additional flags appearing on cross-staff beams after staffs are hidden
Reported version
3.6
Type
Functional
Frequency
Once
Severity
S3 - Major
Reproducibility
Always
Status
active
Regression
No
Workaround
Yes
Project
1) Select m. 70.
2) Go to Staff/Part Properties.
3) After Hide When Empty, select Always.
4) Force a page break after m. 73.
5. mm. 70 and 71 now appear to be 32nd notes rather than 16th notes. Playback is unaffected.
| Attachment | Size |
|---|---|
| Transfiguration.mscz | 36.3 KB |
Comments
I can't reproduce, they do stay 16th for me
In reply to I can't reproduce, they do… by Jojo-Schmitz
I was able to reproduce it, so I'm not sure why it's not doing the same thing for you. Regardless, when I closed and reopened the score, the extra beam was gone, and they were visibly 16ths, so it seems to be a temporary glitch.
I can reproduce. I assume it's a kind of similar issue to #317885: cross-staff beams on first measure of page disappear on edit to previous page, an error in how the cross-staff beams are drawn in the face of partial relayout. Note they also fix themselves if you make any edit to that system, or any other edit that causes that system to be laid out anew.
So related to, or even a duplicate of #314692: Beams of cross-staff notes in first bar of page sometimes vanish
Certainly related to #285233: [EPIC] Cross-staff notation issues
What's weird is the other known issues involve beams not being drawn. Here it appear it's too many beams being drawn. I kind of think it's really about the beam being drawn twice at slightly different locations, perhaps due to autoplace or the vertical justification algorithm needing to adjust staff distance. I note I can't reproduce this particular aspect of the issue with vertical justification off, although it's also hard to recreate the conditions exactly in that case. Setting both min staff distance and grad staff distance to 3.5 sp gets me approximately the same layout, though.
Some versions of these issues predate 3.6, though, so it's not just about that, I know. Still, the extra space that is always added by autoplace happens at more or less the same point in the layout as extra space added for vertical justification. Ideally this would be before cross-staff beams are finalized, but these issue show me this is obviously not the case - or that the beam layout is somehow missing some aspect of this. Looking at the code for collectPage(), it seems things are in the right order with respect to the call to layoutPage vs Beam::layout(), and unfortunately the glitchiness of this behavior makes it kind of hard to trace in a debugger, but I assume somehow this is where things go wrong.
In reply to What's weird is the other… by Marc Sabatella
Just a quick note to say thank you to both of you. I don't know code from a hole in the ground, and I really appreciate the expertise and attention to detail that goes into MuseScore. It is very much appreciated.