On Wed, 2026-08-12 at 15:43 +0200, Lukas-Fabian Moser wrote:
> Dear Richard,
>
> Am 12.08.26 um 12:35 schrieb Richard Shann:
> > > Could you please provide the original, working (with an older
> > > LilyPond
> > > version) state of your code before convert-ly changed it?
> > Thank you for getting back to me:
> >
> > This is the code working in 2.24
>
> Ah, I see. That code makes assumptions on the internal representation
> of
> grace music that are no longer valid in 2.26.
>
> But the original code is also more involved than necessary. If I'm
> not
> mistaken, it can be replaced by something as simple as:
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> \version "2.24"
>
> #(define (after-grace-factor? x)
> (and (rational? x)
> (< 0 x 1)))
>
> #(define (grace-music? m)
> (and (ly:music? m)
> (music-is-of-type? m 'grace-music)))
>
> attachGraces =
> #(define-music-function (frac note graces) (after-grace-factor?
> ly:music? grace-music?)
> #{
> <<
> #note
> {
> \skip $(make-duration-of-length (ly:music-length note)) *
> #frac
> #graces
> }
> >>
> #})
>
> \score {
> \new Staff
> {
> \attachGraces #3/4 g''4 \grace { f'8 }
> }
> }
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> Incidentally, this version should work out of the box in 2.26 as in
> 2.24. It behaves a bit differently from your original, though: Your
> original (I think) strips away the additional code that make up, for
> instance, an \appoggiatura in contrast to a simple \grace. The code
> I'm
> proposing doesn't touch the inner structure of the grace music.
That seems to work just fine, both in 2.24 and 2.26. I haven't heard of
people trying to use it on appogiaturas - I think the main application
is in 18th c music before they came in. So this is probably a benefit.
>
> But actually, are you aware that stock LilyPond \afterGrace accepts
> an
> optional argument that (I think) does just what your code achieves,
> without you having to enter \grace explicitly? Try:
>
> \afterGrace #3/4 g''4 f'8
I noticed that while digging around; I guess the problem is that it
might be difficult for Denemo to make sense of that syntax without
coding something up in C for it (because when you get to the grace note
it doesn't announce that it _is_ a grace note)
Thanks very much for this fix.
You also tipped me off about the use of (_ string) -
I've tested out the G_ and have run into problems - despite (version)
reporting => 3.0.8 and the documents even for guile 2 saying G_ is the
conventional shorthand for gettext I get a "possibly unbound variable
`G_' " warning from using it. I guess I can always just define it.
Thank you once more,
best wishes
Richard
