Correct placement of slash notes lost when score saved
Reported version
3.0
Priority
P2 - Medium
Type
Functional
Frequency
Once
Severity
S3 - Major
Reproducibility
Always
Status
active
Regression
Yes
Workaround
No
Project
MS 3.3 RC3.
- Create a measure consisting of voice 1 and voice 3 notes, such as the following:
- Select just voice 3 notes and apply "Toggle Rhythmic Slash Notation" from the Tools menu:
- Save and reopen the file.
Expected result: No change.
Actual result: Voice 3 notes have shifted downwards and now collide with the staff.
This worked OK in MS 2.3.2.
Comments
There is indeed supposed to be a 0.5sp offset, although it was always debatable as to whether it was needed. Looks like a change was made to Chord::setSlash during the beta - at https://github.com/musescore/MuseScore/commit/96af32eee7d559cce443aba1f… - that is probably what broke this. WIth the offset not being incorporated into the actual offset property, it is no longer written. Perhaps somehow some change needed to be made there to keep in step with the autoplace changes, but that apparently was not the right change. Assuming I'm wrong about when it broke, but I think I am.
FWIW, it was debatable from the beginning as to whether the offset should have been there in the first place, but I felt it made sense, and would like to see it reinstated.
You beat me to it, Marc! I was just about to point to the same change in the same commit, but I have learned to refresh the page before posting (but not before copying my comment to the clipboard!). That is indeed what broke this. I am not sure why Werner felt it necessary to make that change, but I suppose he must have had his reasons. Maybe we could try changing it back and seeing what breaks.
OS: Windows 10 (10.0), Arch.: x86_64, MuseScore version (64-bit): 3.6.2.548021803, revision: 3224f34
Came across this issue again:
P.S. The slashes also look out of alignment with the stems. See #319872: Alignment between slashes and stems needs tweaking.
As far as I understand the analysis above this is the case ever since 3.0, a regression vs. 3.0 Beta, or rather vs. 2.x even