[FFmpeg-devel] libavformat/movenc: Is packets being more than INT_MAX apart always an error?

[FFmpeg-devel] libavformat/movenc: Is packets being more than INT_MAX apart always an error?

[facefunk]

It's commonly the case with forced subtitle tracks that samples are more
than INT_MAX microseconds (35m47.48s) apart. This being the case, packet
duration will be deemed out of range by check_pkt and the packet dts set to
an incorrect value, even if pkt->duration is actually valid.

I had begun working out how to inject extra packets to satisfy this check
but I thought I would ask. Is this check necessary for mov_text packets?
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] libavformat/movenc: Is packets being more than INT_MAX apart always an error?

["zhilizhao(赵志立)"]

> On Jul 20, 2022, at 8:18 PM, facefunk <facefunk@fcfnk.com> wrote:
> 
> It's commonly the case with forced subtitle tracks that samples are more
> than INT_MAX microseconds (35m47.48s) apart. This being the case, packet
> duration will be deemed out of range by check_pkt and the packet dts set to
> an incorrect value, even if pkt->duration is actually valid.
> 
> I had begun working out how to inject extra packets to satisfy this check
> but I thought I would ask. Is this check necessary for mov_text packets?

I think so, it’s still limited by the 32bits sample_delta in stts.

On the other hand, I have an idea that a timescale like 100 can be used in
that case, which is 2^31/100/3600 = 5965 hours.

> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

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

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] libavformat/movenc: Is packets being more than INT_MAX apart always an error?

[facefunk]

Wow! Thanks!

That really works. I had no idea we could just play with timescale like
that.

On Wed, 20 Jul 2022 at 13:41, "zhilizhao(赵志立)" <quinkblack@foxmail.com>
wrote:

>
>
> > On Jul 20, 2022, at 8:18 PM, facefunk <facefunk@fcfnk.com> wrote:
> >
> > It's commonly the case with forced subtitle tracks that samples are more
> > than INT_MAX microseconds (35m47.48s) apart. This being the case, packet
> > duration will be deemed out of range by check_pkt and the packet dts set
> to
> > an incorrect value, even if pkt->duration is actually valid.
> >
> > I had begun working out how to inject extra packets to satisfy this check
> > but I thought I would ask. Is this check necessary for mov_text packets?
>
> I think so, it’s still limited by the 32bits sample_delta in stts.
>
> On the other hand, I have an idea that a timescale like 100 can be used in
> that case, which is 2^31/100/3600 = 5965 hours.
>
> > _______________________________________________
> > ffmpeg-devel mailing list
> > ffmpeg-devel@ffmpeg.org
> > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> >
> > To unsubscribe, visit link above, or email
> > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".