Changing Time Signature causes existing measure rest to move
I have experienced on several occasions where changing the time signature moves the measure rest to another spot in my score. The rest will jump to somewhere else in the score. Often it jumps to the end just beyond the staff (in continuous view), but testing this score it jumped to another random measure or simply disapears in every staff in that measure.
I work around this by inserting a key signature into the measure where I'm changing the time signature, insert new time signature, click on measure to select entire measure, "Delete" to delete the measure with the meaningless key signature. It also works if I add staff text rather than a key signature. Sometimes, but not always, I can delete the offending measure without adding anything to it. When that measure is deleted the time signature is still there and the rests are all lined back up where they belong. Here is a link to the score it last happened on:
https://musescore.com/user/6105546/scores/2934426
On this score append a few measures to the end and change one of the new measures time signature and you will see what I mean.
Comments
Hmm. Unfortunately it seems something is corrupt in this score, although I can't quite put my finger on exactly what the problem is. Wish I could say more. Are you using local time signatures anywhere in this score - places where one staff has a different time signature than another for the same measure? This feature is kind of experimental. While it basically works, there are a few known issues where this can lead to corruption that might look like what we are seeing here.
In reply to Hmm. Unfortunately it seems by Marc Sabatella
There are no local time signatures and this is not the first time I've seen this. I've had other scores that had multiple time signature changes where this happened - with no local time signatures. I've had this happen on scores with relatively few (5-6 instruments) as well as huge ones such as this.
Local time signatures are a totally different issue for me and I won't even go down that street in this thread.
In reply to There are no local time by mike320
If you can find a shorter score with this problem, please post it - that will make investigation simpler.
In reply to If you can find a shorter by Marc Sabatella
We were apparently typing at the same time. I can't find a shorter score. You can shorten it as much as you want and the problem should still occur. I just tested it by decreasing the length to 3 measures and the problem was definitely there still.
The current nighly build, 6a0fb3f, crashes on that score. Stack trace:
1 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f1
2 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
3 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
4 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
5 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
6 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
7 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
8 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
9 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
10 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
11 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
12 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
13 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
14 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
15 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
16 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
17 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
18 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
19 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
20 Ms::SpannerSegment::setVisible spanner.cpp 175 0x8409f3
...
Looks like an endless recursion? Not sure whether this is an entirely different bug, or a hint at what the corruption in this score is about?
Theres a recent code change that seems to trigger that endless recursion, in 4e0b0dba
In reply to The current nighly build, by Jojo-Schmitz
If it helps you any I'll tell you what I've done. I deleted everything except the first measure and changed the time to 4/4 (which is what I needed for my next project). When I got to the point where the time signature changed happened I had the same problem. I then remade my instrument list the same from scratch, copied the existing score to the new file and there is no problem so far. Apparently the corruption happens while I'm entering the score and it shows itself through this anomaly. The corruption is not within the data of a singe measure, or deleting all of the measures would fix it. Such as when you get the message that measure 27 is corrupted and the measure contains 15/16ths of a measures data for some reason.
One other thought, the problem did not occur until there were about 300-350 measures in this file. So there were a large number of normal time changes until that point, then all time changes after that presented the same problem. Other files I've had this problem with are far smaller than this. I don't remember having it any of the other song's I've transcribed, and many of those had multiple time changes in them.
If it helps, I'm running this under Windows 10 version 2.0.3 revision 3c7a69d
I don't recall if I had this problem with earlier versions but I don't think I did.
In reply to If it helps you any I'll tell by mike320
Thanks for the info. I think you are right that the corruption is not tied to any single measure. My guess is that it has to do with the "time signature map" that MuseScore constructs internally to track how many beats should be in each measure, and somehow there is a bogus entry in there. Something like that. I am not set up right now to be able to debug using 2.0.3, but the problem is different using the development builds for 3.0, so that's why it's hard to say more.
In reply to The current nighly build, by Jojo-Schmitz
See #142141: Crash when trying to open file with tied unison notes in different voices, same endless recursion
With that being fixed now, maybe we can get to the bottom of this corruption?