Re: \partcombine terribly broken

[Erik Sandberg]

On Wednesday 31 March 2004 10.42, Han-Wen Nienhuys wrote:
> address@hidden writes:
> > > >c-part-combining.html#Automatic% 20part%20combining
> > > > In the first example, the stems of the two voices are merged.
> > > > According to the documentation they shouldn't be.
> > >
> > > Why?
> >
> > What exactly happens if the two parts differ? Will they (1) always be
> > merged into one voice (and in that case, to which Voice context), or will
> > they (2) be polyphonised as if << \\ >> had been used, or (3) will either
> > one of (1) and (2) be chosen depending on whether the parts differ
> > rythmically?
>
> Hi!
>
> maybe you can look at input/regression/part-combine*.ly for better
> examples.

Thanks! It's alternative (3) then; everything is done in a clever way. (apart 
from the bugs i found :) )

There's one situation I find curious though. The following scores have 
different music but produce the same paper output:

\score {\notes \relative c'' \partcombine
{g4 g a( b) c c r r}
{g4 g r4 r  e e g g}}
\score {\notes \relative c'' \partcombine
{g4 g a( b) e, e r r}
{g4 g r4 r  c  c g g}}

If I would have typeset the second score manually, I would have used polyphony 
notation for the <<e c>> notes, to mark that the c actually belongs to the 
'lower' part. (<c'' e'> suggests that the upper voice plays the c, while <<e' 
\\ c''>> suggests that the lower voice plays the c)

I.e., I suggest an additional rule saying that polyphony notation should be 
used instead of chord notation if the pitch of the second part is higher than 
the pitch of the first voice.


Also, I still find the docu a bit confusing, since the 'up stems' implicitly 
refers to a situation that never is mentioned. I'd suggest to add half a bar 
of polyphony in the first example (like the example from my last bugreport), 
and maybe rewrite the section:
"The first part (with context called one) always gets up stems, and `solo', 
while the second (called two) always gets down stems and `Solo II'."
to something like
"The first part (with context called one) always gets up stems during 
polyphonic sections, and its solo parts are marked `Solo', while the second 
(called two) always gets down stems and `Solo II'."

Erik