Re: rests in partcombine

[Reinhold Kainhofer]

Am Freitag, 4. Februar 2011, um 07:17:43 schrieb Keith OHara:
> Reinhold, and list,
>   An aesthetic flaw slipped in with the improvements to the part combiner. 

Nope, that problem has always been there... My patches didn't change how the 
part-combiner detected combined/apart periods.


> Rests sometimes adopt voiceOne or voiceTwo settings at surprising times. {
>    \partcombine
>    { R1*4 R1 }
>    {R1*2 R1 c' R1}
> }
> The split R1*2 R1 comes from the second part having a one-measure
> cueDuring.  

I'm running into these problems all the time, too.

> If I back out the commit that gave voice-positions to
> multi-measure rests, b5603acc..., then of course the rests are all
> centered.
> 
> Possibly, the automatic part combiner is including the R1 as part of a
> brief both-parts-playing-but-apart period.  In reality it is either
> parts-together or SoloII, and it would be more logical to have the R1 at
> center-height.

And thanks to my improvements you can now override the part-combiner's wrong 
decision... You can now use \partcombineUnisonoOnce or \partcombineSoloIIOnce.
Unfortunately, \partcombineUnisonoOnce before the R1 shows another problem in 
the part-combiner: The rest is missing.

What does work is using \partcombineSoloIIOnce before the R1 in the second 
voice:
{
   \partcombine
   { R1*4 R1 }
   {R1*2 \partcombineSoloIIOnce R1 c' R1}
}



Cheers,
Reinhold
-- 
------------------------------------------------------------------
Reinhold Kainhofer, address@hidden, http://reinhold.kainhofer.com/
 * Financial & Actuarial Math., Vienna Univ. of Technology, Austria
 * http://www.fam.tuwien.ac.at/, DVR: 0005886
 * LilyPond, Music typesetting, http://www.lilypond.org