Don't eliminate layout elements upon some cases of time signature change
Type
Functional
Frequency
Many
Severity
S5 - Suggestion
Status
active
Regression
No
Workaround
Yes
Project
See the score attached to http://musescore.org/en/node/34041#comment-144441, delete the time sig in measure 1.
All system breaks up to the next timesig get deleted
1fb21eb, Windows 7 Enterprise, 64bit
Comments
Not just deleting time signatures - any time signature change. This is "by design" - there is no guarantee that barlines will still be in the same places, so breaks attached to them may no longer be meaningful.
That doesn't mean we couldn't try to make guesses. See http://musescore.org/en/node/31851.
OK, what is the timesig once it is deleted from score? It changes ro 4/4. If it was 4/4 before, or any 'compatible' version, common or cut time, there is no need at all to delete the system breaks. Same for changes between these 'compatible' (mathmatical identical) types, not much reason for shorter types, only longer ones may not fit anymore
Yes, I'm sure it would be possible to detect changes that are not actually significant and then avoid calling the rewrite code (which is where the breaks are removed, I believe).
For most people, though, I think it's going to be a bigger issue that time signature *changes* delete breaks than the fact that time signature *non-changes* do. So if we're to come up with special case code, I'd rather it address the problems more people are likely to see.
Hmm, wierd (actually good), changing cut time to common time or 4/4 or the other way round doesn't delete the system breaks
How much you wanna bet that's related to the fact that cut time doesn't register properly in Measure Properties?
possible
Ir should not do this if the time signatures are mathmatically identical, i.e. don't lead to different measure boundaries. Cut time is 2/2, common time is 4/4, so there's no need at all to loose the system breaks, as all measures stay the same length.
To my understanding, system break are always at the and of a measure, am I correct?
In that case Ir should not do this if the time signatures are mathmatically identical a solution could be: Remove system break which are not at the end of a measure anymore (or even move them the the end of a measure, although I'm not sure that's a good thing to do).
That'd push it even further than just leave them at mathmatical identcal time sigs, like a system with 3 measures of 4/4 are the same as 4 measures of 3/4. May or may not fit the same page width.
In reply to That'd push it even further… by Jojo-Schmitz
But makes it easier to implement :-). For each system break, just test whether it coincides with a barline. If not, remove it.
In reply to But makes it easier to… by njvdberg
If the user really wants to delete these breaks, he can do it quite easily (select all similar blablabla...)
If he does NOT want to delete them and MuseScore did it for him, there is no way to make them come back other than reputting them one by one.
So, in case of hesitation (should MuseScore delete or not), I would rather say not.
OK by me
Came up in #51176: Change the time signature remove the score elements, #123886: 4/4 and cut time not considered equivalent and https://musescore.org/en/node/303220
See also #287071: Changing Time Signatures removes symbols from barlines
In cases where the duration of the measures isn't actually changing (i.e. 2/2 to 4/4, 3/4 to 6/8), a workaround is to select the time signature at the left of the staff, right-click (or two-finger tap, or whatever input brings up your contextual menu) —> Time Signature Properties, then change the "Text" setting in the Appearance area at the upper right of the dialogue box that appears. Don't change anything else. That should change the time signature marking to whatever you want it to be without disrupting any of your formatting. You'll have to do this individually for each staff, because it's purely cosmetic and staff-specific.
If you want to insert a change from one compatible time signature to another, just drop an extra instance of the current time signature where you want the change to be, then (cosmetically) alter that instance.
Closely related but not quite the same: if you want to change a time signature for a certain passage, but don't want the change to disrupt formatting beyond a certain point, create a new instance of the current time signature at the point beyond which you don't want any disruption to occur. Do this before you insert the new/different time signature. Then insert the new/different time signature. This way notes won't get pushed along and split by barlines etc. Instead, Muse Score will create an extra measure with some rests in it as a buffer, re-syncing the barlines to the later, unchanged content.
It seems like this issue has come up in several threads. I don't want to try to hunt them all down, but I hope this solution propagates to them, or they end up leading here.
In reply to In cases where the duration… by davidevankrebs
@davidevankrebs
Thanks for clearly illustrating so many scenarios and options for dealing with them.
As of today What Links include this related link:
#328098: Linked tablature staff doesn't show Treble clef's beaming pattern / And its beam pattern cannot be set with steps that worked for the treble clef.
But the salient point there, related to this issue, is in this reply regarding a meter change deletion of 4/4 (where that measure was surrounded by 4/4, fore and aft) yet MuseScore needlessly tossed all line breaks and more.
https://musescore.org/en/node/328098#comment-1111871
scorster
In reply to But makes it easier to… by njvdberg
njvdberg wrote > > For each system break, just test whether it coincides with a barline. If not, remove it.
Makes sense to me.
And as mentioned int recent post, in such cases, when the user see one or more undesirable affects on invoking a change, like a meter change, I think there's merit in affording the user a means retry with an intervening dialog that shows pending property changes with the option to toggle some of them to the preferred outcome.
scorster