[MU4 Issue] Stem disappears when adjusting beam height over rests (MU3 import)

[denischevalier]

Describe the bug When adjusting beam height in tuplet where the first note is a ~silence~ rest, the stem disappears

To Reproduce Steps to reproduce the behavior:

1. On MuseScore 3.6.2, create a new score.
2. On the first bar, create a triolet on the first beat.
3. Fill a middle c at the second part of the triolet.
4. Select the first note (silence) of the triolet, and in the palette, go to "beam properties", and select the "bean start" option.
5. Select the last note (silence) of the triolet, and in the palette, go to "beam properties", and select the "bean middle" option.
6. Save your score.
7. Open the score on MuseScore 4.
8. The beam crosses the ~silences~ rest.
9. Try to adjust the beam height.
10. The C4 stem disappears.

Expected behavior At step 8: The bean does not cross the ~silences~ rest. At step 10: The stem stretches to the beam.

Screenshots At step 8:

At step 10:

Platform information MuseScore 3:

• OS: macOS 10.16,
• Arch.: x86_64,
• MuseScore version (64-bit): 3.6.2.548020600,
• revision: github-musescore-musescore-3224f34

MuseScore 4:

• OS: macOS 12.6,
• Arch.: x86_64,
• MuseScore version (64-bit): 4.0.0-223472200,
• revision: github-musescore-musescore-5485621
• Installed through MuseHub

Additional context The bug does not apply if only one of steps 4 and 5 is applied. There is another bug: On MU4, this behavior (applying "beam start" or "beam middle" to a silence) doesn't seem to be doable.

[denischevalier]

After continuing the import of the same score, I see that the bug appears also outside of tuplets, on this pattern:

and this pattern:

[miwelte]

Confirmed. Same here: OS: Windows 10 22H2, Dell 5520 Xeon E3-1505M 32GB RAM MU4 revision: github-musescore-musescore-5485621

[asattely]

~~This is fixed by PR https://github.com/musescore/MuseScore/pull/15383, please confirm~~

Looks like it's not actually. I'll look into this now

[asattely]

#15383 Now fixes this one, if you'd like to use that build to verify the issue has been fixed

[denischevalier]

#15383 Now fixes this one, if you'd like to use that build to verify the issue has been fixed

I'll try it as soon as I have some time, thanks!

[bkunda]

Re-tested on macOS. Certainly, the stem no longer disappears, which would suggest this problem has been fixed. @oktophonie there's obviously some work still to be done here, but I expect what we're seeing in the below video lies outside the realm of the present issue (please confirm):

https://user-images.githubusercontent.com/86290556/222765532-6e43dfa1-894d-4682-8b7c-343d4ca7f187.mov

If addressing the behaviour of the stem (i.e. so it doesn't disappear) does mean this issue has been fully resolved, then we can move it to the Done - Tested again project.

[its-not-nice]

The other issues are indeed quite separate (but could perhaps do with being filed independently!) 🥺