Re: [FFmpeg-devel] [PATCH 11/31] fftools/ffmpeg_filter: stop accessing encoder AVCodecContext

From: Anton Khirnov <anton@khirnov.net>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH 11/31] fftools/ffmpeg_filter: stop accessing encoder AVCodecContext
Date: Fri, 05 Apr 2024 19:09:53 +0200
Message-ID: <171233699309.28449.1024041800145583511@lain.khirnov.net> (raw)
In-Reply-To: <CAKKYfmHzWEq=eCezTQSwLuais+8xUqjBsq9SSvbLBtWi45YL8w@mail.gmail.com>

Quoting Dennis Mungai (2024-04-05 19:07:19)
> On Fri, 5 Apr 2024 at 19:54, Gyan Doshi <ffmpeg@gyani.pro> wrote:
> 
> >
> >
> > On 2024-04-05 10:20 pm, Dennis Mungai wrote:
> > >
> > > Does this imply that down the line, with this and additional patchsets,
> > > that FFmpeg can handle tasks such as stream copy and filtering in the
> > same
> > > invocation?
> >
> > What do you mean? If you map a stream twice, you already can do that now.
> >
> > Regards,
> > Gyan
> >
> 
> What of doing the same *without* mapping the stream twice?

What's the issue with mapping a stream twice?

> A complex filtergraph could have a stream mapped, split and then passed to
> a copy codec operation, for example, without needing the double re-map
> implied above.

Filtergraphs process decoded frames, streamcopy bypasses them entirely.
None of this affects streamcopy.

-- 
Anton Khirnov
_______________________________________________
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".