Musescore file is huge due to one measure
Reported version
2.1
Priority
P2 - Medium
Type
Functional
Severity
S3 - Major
Status
active
Regression
No
Workaround
No
Project
This musescore file is huge (776kb) even though the score contains only four measures. The problem is due to an accent in measure 3. This significantly slows down the program. Is there any way to quickly identify problematic measures like this one that vastly increase file size? I believe this was originally caused by an import through music XML
| Attachment | Size |
|---|---|
| SO HUGE.mscz | 775.78 KB |
Comments
Is there some relation to the problem described in https://musescore.org/en/node/138061, could it be orphaned slurs causing this?
This is related in a way to https://musescore.org/en/node/138061, instead of slurs,
this involves hairpins. There are 65K hairpins defined in measure 3.
I suspect this is just a different manifestation of the slur model.
--Chuck
Hairpins that start but don't end, i.e. don't have an end tag, by any chance? Both are spanners, so that's the common denominator
Hmm, no, looks as if all hairpins do have start and end, and they are all visible and on top of one another (except for 2 longer hairpins). So this seems to be a different issue from those slurs that lost their end tag.
But the file does stem from an XML Import. @daloonik: would you be able to dig up and share the XML source for this?
Raising to Major as it renders MuseScore almost unresponsive and next to impossible to work with, seems to be some kind of corruption.
Need a better title...
Do you have Sibelius ? I have the original Sibelius file, created with Sibelius 2 which I used a long time ago, and used Sibelius 7 to export to Music XML temporarily which I then used to edit in Musescore. However I don't think I'm allowed to upload sibelius files here.
I think I have a small file to reproduce. (I cut out the huge file from previous comments)
It could be solved by not saving spanner with ticks=0.
here? https://github.com/musescore/MuseScore/blob/b8fe9e63328098fbd71645c48e2…
There have been situations where I've taken advantage of - or at least considered doing so - slurs from a note to itself to mimic a one-sided tie. Currently I'm not sure how to create these, but there used to be ways. Not that this is a good reason why we should support zero-length spanners, particularly if they don't work, but there *could* be scores out there actually depending on them.
I usually do this by making one end of the slur a grace note, then hiding the note, but not the slur. I use the technique less for things such as percussion and more for showing that a second ending in a volta ties into the preceding measure.