My idea is to treat that as equivalent to 2,3/8 -> '(2 (3 . 8)) and let the function figure out how to interpret it. If the + is objectionable syntax because of confusion with infix arithmetic, a comma could be used instead just like we do for integer lists. Personally I wouldn't have an issue with the use of + as a separator given that Lilypond doesn't use infix arithmetic syntax in any user-facing context. But I can also understand why it might be confusing. And I concede that + as a separator isn't applicable to other uses. Allowing fractions in comma separated lists on the other hand seems quite useful.
Saul On Tue, Jan 28, 2025 at 7:11 PM David Kastrup <d...@gnu.org> wrote: > Saul Tobin <saul.james.to...@gmail.com> writes: > > >> On Tue, Jan 28, 2025 at 6:18 PM David Kastrup <d...@gnu.org> wrote: > >> > >>> Saul Tobin <saul.james.to...@gmail.com> writes: > >>> > >>> > I think merging \compoundMeter into \time as a single command would > be > >>> > great. IMO an even bigger improvement would be to support compound > meters > >>> > without requiring Scheme syntax. The parser already supports > >>> > comma-separated integer lists and dot-separated symbol lists. How > feasible > >>> > would it be to support arguments to \time (or to music functions > generally) > >>> > of the form 3/8+2/4 or 2+3/8? > >>> > >>> Nightmarish? > >>> > >>> Don't really see anything that would generalize sensibly. > >>> > >>> In particular since you likely want the above to be (3/8)+(2/4) vs > (2+3)/8. > > > > In terms of the parser, wouldn't it just entail treating + as a separator > > and allowing a list of (fraction | integer)? > > > > Interpreting that data as a time signature I would think isn't > > too different from what compoundMeter does already. > > > > Maybe I'm missing some hidden pitfalls. > > What about 2+3/8 ? What about the whole "+ is not really addition" > concept? This gives a whole lot of garbage unexpected meanings. For > the sake of a single infrequent command. > > -- > David Kastrup >
