On Thu, Oct 4, 2012 at 3:26 PM, David Kastrup <d...@gnu.org> wrote:
> It's likely a more important consideration to divide the Guile and
> LilyPond namespace rather than the LilyPond and user space.
>
> To wit, convert
>
> xxx = \something
>
> into
>
> #(define $xxx $something)
I assume that this would mean having to use #(define ....) every time
we wanted to define a variable/command (e.g. notes = { c d e } and red
= \twear #'color #red). From a user's point of view that would be
both cryptic and a bit inconvenient.
I consider manupilating variables a fundamental aspect of using
Lilypond, and its syntax should be as simple as possible.
> However, this would imply
> serious incompatibilities at the Scheme level, and the module system of
> Guile decreases the actual impact of naming collisions since precompiled
> code continues to refer to the variables/identifiers in the Guile
> module.
I don't understand.
Janek
_______________________________________________
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel
