Continue of line begins at wrong place

• Mar 11, 2015 - 22:43
Type
Functional
Frequency
Few
Severity
S4 - Minor
Status
active
Regression
No
Workaround
No
Project

1. Open attached score (produced in 1.3).

(Additional steps:)
2. Click ottava.
3. 'Layout'>'Reset'.

Expected Result (taken from LilyPond 2.18.2): The continuing ottava begins at the first note of the line.
Continue of line begins at wrong place - Expected result.png

Actual result: The continuing ottava begins before the first note (after clef?).
Continue of line begins at wrong place - Actual result.png

Note: It is also reproducible in a 2.0 score.

Using MuseScore 2.0 Nightly Build d4e22ed - Mac 10.7.5.


Comments

At some point a few months back, the rule for how lines are continued was changed from starting even with with first note to starting just to the right of the clef/key/time signature. I think in the beta you are seeing the "before", which isn't really correct, but happens to look better in this case. The effect you are seeing here is because the key signatures have different widths; probably the code should start after the widest key signature, not the current key signature. Shouldn't be an issue in the parts;

I saw it in a score with a baritone saxophone and piano, and in preparing the report, I added it so you could see the problem more clearly, but didn't properly think through the possibility of it being the actual problem.

Additionally: Removing the baritone saxophone seems to position the ottava at the start of the clef:
Continue of line begins at wrong place 2.png

Using MuseScore 2.0 Nightly Build d4e22ed - Mac 10.7.5.

FWIW, this is kind of wrapped in in #50316: Concert Pitch modifies layout and forgets key signature in one line, actually, in that it has to do with interactions betwene key signatures on different staves. I can fix this but it won't look good with hide empty staves until that other issue is fixed. Here's the branch, but there is no point submitting a PR right now, I don't think:

https://github.com/MarcSabatella/MuseScore/tree/50391-continue-line

There is, I suppose, some small chance of the other issue being fixed for 2.0, but this one is not really worth it unless for reason we end up doing more bug fixes.