Thank you for looking at what I sent earlier, Carl. Turns out that it *doesn’t* actually work like I thought it did.
Here is a real tiny example of the problem:
\version "2.18.2"
\relative c'' {
\tieUp
<d f>2 <cis e>2 ~ |
<cis!-\shape #'((0 . 2) (0 . 2) (0 . 2) (0 . 2))~ e-\shape #'((0 . 2) (0 . 0) (0 . 0) (0 . 2))~>1 |
<cis! e>1 |
}
That yields this:
First tie moved as expected, but the second one (E to E) doesn’t budge (it should be inverted with the values I gave it).
(This would actually solve my problem graphically, but when I tried it in my actual score, things didn’t work as expected at all, with things moving in seemingly random ways I cannot replicate in a tiny example.)
But if I insert a \break into the mix:
\version "2.18.2"
\relative c'' {
\tieUp
<d f>2 <cis e>2 ~ |
\break
<cis!-\shape #'((0 . 2) (0 . 2) (0 . 2) (0 . 2))~ e-\shape #'((0 . 2) (0 . 0) (0 . 0) (0 . 2))~>1 |
<cis! e>1 |
}
I get this:
Now the second \shape command works, but the first one no longer does.
I think I’ve uncovered a genuine bug. I doubt very much that this is anyone’s idea of how this should work. I guess I’ll have to report it.
Best
-Arle