ties starting in nested tuplet lost after save and reopen
Reported version
3.0
Priority
P1 - High
Type
Functional
Frequency
Few
Severity
S3 - Major
Reproducibility
Always
Status
closed
Regression
No
Workaround
No
Project
see attached file (fragment from a well known orchestral work):
1. try to set tie to following positions:
- between bar 2 and bar 3 the "c"
- bar 6 above the two "b-flat"s
- between bar 9 and bar ten the "d"
- bar 10 the second "c-sharp" and the third
- bar 11 the first "c" and the second
2. save it
3. reopen it: the ties are gone.
| Attachment | Size |
|---|---|
| tie-problems.mscz | 16.06 KB |
Comments
See also https://musescore.org/en/node/285242.
The problem is that the ties start in a nested tuplet.
This appears to work in master for all cases I have tested. If someone has a counterexample, feel free to attach it and change the status. But my guess is, this was due to integer tick calculation rounding - both for nested tuplets and for other tuplets that don't divide nicely into 480 - and the problems are likely all fixed with the change to fractions.
I would suggest that this be marked as fixed with a note to the PR that fixed it in master (future 3.1).
Agreed. Again, we can re-open if we find other examples that fail. I'm actually amazed we the Fraction transition has gone as smoothly as it has, only a couple of regressions found so far that are connected to it, most already fixed. But, then, not that many people are trying the nightly builds.
Here is the PR that implemented fractions for ticks: https://github.com/musescore/MuseScore/pull/4630
Automatically closed -- issue fixed for 2 weeks with no activity.