Re: Oodles of bugs

[Reinhold Kainhofer]

Am Samstag, 8. Januar 2011, um 22:25:51 schrieb David Kastrup:
> So what do we get?  The part combiner messes with the beaming in a
> manner that does not look like an improvement (arguably by design), 

The "problem" with beaming is that the chord (<cis e>16) on the second 8th of 
the third beat is placed in a different voice than the previous eighth and the 
following 16th and 32th...

The part-combiner uses three different voices: One for combined melodies (like 
a2, solo, solo 2 and chords), and two for separate melodies. 
As lilypond lacks any way for cross-voice beaming/slurin/tieing, there is no 
chance to get what you want without forcing the two voices to be apart for 
basically the whole measure (i.e. place a \partcombineApart at the begin of 
the measure and \partcombineAutomatic at the end to use standard part-
combining again after the measure)


> and
> it starts off a spontaneous "Solo" on the last note from violin 1 that
> is not quite called for since violin 2 is not yet finished.

Because the part-combiner does not check how long notes are... It just checks 
onsets of notes.
You can, however, force the partcombiner to a given combination strategy by 
inserting \partcombineApartOnce before the final e16 of the first voice of the 
measure. 

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