[FFmpeg-user] Rewriting 'fieldmatch' using the State Machine Method

[Rob Hallam]

On Wed, 11 Jun 2025 at 18:33, Mark Filipak
<markfilipak.imdb-at-gmail.com at ffmpeg.org> wrote:
>
> On 11/06/2025 11.48, Rob Hallam wrote:
> > On Wed, 11 Jun 2025 at 02:45, Mark Filipak wrote:
> >
> >> Apparently, the people that I meant to reach have not been reached. So, in order to gain
> >> credibility, I'm going to show you what I did for 25 years.
> >
> > This is the user list. Did/do you intend to reach the ffmpeg developers?
>
> Yes, I did, Rob. I've applied to ffmpeg-devel three times in the last week and got no response.

It is unfortunate you were discounted, but given the venue I am not
sure what we can achieve.

I say this not to discourage you but with an eye on where the
discussion is likely to go,

  - debates over what 'trial-and-error coding' is/means/how it works
(has already happened)
  - pros and cons of FSM ( / FSA / DFA)
  - bugs in ffmpeg and how critical those are
  - debates about how well fieldmatch works (and its friend ↓)
  - debates about how well documented fieldmatch is
  - debates about other filters
  - ... etc ...

None of these bring you closer to what I presume -- please correct me
if I'm wrong -- is your goal: a rewrite of some part of ffmpeg
implemented in an FSM.

Bearing in mind the audience: what can we concretely do, in your view?

Cheers,
Rob