Possible bug in partCombine
I have found a strange corner case which partCombine doesn't handle
correctly. I ran into it setting a complex orchestral score, but I've
narrowed down the demonstration to something more reasonably sized.
In this example, I have to do the known workaround in bar 2 of adding
invisible grace notes when one staff has one right after a time signature
change. The bottom line represents the piano part (with the visible
slashedGrace), the top two represent two clarinets, and the third line is a
partCombine of the first two.
Once this happens, partCombine starts putting slashes on any unison flagged
notes from that point onward. Quarter notes are fine; 8th, 16th and so on
get slashes.
The critical thing that triggers this is that one of the two combined parts
has to follow the invisible slashed grace with a flagged note, and the
other part has to be doing anything else, like rests or a quarter note.
This is simple to work around--just change the silent slashed grace notes
to silent plain grace notes--and that's what I'm doing. But it might
indicate something subtly wrong with the code, so I'm reporting this in
case anyone cares.
tiny-example.pdf
Description: Adobe PDF document
\version "2.22.2"
one = \relative c' {
\time 4/4
R1
\time 2/4
\slashedGrace s8
e8 r4.
e8 r4.
e4 r4
e8 r4.
}
two = \relative c' {
\time 4/4
R1
\time 2/4
\slashedGrace s8
%c4 r4
r2
c8 r4.
c4 r4
c8 r4.
}
reference = \relative c' {
\time 4/4
c1
\time 2/4
\slashedGrace g8
c2
R2
R2
R2
}
\score {
<<
\new Staff \one
\new Staff \two
\new Staff \partCombine \one \two
\new Staff \reference
>>
}
Re: Possible Bug with \partcombine?
Joshua Nichols josh.d.nichols at gmail.com writes:
I was recently typesetting a song, using part combine, from a hymnal. I ran
into a problem: everytime I ran \partcombine, I found that unterminated
slur and cannot end slur, were common warnings. I was confused, and so I
created this small snippet that should reproduce the same issues.\version
2.18.2
I would call it a limitation of the design \partcombine.
\partcombine takes its two inputs and combines them when possible
into one LilyPond Voice. The resulting music goes through the
rest of LilyPond as usual, and in this case a pair of slurs starts
when \partcombine has produced two Voices, and ends when it only
has one (differnt) Voice holding the chords, so the slurs are left
dangling.
\partcombine on its own is smart enough to avoid switching from
separate stems to chords in the middle of a slur, but when we need
to override its choices, we need to take over completely.
The simplest solution here is to keep the voices separate through
the slurs
two = {
b a b \partcombineApart c'( b) \partcombineAutomaticOnce a b( c')
\partCombineAutomatic }
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user
Re: Possible Bug with \partcombine?
The simplest solution here is to keep the voices separate through the slurs... Thanks, this is helpful. Is there talk about improving this feature? Or do you know of a hack for getting the desired output? Your solution will be a solid work around from here on out. Thanks. Josh ___ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user
Re: Possible Bug with \partcombine?
Joshua Nichols josh.d.nichols at gmail.com writes: Is there talk about improving this feature? Yes. Someone recently gave \partcombine an option to double-stem unisons http://code.google.com/p/lilypond/issues/detail?id=4112 The example at the top of this follow-up might make more sense http://code.google.com/p/lilypond/issues/detail?id=4198 In the next experimental version (available on Windows and macOS after the one person currently practiced at cross-compiling is back from a trip) you can tell LilyPond just once to double-stem unisons, and then any notes slurred to those unisons are also kept in separate Voices so the slurs have a place to go. The basic mechanism of \partcombine, distributing your music into Voices and letting the rest of LilyPond work on those, will almost certainly stay. That is awkward sometimes, because we are thinking of two voices in the input whether set as chords or double-stemmed. ___ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user
Possible Bug with \partcombine?
I was recently typesetting a song, using part combine, from a hymnal. I ran
into a problem: everytime I ran \partcombine, I found that unterminated
slur and cannot end slur, were common warnings. I was confused, and so I
created this small snippet that should reproduce the same issues.
\version 2.18.2
one = {
c' c' c' c'( c') c' c'( c')
}
two = {
b a b \partcombineApartOnce c'( b) a b( \partcombineApartOnce c')
}
\score {
\new Voice {
\partcombine \one \two }
}
Anybody else having the same problem?
Is it a bug? Or is it a amateur mistake (hoping for the latter).
Thanks,
Josh
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user
