use default directions for \voiceN?

[Evan Laforge]

Hi everyone, I'm new to lilypond, so sorry if this is a simple
question, but I wasn't able to find an answer in either the
documentation or the archives.

I have a piece which is mostly monophonic, but occasionally has
another voice.  Up until now I have been using << \\ >> to split the
voices as needed, but it's awkward for several reasons.

One is that lilypond won't tie notes across what it considers separate
voices.  So when switch from \oneVoice to having two voices, it
doesn't know that \voiceOne is actually the same voice as the one in
\oneVoice.  The other thing is that due to pick up notes and the like,
it becomes awkward to factor out repeated patterns.

Both of these problems should be solved if I switch to using two
voices throughout.  However, it's quite ugly to have all the stems
forced in a certain direction and rests in odd places for long
stretches of music that effectively only has one voice.  Ideally I
could write everything as two voices, but with a setting that says if
I measure has visible notes from one voice only (say all the others
have 's' hidden rests), then to treat it as a \oneVoice for the
purpose of beam direction and rest placement.  Or, simpler, a way to
override beam direction for a voice to restore automatic directions
within a given expression.

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?

thanks in advance!