On 11/22/05, Han-Wen Nienhuys <[EMAIL PROTECTED]> wrote: > Erik Sandberg wrote: > > > It seems to me that all brackets on a given nesting depth must be the same > > kind of bracket. I.e., the following bracket configuration can't be done by > > a > > single StaffGroup: > > > > \new StaffGroup << > > \new InnerStaffGroup << > > \new Staff c'1 > > \new Staff c'1 > > >> > > \new Staff c'1 > > \new PianoStaff << > > \new Staff c'1 > > \new Staff c'1 > > >> > > > > > > I don't know how common this kind of notation is, but wouldn't it be more > > generic to ditch the systemStartDelimiters property and give the delimiters > > explicitly in the hierarchy? I.e., for the example in NEWS, something like > > this: > > \set StaffGroup.systemStartDelimiterHierarchy = #'(SystemStartSquare > > (SystemStartBracket a (SystemStartSquare b)) c) > > > > (I don't have a particular need for this myself, I just want to bring up the > > issue before backward compatibility becomes a problem) > > YEs, you're right. I contemplated this before, but didn't do it out of > a. lazyness, b. elegance. If we go for this, then a syntax like > > #(Square ((Bracket (x (Square y)) > c) > > looks more logical to me. > > Trevor?
Yup, Erik's exactly right on this one ... I was just at the desk going over this same point. IMO a labelled tree for systemStartDelimiterHierarchy (like above) is the way to go, getting rid of the need for a separate systemStartDelimiters property. I'll sponsor the change if Han-Wen's got the time. -- Trevor Bača [EMAIL PROTECTED]
_______________________________________________ lilypond-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/lilypond-devel
