Joseph Rushton Wakeling <[email protected]> writes:

> On 10/08/12 02:23, David Kastrup wrote:
>> It would have been 3+2/8 at any rate since throwing parens into the
>> token syntax would have further messed up the ambiguities, and forms
>> like 3/2+2/5 would not likely have worked.
>
> Could it improve matters to have instead something like,
>
>    3:2 + 2:5
>
> ... i.e. to use a different symbol from / to separate the upper and
> lower values of time signature?

No.

> Then if you have something like,
>
>    3+2/3:4 + 3+2/3:8 + 3:10
>
> ... there should be little ambiguity for the parser.

This is not something to be done in the parser but in the lexer.

> I accept it's a major breaking syntax change (although the deprecation
> path and conversion rules ought to be pretty simple), but could this
> be a case where it's worth it?

No.  There is no reason to throw further characters into the mix that
was already too heavy-handed just for the sake of \time.

-- 
David Kastrup


_______________________________________________
lilypond-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Reply via email to