Sanity Check
Regarding the new "Sanity check" test on score loading in 2.0b2 (git 53fc23e), I've only come across two scores (so far) that fail on the final measure. These are both sample scores I wrote to test features (so don't criticize the musicality of them), and see how 2.0 imports more complex 1.x scores. V1 was written in 1.1 (rev 4801), and V2 was written in a very early nightly of 2.0 (rev 4814), likely almost 2 years ago.
The error for the V1 file says
Measure 33 Staff 1 incomplete. Expected: 2160; Found: 2400
Measure 33 Staff 2 incomplete. Expected: 2160; Found: 2400
and for the V2 file says basically the same thing. The final measures do add up though!
The V1 score also shows a bug in the placement of the volta lines. On page 1, measure 9, the second volta is over too far right, and a reset won't move it. On page 2, the first volta ending (1., 2.) should be from 18-19, and the second (3.) should only be on measure 20. Unless someone sees some corruption or a reason not to, I will post these in a bug report regarding the faulty volta placement.
| Attachment | Size |
|---|---|
| Test Piano Score V1.mscz | 7.64 KB |
| Test Piano Score v2.mscz | 5.5 KB |
Comments
V1 doesn't sum up, it should end with a half and an 8th rest, but ends with a dotted half rest (half and quarter), so is one 8th too long.
V2, the 2nd to last measure is nominal 6/8, actual 1/1, the last is 6/8 both.
Selecting the last measure and hitting 'Del' replaces the hald + quarter rest by a full measure rests, the result then loads cleanly.
It looks like you had manually adjusted the volta in 1.3 to start after the clef & keysig. In 2.0 that happens automatically, so your manual adjustment went too far. While Ctrl+R doesn't reset that, the Inspector does.
In reply to It looks like you had by Marc Sabatella
Shouldn't Ctrl-R reset the same way? I would consider this a bug.
The second set of voltas are a whole different problem. I can't reproduce the issue, so there must be some subtle corruption of the start/end points.
Twill be interesting to go over _all_ my 1.x scores, to make sure things are the way I expect, and to "fix" my workarounds and manual adjustments so 2.0 can auto-layout better.
In reply to Shouldn't Ctrl-R reset the by schepers
Currently, Ctrl+R resets certain types of manual adjustments but not otherwise. I am not actually sure what the distinction is or is supposed to be. It does seem like this should work; you may wish to consider filing this to the issue tracker.
In reply to Shouldn't Ctrl-R reset the by schepers
As for the second set of voltas, I don't see the cause, but looking at the Object Debugger, I can tell you the first one claims to start with an element at beat 54, the second at beat 59. They should have been 8 beats apart. I don't know if something went wrong during the read, or if the file itself was corrupt but 1.3 happens to render it differently than 2.0.