Remove selected range/note/rest in a score with parts results in corruption/crash
Reported version
3.3
Type
Functional
Frequency
Once
Severity
S2 - Critical
Reproducibility
Always
Status
duplicate
Regression
No
Workaround
No
Project
Current 3.3.4 (and 3.4 beta) /Windows 10
1) Load this test file (basically, four measures of SATB with some notes): selected remove parts.mscz
2) First staff/Soprano: Delete (Ctrl + Del) the quarter rest, fourth beat, in first measure.
3) Select the TWO first measures, same first staff always, and remove its content (Del)
4) Toggle in Soprano part, result: big problem ie corruption
5) Select the "weird" first measure with a few rests, and delete (Del)
6) Return in main score: empty measure
The file at this state (with warning and corruption): 1 selected remove parts.mscz
- There is other scenarios (included by doing the deletion in first Soprano part) that lead to the same harmful result, and whose source seems to be that the total deletion (Ctlr + Del) combined with that of a content (Del) is not supported, updated in parts, and therefore vice versa.
Fix version
3.4.1
Comments
When deleting the rest (step 2) the measure property
lenis changed from 4/4 to 3/4. But in the parts,lenis not changed. Deleting the parts after step 2 and recreate the parts solves the issue.By the way, same issue can be reproduced in 3.0.
Is this maybe the reason why we're seeing those score where the parts are corrupt and crashes happen on (auto)save?
Isn't this the same as #290096: Remove select range in a score with parts results in corruption ?
I'm working on that. Basically, the measure length and the chord/rest ticks in the parts are not adjusted to the new values. This causes differences (corruptions) between main score and parts, which eventually lead to crash.
In reply to Isn't this the same as … by ABL
From what I found for this issue and what I'm reading at #290096: Remove select range in a score with parts results in corruption it looks the same. So, if you looking into that, I'll stop and find another issue to work on :-)
most probably, reopen if not
Yeah, it's a duplicate. When I filled out the last report, I was pretty sure I filled one or two with this "Timewise" problem in scores with parts. I wasn't sure exactly when. I checked a few minutes, though. I usually find them into the Search. Not this time, I don't know why exactly (my list of bugs report over the last years is probably too long, and some are fixed, some are in progress, some are with PR, in short, it's a bit of a mish-mash, where I got fooled, sorry).
@cadiz1 : no need to apologize :-)
Your testing work (and manual bi-section) is always an extremely valuable resource in finding and fixing bugs. Thank you.
Better two reports about the same bug than none at all ;-)
In reply to @cadiz1 : no need to… by ABL
Thanks Antonio.
"Better two reports about the same bug than none at all ;-) "
Yep. Good point! :)