
> stream_index,
> }
> /* check if the timestamp is valid for the playlist with the
> * specified stream index */
> - if (!seek_pls || !find_timestamp_in_playlist(c, seek_pls,
> seek_timestamp, &seq_no))
> + if (!seek_pls || !find_timestamp_in_playlist(c, seek_pls,
> seek_timestamp, &seq_no, &seg_start_ts))
> return AVERROR(EIO);
>
> + if (s->streams[stream_index]->codecpar->codec_type ==
> AVMEDIA_TYPE_VIDEO &&
> + flags & AVSEEK_FLAG_BACKWARD && !(flags & AVSEEK_FLAG_ANY)) {
> + /* Seeking to start of segment ensures we seek to a keyframe
> located
> + * before the given timestamp. */
> + seek_timestamp = seg_start_ts;
> + }
> +
> /* set segment now so we do not need to search again below */
> seek_pls->cur_seq_no = seq_no;
> seek_pls->seek_stream_index = stream_subdemuxer_index;
> @@ -2423,7 +2433,7 @@ static int hls_read_seek(AVFormatContext *s, int
> stream_index,
>
> if (pls != seek_pls) {
> /* set closest segment seq_no for playlists not handled above
> */
> - find_timestamp_in_playlist(c, pls, seek_timestamp,
> &pls->cur_seq_no);
> + find_timestamp_in_playlist(c, pls, seek_timestamp,
> &pls->cur_seq_no, NULL);
> /* seek the playlist to the given position without taking
> * keyframes into account since this playlist does not have
> the
> * specified stream where we should look for the keyframes */
> --
> 2.25.1
>
>
More information about the ffmpeg-devel
mailing list