Re: Fix temporary polyphony "for good" (was 'Re: Ties in complex polyphony')

[Kieren MacMillan]

Hi David,

> If it was that easy to do, don't you think this would already have been done?

Not necessarily: different priorities for different people, and the resources 
of the development community (i.e., mostly you) are greatly limited. I can 
certainly imagine that many relatively trivial things have not been done, in 
favour of other (possibly less trivial, but more 
important/critical/fundamental) development.

> What do you think the groundwork of issues 3553, 3303, 3292, and
> particularly 3225 is supposed to be for?
> Have you read the issue description of issue 3225?

1. Don't know.
2. No.
Hence why I asked.  =)

> to do this properly will likely require a different translator group
> with a different way of routing its grobs than Engraver_group does.

Good to know!

> I think it's more inconsistent than difficult.  With the current setup,
> at least you can pick your main Voice to be named "1" or "2" and then
> have the respective split voice fall into place.

Well, for the list archive in general — and the OP (Joshua) in particular — 
here's the \split function I use:

split =
#(define-music-function (parser location music1 music2)
   (ly:music? ly:music?)
   #{ << { \voiceOne $music1 \oneVoice }
         \context Voice = "2" { \voiceTwo $music2 } >>
   #})

(n.b. According to this thread 
<http://lists.gnu.org/archive/html/lilypond-user/2013-03/msg00166.html>, this 
code now correctly handles repeatchord, etc.)

> I am hauling this one in quite slowly. […]
> the code base is not yet where this would be trivial to do.

Fair enough — the \split function will have to do for now.
Any use making it (and maybe its variants) part of the regular distro? Maybe 
named \polyphony, or something helpful?

Thanks,
Kieren.