Reading 2.x score ignores all text style offsets
Reported version
3.3
Type
Functional
Frequency
Once
Severity
S2 - Critical
Reproducibility
Always
Status
active
Regression
Yes
Workaround
No
Project
Load any score created with MS ver. 2.x and check the various elements of the "Format | Style... | Text styles" list: none has any offset value from the score.
I am attaching a 1-measure score created with ver. 2.3.2 in which each text style has been set to a rather unlikely (1.23, 1.23) offset.
If read with 3.x, none of the text styles has such an offset. Some have some offsets, which are presumably supplied arbitrarily by the programme.
Set to critical as it makes any work continuity with ver 2.x almost impossible.
| Attachment | Size |
|---|---|
| TestOffsetsTo3.mscx | 12.73 KB |
Comments
FWIW, the code is there in read206 and seems to be getting hit, somehow it's getting cleared later, maybe during the reset stage?
See also #286129: (Some?) text styles imported with wrong (Y?) offsets from 2.x scores?
In reply to FWIW, the code is there in… by Marc Sabatella
The code was there, but it has been 'if-zeroed' out: (https://github.com/musescore/MuseScore/blob/master/libmscore/read206.cp…)
Ah, I saw the code above that where the tags themselves ("xoffset" and "yoffset") are processed, and I verified that code is hit. But this is indeed where it falls apart. Looks like that change was made during the beta period.
Maybe I'm missing something, but I don't think the series of tests for different styles should be needed any more. FWIW, I think all that code can probably be replaced by just the "case Pid::OFSET: value = offset; break;"
Seemed to work when I just tried it, although I didn't test very exhaustively.
@Jojo-Schmitz : I know this issue came after #286129: (Some?) text styles imported with wrong (Y?) offsets from 2.x scores, but I think this is more documented and clear, so I believe the other issue should be marked as duplicate of this one rather than vice versa (just my opinion, though).