Re: [Lilypond-auto] Issue 1316 in lilypond: Enhancement: improved << \\ >> polyphonic shortcut notation

[lilypond]

Comment #4 on issue 1316 by address@hidden: Enhancement: improved <<  
\\ >> polyphonic shortcut notation
http://code.google.com/p/lilypond/issues/detail?id=1316

I think we need an example of desired input and desired output, where it  
helps for << \\ >> to "Do the Right Thing," so we know what the right thing  
is.

The automatic voice-splitter is "voicify-music" and helper functions  
in 'music-functions.scm'.  It is functional-programming-style Scheme that I  
can't quite follow, but it *does* seem to expand  << a \\ b \\ c >>  into  
the Scheme representation of
<< \context Voice = "1" {\voiceOne a}
   \context Voice = "2" {\voiceTwo b}
   \context Voice = "3" {\voiceThree c} >>

That is rather predictable, so if I really want to connect slurs into the  
<< \\ >>, I just tell Lilypond which context the beginning of the slur  
should have:
 \relative f { d'4 \context Voice="2" g_( << b) \\ d,) >> g4 }

Do we always want the /first/ context to be the same as the existing  
context ?
 "1" -> "PianoUpper"  and "2" -> "PianoUpper2" ?
If so, do we always want to reset the existing to \oneVoice, or would we  
need to save and restore the existing context settings ?