More Post-Cadenza Music

[Will Oram]

Somehow the discussion of this subject got bumped to bug-lilypond, so 
I'm resetting it to here.

To re-describe: I have music with solo and accompaniment. The solo has 
a cadenza that the accompaniment has to sit through. By using skips to 
round to the nearest measure, the cadenza works out to exactly 44 
measures (by manual count). If I tell the accompaniment to skip the 
cadenza (s1*44), it looks OK in print but ends up playing between a 
quarter and half a beat early in MIDI. If I extend the skip any amount 
between 0 and 1 measure, I get barline warnings and the MIDI doesn't 
change. Extending it 1 full measure (*45) makes the accompaniment start 
1 measure late, as expected.

Han-Wen offered a Scheme function to count for me, viewable here: 
http://mail.gnu.org/archive/html/bug-lilypond/2004-03/msg00345.html . 
Using this, the cadenza prints, then post-cadenza solo, then a few 
measures of rest, then post-cadenza accompaniment.

Sorry to post this again, but it's important to have the PDF look good 
and the MIDI sound right in this case.

Will Oram
spamguy ^^AT^^ foxchange.com // william.oram ^^AT^^ cwru.edu // AIM 
spamguy21