[FFmpeg-devel] [PATCH 0/3] Clean up build spam from graph css builder

[softworkz .]

> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces at ffmpeg.org> On Behalf Of Derek
> Buitenhuis
> Sent: Mittwoch, 21. Mai 2025 15:27
> To: ffmpeg-devel at ffmpeg.org
> Subject: Re: [FFmpeg-devel] [PATCH 0/3] Clean up build spam from graph css
> builder
> 
> On 5/20/2025 7:44 PM, softworkz . wrote:
> > Hi Derek,
> >
> > thanks a lot for the patch. This partially duplicates what Timo had
> > already submitted:
> >
> > https://patchwork.ffmpeg.org/project/ffmpeg/patch/20250516230202.355445-1-
> timo at rothenpieler.org/
> 
> Dropping this set, then.
> 
> > Regarding patch 3/3, would you mind taking a look at the patch that I
> > have submitted in this regard, from which I believe that it's the "most
> > correct" way:
> >
> >
> https://patchwork.ffmpeg.org/project/ffmpeg/patch/pull.80.v2.ffstaging.FFmpeg.
> 1747549830700.ffmpegagent at gmail.com/
> 
> I am actually unclear on which is "most correct", myself.
> 
> [...]
> 
> - Derek
> _______________________________________________

I can't say that I have great expertise with Makefile development. I read the 
documentation and then I created a super-simple test to verify my understanding.
The Makefile content is this:

# Makefile
%.t: %.r.tmp
	./cmd1 $< $@

%.r.tmp: %.q
	./cmd2 $< $@

%.q: %.r
	./cmd3 $< $@


and every cmd1, cmd2 and cmd3 does nothing else but copy the file. The "source"
file is test.t and no other files exist at the beginning.

When running make, it copies test.t to test.tmp and then test.tmp to test.r and
finally it deletes test.tmp.

When you run make again, it says everything is up to date. When you touch test.t,
it re-builds the result.

That's why I said I believe it's the "most correct" way and that's the behavior
that my patch is introducing. At the same time it allows to get rid of a few
lines (the filtering of files and adding them to the SECONDARY special target).

Thanks
sw