lilypond-user
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Invisible notes, Scheme contexts


From: Erik Sandberg
Subject: Re: Invisible notes, Scheme contexts
Date: Wed, 14 Dec 2005 12:49:31 +0100
User-agent: KMail/1.8.3

On Tuesday 13 December 2005 22.27, Nicolas Sceaux wrote:
> Han-Wen Nienhuys <address@hidden> writes:
> > I wonder whether this should be changed too. This would mean that
> >
> >    \clef alto
> >    \clef "alto"
> >
> > becomes
> >
> >    \clef #"alto"
> >
> > or
> >
> >    \clef #'alto
> >
> > This will simplify the syntax a bit, at the expense ease of entry.
> >
> > What do you think?
>
> I see some cases where this would be clearly a win. Few weeks ago, a
> singer asked me to change all alto clefs to treble clefs in a
> voice+piano reduction score. Just redefining the \clef music function
> would have made that really easy and quick.
>
> The more music functions there are, and the less hardcoded syntax is,
> the more extensible and flexible LilyPond is. Ideally, even \include
> should be some kind of (not only-)music function, so that users could
> define their own \include version (for including different files
> depending on some parameters for instance, or adding a path).
>
>
> BTW, Erik, weren't you investigating on a more flexible music function
> system, where for instance a "simple_string" argument could be used
> instead of a #"scheme-string"?

I looked into it, but I didn't feel I knew enough to make up my mind.

(it is easy to implement changes, but it is difficult to know what we want our 
grammar to be like)

The most important thing is that we start allowing the #"foo" and #42 syntaxes 
globally for all ints and strings.

The question is whether we should continue to allow the clean "foo" syntax. We 
will run into problems if we allow the 42 syntax for numbers (it doesn't 
allow us to clean up the grammar of music functions, because "c 4" can be 
interpreted either as c4 or as c #4). It would also be slightly inconsistent 
if we allow the "foo" syntax but not the 42 syntax. On the other hand, the 
"foo" syntax is still strictly(?) an improvement over #"foo", and it _might_ 
be good to reserve the "foo" string syntax for future syntax cleanups.

Any opinions on this?

-- 
Erik




reply via email to

[Prev in Thread] Current Thread [Next in Thread]