Re: [FFmpeg-devel] [PATCH 0/6] avformat/movenc: normalize on AC-3 parser usage

From: "Jan Ekström" <jeebjp@gmail.com>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH 0/6] avformat/movenc: normalize on AC-3 parser usage
Date: Thu, 9 Jun 2022 10:00:25 +0300
Message-ID: <CAEu79SaXYNu=wrbQ9h28Ur-a8yr6N86O4pRt2Un_sgNsyavWfg@mail.gmail.com> (raw)
In-Reply-To: <CAEu79SYPr0bm5vAToxoV6mQxsuObTro_ooLu0OWR-xP5U8VVuA@mail.gmail.com>

On Tue, Jun 7, 2022 at 9:58 AM Jan Ekström <jeebjp@gmail.com> wrote:
>
> On Fri, Jun 3, 2022 at 3:46 PM Jan Ekström <jeebjp@gmail.com> wrote:
> >
> > The simplified parsing currently in `mov_write_ac3_tag` trusts the content
> > of the packets a bit too much (the AC-3 parser returns all data fed to it,
> > including any possible data before the start code), while the existing E-AC-3
> > logic does proper header validation by utilizing the (E-)AC-3 parser.
> >
> > Thus, normalize on AC-3 parser usage for both AC-3 and E-AC-3.
>
> Ping on this patch set, as this seems to improve the state of affairs
> with the input being MPEG-TS muxes where PES packets do not start with
> a start code (the previous ad-hoc parsing was way too trusting and
> would write bogus values based on the last N bytes of a previous
> packet into the AC-3 sample description).
>
> When applying, `avcodec/ac3_parser{,_internal}: expose AC-3
> bit_rate_code` will have a minor bump as technically a
> private-yet-utilized-from-lavf structure has something appended to it.

If there are no further comments, I will pull this in tomorrow, as
this has been generating valid sample descriptions for a piece of
software which is picky that the data is correct.

Jan
_______________________________________________
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".