Re: Fixing regressions and serious issues

[Thomas Morley]

Am Mi., 20. Juli 2022 um 11:40 Uhr schrieb Jonas Hahnfeld via
Discussions on LilyPond development <lilypond-devel@gnu.org>:
>
> On Thu, 2022-07-14 at 17:38 +0200, Jean Abou Samra wrote:

> I agree, even though I'd like to note that only those marked ~Critical
> actually block the release. There are a number of ~Regression issues
> that have been there for 2.22.x or even 2.20.0. Whether we want to re-
> classify some of the more recent ones as ~Critical, is of course up for
> discussion. And fixing issues is obviously still a good idea 😉
>
> > [...]
> > * “Heisen-crashes on Windows with large scores”
> >
> >    https://gitlab.com/lilypond/lilypond/-/issues/6361
> >
> >    A nasty and poorly understood GC problem on Windows,
> >    needs some tough debugging.
>
> This is currently the only issue marked ~Critical, and I agree this
> must be addressed before a stable release.

+1

> I hope I can come back to
> this in the next weeks.
>
> > * “GUILE 2.2 doesn't provide source locations”
> >
> >    https://gitlab.com/lilypond/lilypond/-/issues/5992
> >
> >    Needs figuring out if executing Scheme code with
> >    `compile` is OK performance-wise and how to get
> >    it to display source location info.
>
> I agree this is annoying, and it would be great to improve the
> situation. I'm not sure if this should block the release though.

I beg to differ.
For me it's the most pressing and annoying issue. A stable release
without good error messages is unusable for huge custom scheme
codings.
Granted, recently some workarounds were discussed on -user. Alas, I
can't imagine a stable version needing those workarounds right from
it's release date.
Even it does not crash lilypond, I'd label it Critical as well.

Cheers,
  Harm