Dissappearing elements (dynamics and ties)
I'm afraid this bug report is pretty useless in and of itself: I cannot think of a way to present this issue in a helpful way as I cannot reproduce it. I've been experiencing this on and off for some time and haven't reported it because of the aforementioned lack of reproducibility. Thus, I do not expect anyone to attempt to help me specifically with this issue; I don't see how uploading a file would help, all I could say is "there aren't elements here but there used to be", what does that tell any devs that they don't already know?
So this is more of a 'touching base' report to see if anyone else has had this issue? Sometimes when i load up a score previous elements that I added have disappeared. They do not appear to disappear in any systematic fashion, as some of the same type of element surrounding them are unchanged. Usually they are only the odd dynamic, but today I've had quite a few ties disappear. Sometimes it's only a single element and I don't notice until such time as I have no idea when it happened (or question my memory of whether I actually added anything), however sometimes it's more significant and I have to spend a few minutes adding back the elements (like the ties this morning that triggered this report).
I apologise for being so vague in this report and being unable to provide any specifics. Regardless, I hope this report is at least vaguely useful perhaps in flagging a known issue that is easy to fix in future releases. Alas, I suspect not. If there's anything I can do to be more helpful, please do let me know!
OS: Manjaro running MS V3.4.2 Revision 148e43f.
Comments
Rather than questioning your recollections of whether you made the edits, perhaps we ought to rule out the possibilities that a) you made the edits but forgot to save the score afterwards; or b) you made the edits and saved the score but opened a previous pre-edit version.
In reply to Rather than questioning your… by SteveBlower
Thanks for your reply! Of course in the case of the odd dynamic I cannot rule out those two possibilities, even though I might doubt them. However, in the case of the ties today I can unambiguously rule those out - these particular ties were added long before the changes I made yesterday, and it is the only copy of this file I have.
I should add that in my experience this only happens when opening files for the first time in the morning, and hasn't randomly happened if I've closed and reopened the file before shutting down the machine (thankfully). It doesn't consistently happen opening files after a reboot, however.
I can believe this might happen with ties is the conditions are "right". The code to find the end note for each tie is trickier than you might think, and it's conceivable some sort of sequence of operations involve transposition passages, changing enharmonic spelling, use of linked parts, and possibly other things might have led to a situation where the two notes that were tied together are somehow no longer the same pitch. You might see this right away - the tie might still appear - but on save/reload, we'd be unable to hook the tie back up.
The challenge then would be to figure out what you might have been doing that got the score into this situation before saving. It is possible in fact that attaching the score would help. Because it will tell us if the tie is indeed present in the file but just cannot be hooked up (in which case you won't ever see it). And we might see an error message loading the file into the debugger that gives us a clue as to why the tie can't be hooked up.
Although if you've saved the score again since this happened, it's probably too late - we'd need the score as it existed the first time you loaded it and found the ties goes. I guess the backup version (".scorename.mscz,") would then have the file you'd want, unless you then saved the file, closed it, and opened it again, and saved again.
Of course, it is always possibly you inadvertently deleted them or loaded the wrong version, but somehow it seems kind of unlikely you'd have two versions of a score that differed only in the presence of some ties.
In reply to I can believe this might… by Marc Sabatella
Thanks a lot for your comprehensive reply!
Yes, now you say it I should have copied the file before readding the ties. Alas, it's now useless to you because I've added the ties back, and the backup is overridden. I shall update this post when the issue next happens and attach a file.
That does indeed sound like what it is: I haven't done any of those operations on this particular score, though as you say others may be the culprit. If it's relevant I make extensive use of microtones that I tune with the "AccidentalTuner" plugin (I really should have said that on the original post shouldn't I), which perhaps played a role (although that 'pitch' difference surely shouldn't effect ties being just a change of cents in the inspector?); I also use a fair few tuplets (although not all the deleted ties were within/between tuplets). There are transposing instruments in the score (that I have selected as 'concert pitch'), although the ties also disappeared on non transposing instruments.
I suppose until I get the issue again and attach a file this is all fruitless 'muse'ing (I'm sure you're fed up of that pun), so I shall update when I get the error again. Thanks again for your help.
In reply to Thanks a lot for your… by Nallados
In theory the plugin shouldn’t have caused a problem, but it could be the bug is there. And it doesn’t surprise me that triplets might be involved either.