Re: use default directions for \voiceN?

[Urs Liska]

Am 15.08.2011 04:57, schrieb Evan Laforge:
> On Sun, Aug 14, 2011 at 7:27 PM, Kieren MacMillan
> <address@hidden>  wrote:
> > Hi Evan,
> >
> > > I'm guessing both of these are possible, but I don't know quite the
> > > right knobs to tweak to modify a voice in this way.  And I imagine it
> > > must be a fairly common problem, so surely someone else has already
> > > done something like this?
> Ha, I should really do more research.  Just a bit more archive
> trolling turned up \stemNeutral, which appears to be just what I want.
Not really. If you would go for having two voices throughout (which is a 
quite common idiom),
you should use
    \oneVoice
for the sections with only one voice, and
    \voiceOne / \voiceTwo
for the polyphonic sections.
\oneVoice doesn't only affect the stem direction but all relevant 
settings (like rest placing and slur directions - try out <c c'>~ <c c'> 
with \oneVoice and \voiceOne respectively to see the difference).

But if you only have short polyphonic sections you should rather use 
Kieren's example.
To make it more explicit: This construct tells Lilypond that the first 
of the two voices actually continues the single voice before and the 
second voice is added besides it. Therefore you can tie the notes, and 
therefore the first voice also has all the properties from the single voice.

HTH
Best
Urs
>   Looking at the code in there makes a lot of things clearer.
>
> > Instantiate the voices explicitly:
> >
> > \relative g' {
> >   g4 a b c ~   |
> >   <<  { \voiceOne c4 d e f } \new Voice { \voiceTwo a,2 a }>>  \oneVoice
> > }
> Right, my problem was getting a tie from the end of voiceTwo to tie
> into the beginning of oneVoice.  It didn't want to tie even when the
> notes were the same pitch.  But keeping them as two voices solves the
> problem, and \stemNeutral deals with the ugliness of staying in two
> voices when one is silent.  So I think I'm good now, thanks :)
>
> While I'm here... "assignment" in lilypond makes macros, but they
> don't take arguments.  However, it's very common for music to repeat
> except a little bit at the end, or a few notes in the middle or
> something.  It looks like you can do full on functions by dropping
> down into scheme, but then you (apparently?) lose the nice syntax.  Is
> there a nice way to factor "almost the same" chunks of music?  Also,
> the traditional kind of macro expansion requires you to know
> beforehand where all the "holes" are to fill and provide them
> explicitly, and that doesn't apply so well to music.
>
> Also it feels like I would like some kind of "negative start time" for
> pick-ups, so I could say e.g.
>
> partTwo =<<
>    \new Voice { \voiceOne { ... } }
>    \new Voice { \voiceTwo { \pickup 1 measure { 2s c4 c | ... } } }
> music = { \partOne \partTwo }
>
> This way the pickup from partTwo would overlap with the end of
> partOne.  Otherwise, you have to split the parts up in a somewhat
> unnatural way, just to insert a partOnePointFive that mixes the pickup
> of partTwo with the tail of partOne.  And that leads to more messiness
> if you have a rondo kind of A B C B D B thing.  Does that make sense?
> Am I doing it wrong?
>
> _______________________________________________
> lilypond-user mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/lilypond-user