Fingering repositions unexpectedly when multiple elements set to same offset using Inspector

• Aug 17, 2019 - 20:53
Reported version
3.2
Priority
P2 - Medium
Type
Functional
Frequency
Once
Severity
S3 - Major
Reproducibility
Always
Status
active
Regression
No
Workaround
No
Project

MS 3.2.3.

  1. Open the attached score.
  2. Right-click on any of the RH fingering elements and chose "Select All Similar Elements."
  3. In the Inspector change the X-offset to 0.5 and the Y-offset to -0.3.
  4. Save and reopen the file.

Expected result: All RH fingerings should be in the same position they were when saved.
Actual result: A considerable number are out of position, but It's difficult to see what the common factor is. Probably also affects "Fingering."

See also: #289943: Inspector: min distance and offset values do not update immediately when moving into the skyline using Inspector controls.
And #291990: Recently adjusted fingering can shift when changing style setting.

Attachment Size
fingering_adjustment.mscz 35.27 KB

Comments

Confirmed.

The problem is here. => some minimum distance values are automatically set to "-999", but don't appear on the score at this time.

Workaround:
Normally, after saving and reopening the file, selecting all RH fingerings and resetting the minimum-distance (with the recycle button) should solve the problem.
However, if you try to correct it in another way, the ​​ x, y, and minimum-distance are vary in ranges you do not set.

Priority P2 - Medium

Yes, I think that trying to set literally the same X & Y for all the fingerings is problematic because that would normally require reducing the minimum distance for some of them, but not all. By using the Inspector to do it all at once, you are effectively forcing them to have the same minimum distance as well, which results in the issue that appears on next relayout even without save & reload. Even the offsets themselves can't literally be same - some of them are already at the default but others were autoplaced higher, and now you're forcing them to a specific offset that places them on the note.

Basically, you really can't set offsets for fingerings the same like that. I guess maybe we should figure out a way to disable it.

Title RH Fingering repositions unexpectedly when file is saved and reopened Fingering repositions unexpectedly when multiple elements set to same offset using Inspector

I'm having a possibly related issue issue with 3.3 RC (3.3.0.23833) as well. In my case, at least, it is not necessary to select multiple fingerings for this bug to manifest, and the Y offset is reliably set to the incorrect value of -2.09sp (I generally want to set it to -0.50sp if the default offset isn't sufficient). Does that sound like the same bug, or should I raise a new issue for it?

Hard to say, but it's probably something different, could well be not a bug at all (could be automatic placement jus performing its job). Best to attach your score and precise steps to reproduce what you are seeing.

In reply to by Marc Sabatella

Precise steps:

  1. Use note edit mode to enter some notes.
  2. Attach a fingering to a note by double-clicking it in the palette.
  3. Use the Inspector to adjust the Y offset of the fingering (in my case, usually to -0.50 sp, because usually I only need a small change from the default positioning).
  4. Continue entering music.
  5. At some point (not every time—this is an intermittent bug), watch the fingering in question suddenly jump to (always) -2.09 sp.

At that point, things are dicey. Sometimes I can edit the fingering to restore the desired offset of -0.50 or 0.00 sp. Sometimes I can't. Sometimes deleting the fingering in question and recreating it helps. Sometimes it doesn't. Sometimes if I try to restore the offset I wanted, it interprets 0 sp as right in the notehead instead of wherever the default positioning puts it. There's no obvious pattern. Sometimes I have to edit the MSCX file to get the right positioning.

This worked fine up through 3.2, I believe (that is, fingerings didn't move after they were entered), so I think the behavior I'm seeing is a regression.

It's still not clear what you mean - which note (it matters, because positioning is different according to which staff line, which voice, what other notes are present etc)., also which finger (piano fingering has different rules than guitar fingering, etc). Alos, the style settings you have in your score are relevant. So please, attach an actual score and tell us precisely which note you are adding a fingering to, which fingering you are adding, etc.

In reply to by Marc Sabatella

It’s not clear to me either! I can’t predict whether this will happen in any particular case; I just know that it happens sometimes. These are single-line lead sheets with violin fingering (not guitar fingering). I’ll attach a score when I’m at my computer, but anything in https://github.com/marnen/kostakowsky/tree/master/source/tunes is likely to exhibit the issue. The stylesheet I’m using is https://github.com/marnen/kostakowsky/blob/master/source/tunes/default… .

I should say that even entering notes on other systems makes it jump to -2.09 again. Even if this is something to do with automatic positioning, that surely can't be correct behavior.

I can reproduce in the release candidate of 3.3, but not with a current debug build. What's weird I can't I recall changing anything since the RC (or, for that matter, since 3.2.3) that would explain why it works now if it didn't work in the RC - or why it wouldn't work in the RC if it did in 3.2.3.

But somehow I do think the issue geetar mentions, where the Inspector doesn't update immediately, provides a clued. If I adjust the fingering in 3.3. RC, then press Esc to deselect the fingering, then select it again, I see it is already set fo -2.09, it just hasn't taken effect yet. This tells me that the effect is actually connected to me fix for #291287: Fingering behaves unpredictably when dragged. Not that this fix caused it, but somehow some of the same code is involved. The weird thing is, if I do the same in my current build - adjust offset with Inspector, Esc, then select again, the offset does not show as changed.

BTW, the fact that the offset changes is not in itself a bug - the original offset of "0" is a "relative" offset whereas the -2 is absolute. The code is supposed to deal with this properly but apparently sometimes it doesn't. Not being able to reproduce it in my debug version is not helping.

Anyhow, please do submit a new issue for this. I don't believe for a minute that it is truly fixed just because i can't reproduce it in the current build.

OS: Windows 10 (10.0), Arch.: x86_64, MuseScore version (64-bit): 3.3.0.8624, revision: f85163f

I can confirm the "jumping" behaviour. A workaround:

  1. Open Freilach_41.
  2. Delete the fingering in m12, and replace it with the same by double-clicking from the fingering palette.
  3. In the Inspector, click once on the Y-offset up-spinner, then use the down-spinner to place the fingering correctly.
  4. Add another note to the measure.

Result: This time the fingering stays put. The first adjustment places the fingering inside the skyline (?) and, somehow, this resets the Y-offset value to its true (i.e. WYSIWYG) value. Thereafter, the WYSIWYG value is retained and there is no jumping. See #289943: Inspector: min distance and offset values do not update immediately when moving into the skyline using Inspector controls?