[trunk] Wrong staff/coords in symbol placement with multiple staves (regression)

• Sep 23, 2010 - 20:00
Type
Functional
Severity
S4 - Minor
Status
closed
Project

Checked in rev. 3499 under both Windows XP / Qt 4.7.0 and Ubuntu 10.04 / Qt 4.6.2.

When opening a score with some symbols added from the symbol palette (Z), symbols are placed much far right than expected (see .PNG image attached).

A small score with a few notes and a (silly!) symbol is attached for testing.

Thanks,

M.

Attachment Size
Offsets.mscx 4.69 KB
Offsets.png 8.64 KB

Comments

Title [trunk] Wrong X coord in symbol placement [trunk] Wrong staff/coords in symbol placement with multiple staves (regression)
Status (old) fixed active

Oh damn! I was very happy this bug had been fixed, but things are a bit more complicated when there is more than one staff only.

Setup: trunk rev. 3527, Windows XP w/ SP3, Qt SDK 2010.04 (Qt Creator 2.0.0, Qt lib 4.7.0)

Steps: Load the attached score (symbols ('C' and '3') are misaligned on purpose to have them 'show through' in trunk screen shot).

Result: Symbols on staves after the first are 'combined' all in the first staff (and displaced to the left).

Expected result: Symbols remain in their respective staves at their X and Y coords, as in 0.9.6.x. A partial screen shot with 0.9.6.2 and with trunk rev. 3527 is attached.

Thanks,

M.

Attachment Size
TestSymbols.png 25.03 KB
TestSymbols.mscx 15.63 KB
Status (old) active fixed

fixed in r3534. The horizontal position changed bc. the layout is different (measure stretch).

Great, thanks!

I do not understand your comment about different layout, though: it is the same score, how can the layout be different? Measure stretch should be in the score, isn't it? and then be the same in both version.

Do 0.9.6.x and trunk have some differences in (some) defaults, not covered by the styles explicitly set in the score?

M.

I did not investigate why the trunk layout is different from 0.9.6 for your test file. But the layout may change in future versions due to optimizations in the used algorithm and bug fixing (like #7303).