Re: parser variables persist beyond { } scope

[Mark Polesky]

Ian Hulin wrote:
> Is there a way of do this sort of thing sometime in the future? 
>
> \afterGrace {d1 {c16[ d] }} % works as now - except extra braces
> \afterGrace {d1 {c16[ d] } 15 16} % extra optional parameters

Well, now I think we're pretty much back at Han-Wen's original
suggestion, which was to have two commands, one for temporary
afterGraceFraction overrides, like

  \afterGrace <music> <music>
  \afterGraceCustom <pair> <music> <music>

Though I admit, I'm not particularly fond of this solution.

I considered what it would take to truly implement your suggestion
above. It can't be done precisely as you've laid it out. At first
I considered this syntax:

\afterGrace { d1 { c16[ d] } }
\afterGrace { d1 { c16[ d] } #'(1 . 2) }

But that would change the 2-argument signature to one argument
with two (or three) parts to it. Presumably the typecheck for the
one argument would be "ly:music?", but that wouldn't work since
lone scheme-pairs aren't incorporated into music-expressions. You
can see this for yourself if you do:

\displayMusic { c c #'(1 . 2) }
==>
(make-music
  'SequentialMusic
  'elements
  (list (make-music ...)
        (make-music ...)))

The scheme-pair simply isn't included in the music-expression.
Maybe there's another typecheck that would allow this, but I'm not
aware of one.

Perhaps it could be done by enclosing the \afterGrace along with
its own arguments in a set of curly-braces:

{ \afterGrace d1 { c16[ d] } }
{ \afterGrace d1 { c16[ d] } #'(1 . 2) }

The idea then would be to allow an optional argument in the
signature:

afterGrace =
#(define-music-function
  (parser location main grace . fraction)
  (ly:music? ly:music? . pair?)
  ...)

This would require modifying the define-music-function defmacro
(on line 764 of music-functions.scm), which may be interesting,
but most likely would create far more problems than it would fix.
Not even sure if it would be possible. I could be wrong, but I
expect a firm veto from Han-Wen...

But I am curious; this is interesting. Could it be done?
- Mark