Re: Patch: issue #659

[Marc Hohl]

Neil Puttock schrieb:
> 2010/1/8 Marc Hohl <address@hidden>:
>
>   
> > Yes, the picture in the tracker has one loop missing, but in
> > all other sources, the sign looked (mainly) as mine.
> >     
>
> How do they deal with the segno at the end of a line though?  The
> tracker example is aligned as a normal barline (i.e., with the
> double-bar at the right edge), but this is going to be tricky to
> achieve just for this case using BarLine grobs.
>   
Hmmm - you are right. Is it possible to whiteout this small part
of the barlines?
> [...]
> > Phooey, I don't know if I am able to cope with this...
> > Generally, I like this sign, because it sticks in the staff lines and
> > therefore doesn't occupy space above the staff (in comparison
> > to the normal segno mark), but it seems not to be
> > used very often.
> >     
>
> I agree it's not really worth the bother since it likely to be a niche
> symbol; on the other hand, your current implementation represents a
> departure from the default behaviour of all the existing barlines: all
> of them scale properly (either using 'line-count/'line-positions or
> 'staff-space).
>   
I am not sure whether a smaller variant (with less loops) even exists in
current scores - perhaps there are people outside with more knowledge of
orchestral pieces. Is there something like a shortened varsegno sign?
>   
> > I also had the idea to enhance the \repeat command, so one can code
> > \repeat segno { ... } and lilypond inserts .S and S. instead of ||: and :||
> >     
>
> This would be great, and a much better fit for the new segno.
>
>   
> > but I don't have a clue (yet) how to cope with the coda, because coda
> > signs are not barlines, so I'll have to deal with \marks and stuff.
> >     
>
> I don't think that would be too difficult, since it's no problem for
> an engraver to create more than one type of grob.
>   
So the "\repeat segno" command would change the bar lines accordingly,
and \alternative { { .1. } { .2. } } should handle the coda constructs, 
right?
Ideally, there should be some kind of user interface to handle the different
segno possibilities, like

http://lsr.dsi.unimi.it/LSR/Snippet?id=198

but with a more generic, "lilypondish" approach.

Greetings

Marc