Join measures should inherit breaks and takes properties from last measure
Reported version
3.5
Type
Functional
Frequency
Once
Severity
S5 - Suggestion
Reproducibility
Always
Status
active
Regression
No
Workaround
No
Project
If two measures are joined when the last measure has a system break, this is lost when the measures are joined. One would expect such properties to be kept. The same goes for things like double barlines, etc.
Also, which is more serious: if two measures that are followed by a change in time signature in the following measure, the joined measure gets the time signature of the following measure, it does not keep the current time signature, as one would expect.
Comments
I believe this is actually the first measure that gets made longer and the notes of the 2nd moved into it, then the 2nd getting deleted, along with the system break. So this might be by (current) design.
That is possibly why any modifications to the second measure gets lost, such as system break and barlines, but it doesn't explain why the new combined measure gets the time signature of the following measure, and not of the first.
In fact, modifications (breaks, barlines) to the first measure are also lost.
The current behavior seems to be to treat this as a new measure. And it's unclear which properties really should be preserved when they are not the same. Probably breaks and barlines from the second, but other properties from the first? What about joins of more than two measures
Chances are whatever we did would be right sometimes, wrong other times, but I'd be fine with preserving breaks and barlines from the last and doing nothing else.
In reply to The current behavior seems… by Marc Sabatella
If one regards a block of measures marked for joining as equivalent to a single measure - which is what is intended, after all - would it be reasonable to preserve any in-going properties at the start of the measure, and any outgoing properties at the border to the next - as it is before the joining? Any changes in intermediary measures would not have any effect in any case. E.g., if there are measures included with a different time-signature, those changes would be eradicated in any case.
In reply to The current behavior seems… by Marc Sabatella
So in addition to preserving barlines and breaks, time signatures at the end of the block should be preserved.