Re: [FFmpeg-devel] [PATCH] avutil/log: Replace addresses in log output with simple ids

From: Soft Works <softworkz-at-hotmail.com@ffmpeg.org>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH] avutil/log: Replace addresses in log output with simple ids
Date: Thu, 6 Mar 2025 17:02:01 +0000
Message-ID: <DM8P223MB0365D1DCA266FBD80AE5BF66BACA2@DM8P223MB0365.NAMP223.PROD.OUTLOOK.COM> (raw)
In-Reply-To: <Z8l0H8xQNhwcB5sR@phare.normalesup.org>

> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Nicolas George
> Sent: Donnerstag, 6. März 2025 11:09
> To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] [PATCH] avutil/log: Replace addresses in log
> output with simple ids
> 
> Soft Works (HE12025-03-05):
> > Sorry. So - seriously: what would be your recipe then?
> 
> I see not just a little of non-trivial code for a very minor feature,

Whether trivial or non-trivial, it's definitely just very little code.

> that might be a hint that it would be best to let it go.

This is not a helpful comment. I'm trying hard to be friendly and productive and I think it's not asked too much to at least try doing as well. 

> Also, if somebody is debugging a program using the libraries, the
> pointers are relevant for that program. For that reason, I think the
> change is a bad idea in the library.

It's a valid point, I have acknowledged that already and added a log flag in V2 which allows to control it.

As a further compromise, we could also enable it by default in case when DEBUG is defined, how about that?

Generally, debugging is important without doubt, but it doesn't mean that Millions of users need to see something in the output which is only ever relevant to developers - that's the premise of this patchset. 

And even as a developer, those addresses are interesting only in a very narrow range of cases. 
These addresses have been a major pain point for myself and many others over years when comparing logfiles. Even the best diffing algorithms are getting confused by these addresses and I think this patchset provides a huge benefit for both, users and developers in the future, making their work a lot easier.

> On the other hand, you could do that change in the fftools. The point
> about pointers being relevant does not apply for them, and they can have
> as much global state as they want.

You know that it's not easily possible to do it from within fftools because all libs are logging directly to avutil, so it's not quite clear to me what you are up to. 
Do you mean something like a int(* av_log_format_prefix)(...) callback that fftools could register to?

Thanks
sw

_______________________________________________
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".