On Tue, Sep 23, 2014 at 03:21:30AM +0200, Michael Niedermayer wrote:
> This fixes the case where muxing fails hard in case of stream copying damaged 
> input
> 
> Signed-off-by: Michael Niedermayer <michae...@gmx.at>
> ---
>  ffmpeg.c |    8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/ffmpeg.c b/ffmpeg.c
> index 47a4144..5492cfe 100644
> --- a/ffmpeg.c
> +++ b/ffmpeg.c
> @@ -670,6 +670,14 @@ static void write_frame(AVFormatContext *s, AVPacket 
> *pkt, OutputStream *ost)
>                       - FFMIN3(pkt->pts, pkt->dts, ost->last_mux_dts + 1)
>                       - FFMAX3(pkt->pts, pkt->dts, ost->last_mux_dts + 1);
>          }
> +        if (pkt->dts != AV_NOPTS_VALUE &&
> +            pkt->pts == AV_NOPTS_VALUE &&
> +            !(s->oformat->flags & AVFMT_TS_NONSTRICT)) { //FIXME use a 
> AVFMT_TS_NEEDS_PTS
> +            av_log(s, AV_LOG_WARNING, "PTS unknown, using DTS: %"PRId64" in 
> output stream %d:%d\n",
> +                   pkt->dts,
> +                   ost->file_index, ost->st->index);
> +            pkt->pts = pkt->dts;
> +        }

why not abort with a AVERROR_BUG or something so we get report of such
cases?

[...]

-- 
Clément B.

Attachment: pgpm0udjrPsAA.pgp
Description: PGP signature

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Reply via email to