Changing time signature removes repeats and voltas

• Jul 4, 2012 - 11:52
S2 - Critical

1. Load attached file (originated from abc>XML conversion, also attached)
2. Drag Common time to first bar from palette.

Expected behaviour: Common time symbol is inserted for time signature leaving score otherwise unaltered.
Actual beahviour: Common time symbol is inserted but all repeat signs and voltas are removed.

Apparantly the original abc>XML conversion didn't add a time signature although MuseScore appears to know it is in 4/4 - don't know if this is relevant

Marked critical as it results in loss of data

Rd687e22/Windows XP Pro SP3

Incidentally - Apache is still under copyright, so please do not distribute this file outside the development/testing team.

Attachment Size
Apache.mscz 8.2 KB
Apache.xml 47.89 KB


Do you still have access to the ABC file you started with? The XML file posted yields an error on load and I'd like to know why before proceeding.

In the current builds, the 4/4 time signature is displayed in your load the XML file posted. The MSCZ file appears to no longer be valid as it displays poorly.

But I am working on addressing issues with time signature changes that may improve the situation here.

I'm not able to reproduce any problem in any score I create starting from scratch. This includes a MusicXML file I generated from an ABC file with "M:none", which I guess is what you probably had.

Not sure if my inability to reproduce has to do with any fixes that were made or not. I *did* fix some things, and I believe others have too).

I do see the repeats being lost using "apache.mscz", but that file doens't resemble "apache.xml" at all to me, so I'm not sure how it got into the state it is in. Even though apache.xml gives an error on load, it *does* load eventually, and it shows with a time signature.

Apache.xml contains many errors where XML elements are in the wrong order (which causes validation errors), but the data itself seems OK. I tried to fix it manually, but gave up, as the number of errors is simply to high.

It includes a 4/4 time signature.

As this issue is relatively old and there has been no activity for over a year, is this still relevant or can it be closed ?

Critical issues shouldn't stay open for 3.5 years.

I have not been able to reproduce it on my 9c194d5 branch (master+14884

However, adding/removing 4/4 common time in the middle of the first repeat in this mscz example has weird effects:
- the current repeat is removed once the 4/4 common time is removed
- once re- insert on the 6th measure (first after line return) , the 4/4 common time
is display at the end of the previous line after the closing measure boundery of the 5th measure.