score from 53d4ad2 plays correctly but displays bizarre in newest nightly
I entered a score yesterday in the nightly build 53d4ad2, and today in the newest build
it displays really wierd. But it seems to play correctly.
MuseScoreNightly-2017-01-15-0831-master-eb4455c.dmg
Is there a way to fix this, or is my work just lost?
I notice it displays much better in 2.0.x but I see all articulations and ornaments
are displayed as a fermata sign.
Attachment | Size |
---|---|
Screen Shot 2017-01-15 at 12.56.45.png | 104.37 KB |
Screen Shot 2017-01-15 at 13.00.51.png | 122.75 KB |
Comments
The recent nightlies from master are pretty unusable in the Moment (those for 2.1 should be rather stable though). See for example the lack of time sigs (and see #159366: Invisible time- and key signature.
And for real work you shouln't be using those nightlies anyway, they are for testing only.
They are intended for testers and advanced users who want to experiment with new features, or need access to the latest bug fixes and are willing to risk the instability of an unfinished product. A file saved from a nightly build may not open in a future nightly or any released version. Use at your own risk.
Your work is probably not enrtirely lost, just foxe those femata vs. articulations and ornaments glitches
In reply to The recent nightlies from by Jojo-Schmitz
> They are intended for testers and advanced users who want to experiment with new features,
Does that include me if I want to use the experimental grace notes and articulation features?
In reply to > They are intended for by jim.newton.562
If you are a tester and advanced user who wants to experiment with new Features... and are willing to risk the instability of an unfinished product, then yes ;-)
And of course your bug reports are welcome.
In reply to > They are intended for by jim.newton.562
As long as by "want to use..." you realize this means, "and fully expect the scores I created today to be work I can never again open in future builds at some point" - there is no guarantee of future compatibility for scores created in nightly builds, and in fact one can bet pretty safely *against* it at this point.
I tried to download yesterday's and Jan 5's nightly to see if they can read the score,
they both seg fault on startup, with or without the -F flag.
Is that normal?
In reply to I tried to download by jim.newton.562
Yes, it is (on a Mac, that is). See eb4455c, done exactly for that reason, also see #152151: [Mac] Nightlies crash on start
In reply to Yes, it is (on a Mac, that by Jojo-Schmitz
good to know. How far do I have to go back to be able to open in? Or should I just keep trying earlier and earlier until it works?
In reply to good to know. How far do I by jim.newton.562
The one before 0b1aea9 should do
In reply to The one before 0b1aea9 should by Jojo-Schmitz
Can I download the nightly 53d4ad2 again?
Where can I find the link?
In reply to Can I download the nightly by jim.newton.562
http://prereleases.musescore.org/macosx/nightly/
No 53d4ad2 though anymore, that is 25 days old. We can't keep more than a certain amount without running out of quotas
In reply to http://prereleases.musescore. by Jojo-Schmitz
drat! OK, i'll see if I can manage to get it out of my backups. Thanks.
When I install the new nightly, it asks me whether I should keep both or replace the old one.
I never keep the old one. Perhaps I should start keeping it until I know I'm happy with the new one :-(
Live and learn.
Looks like the xml tags have changed.
I see these errors when reading into an earlier nightly
In reply to Looks like the xml tags have by jim.newton.562
Is it really a good idea to change the xml tags? What's the plan for loading scores from 2.0 where the tags are "trill" and "staccato" ?
In reply to Is it really a good idea to by jim.newton.562
There have been some additions lately.
2.0 scores should load just fine in 3.0 (if not that'd be a major Regression and we definitly want to know about it), but not vice versa (and by design), same for scores created in newer 3.0 build, when opening on older builds. Or even newer builds, while 3.0 is in development there's no garantee of any sort that any score created in one version will open in any other version, not even in the same at times
In reply to There have been some by Jojo-Schmitz
Yes I understand the guarantee, or lack thereof.
Knowing the differences, I can edit the xml file with emacs and load it.
especially if it is just a search and replace.
In reply to Yes I understand the by jim.newton.562
Urgs... better use vi(m) ;-)
BTW, check https://github.com/musescore/MuseScore/blob/master/libmscore/read206.cp… , here's the mapping of 2.0's "trill" to 3.0's "ornamentTrill"
In reply to Urgs... better use vi(m) by Jojo-Schmitz
Thanks Jojo, that link does not contain the string "ornamentTrill" as a string, but it contains
SymId::ornamentTrill, as a C++ identifier. Is this the change that started writing "ornamentTrill" into the xml file?
In reply to Thanks Jojo, that link does by jim.newton.562
That is somewhere else, this here is the 2.x comaptibility code, used when reading 2.x scores
Great, I retrieve the version from my local backup.