Simon Albrecht <simon.albre...@mail.de> writes:

> Hello,
>
> I have written a Scheme function to automatically produce all parts
> given through a symbol-list. Here is a boiled-down, dummy example:
> %%%%%%%%%%%%
> \version "2.19.28"
> printParts =
> #(define-scheme-function (abbrs) (symbol-list-or-symbol?)
>    (let ((abbrs (if (symbol? abbrs) (list abbrs) abbrs))
>          (process-part (lambda (abbr)
>                          (print-book-with-defaults
>                           #{
>                             \book {
>                               \bookOutputName #(string-append
> "wrap-multiple-books-"
>                                                  (symbol->string abbr))
>                               \score { c }
>                               \paper { #(set-paper-size "a10") indent = 0 }
>                               \header { tagline = ##f }
>                             }
>                           #}))))
>      (map process-part abbrs)))
> \printParts one.two
> %%%%%%%%%%%%
>
> This compiles fine inasmuch as the output is just as intended, but it
> does sport a:
>
> <0>"/home/simon/lilypond/tests/wrap-multiple-books.ly:17:1: error: bad
> expression type
>
> \printParts one.two"
>
> Is this worth a bug report about this error appearing too easily?

No.  You use map instead of for-each, you use define-scheme-function
instead of define-void-function.

If you bend over backwards in order to pass a value to LilyPond, it
seems absurd to complain about LilyPond trying to make sense of it.

define-scheme-function is _explicitly_ a function returning a value for
LilyPond to interpret.

-- 
David Kastrup

_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Reply via email to