On Mon, May 27, 2013 at 10:44:03AM -0400, Justin Ruggles wrote: > On 05/27/2013 10:31 AM, Kostya Shishkov wrote: > > On Mon, May 27, 2013 at 04:29:19PM +0200, Anton Khirnov wrote: > >> > >> On Mon, 27 May 2013 06:29:15 +0200, Kostya Shishkov > >> <[email protected]> wrote: > >>> --- > >>> libavformat/smacker.c | 1 + > >>> 1 file changed, 1 insertion(+) > >>> > >>> diff --git a/libavformat/smacker.c b/libavformat/smacker.c > >>> index 17c7c52..9084e96 100644 > >>> --- a/libavformat/smacker.c > >>> +++ b/libavformat/smacker.c > >>> @@ -334,6 +334,7 @@ static int smacker_read_packet(AVFormatContext *s, > >>> AVPacket *pkt) > >>> if(ret != frame_size) > >>> return AVERROR(EIO); > >>> pkt->stream_index = smk->videoindex; > >>> + pkt->pts = smk->cur_frame; > >>> pkt->size = ret + 769; > >>> smk->cur_frame++; > >>> smk->nextpos = avio_tell(s->pb); > >>> -- > >>> 1.7.9.5 > >>> > >> > >> I wonder if this is a good idea. > >> > >> It's not a timestamp stored in the file, so the generic code can make it > >> up as > >> well as the demuxer. > > > > In theory in should, in practice it often makes it up to zero so the frame > > is > > dropped later. > > You could also try setting the stream start_time to 0 and see if that > fixes it.
it doesn't _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
