Originally created by: *anonymous
Originally created by: colingh...@gmail.com
Reported on lilypond-user by David Stocker here:
http://lists.gnu.org/archive/html/lilypond-user/2012-07/msg00093.html
and forwarded to bug-lilypond by Trevor Daniels here:
http://lists.gnu.org/archive/html/bug-lilypond/2012-07/msg00015.html
Original post follows:
--------------------------------
I've discovered a strange behavior in 2.15.41 when I alter the size of the staff space in the \layout block.
I've played around with the different settings and overrides I have in the layout block and the layout-set-staff-size and StaffSymbol #'thickness settings seem to be blameless. StaffSymbol #'staff-space is definitely the culprit.
The behavior doesn't occur when I compile the same score with 2.14
Have a look at the attached files.
Thanks,
David
--------------------------------
Originally posted by: k-ohara5...@oco.net
A shorter example:
{ \repeat volta 2 b1 }
\layout {
#(layout-set-staff-size 10.0)
\context {
\Score
\override StaffSymbol #'staff-space = #0.5
}}
Trevor traced the cause of trouble to the patch for issue 2533.
Originally posted by: tdanielsmusic
This bug first appeared in 2.15.40, so is a critical regression.
Trevor
Labels: -Type-Ugly Type-Critical
Originally posted by: tdanielsmusic
I'm not in a position to compile c++ at the moment, so I can't check this and may be completely wrong, but this line in lily.bar-line.cc, introduced by the fix to 2533, looks suspicious:
145 /*
146 gap is measured like line-positions;
147 1.0 for dot diameter, twice the staffline width for the
148 gap above and below and one more staffline width for the
149 two half stafflines
150 */
151 Real const gap_to_find = (1.0 + 3 * staffline) / staff_space;
gap_to_find gets larger as staff_space gets smaller; shouldn't it get smaller?
Maybe try:
151 Real const gap_to_find = (1.0 + 3 * staffline) * staff_space;
Trevor
Originally posted by: benko....@gmail.com
layout-set-staff-size is nasty: see the attached files.
rep-sign-1 demonstrates that the logic of line 151 is not bad.
quote from NR 4.2.2:
”Known issues and warnings
layout-set-staff-size does not change the distance between the staff lines.”
the paragraph above that suggests that the context property fontSize should somehow taken into account to make the logic of line 151 really good. what do you think? rep-size-2, I believe, demonstrates that it's not entirely ignored, as the dot placement in staves with staff-size 0.65 happens to be different (and right) in both cases, because staffline thickness _did_ change.
the real solution would perhaps be to use the actual dot size instead of the magic constant 1.0 in line 151.
Originally posted by: k-ohara5...@oco.net
> the context property fontSize should somehow taken into account
I agree, see my comment on Rietveld, but there might be surprises. Yes, layout-set-staff-size is nasty, but people have to use it, and it is tricky to see how this interacts with scaling.
Revert the appropriate amount of the patch for issue 2533 so you can carefully have dots avoid staff lines during 2.17 development.
Originally posted by: k-ohara5...@oco.net
Reverted the patch for issue 2533 for now, and probably for version 2.16
Labels: fixed_2_15_42
Status: Fixed
Originally posted by: benko....@gmail.com
issue 2533 has an improved fix, see
http://codereview.appspot.com/6431044/
Originally posted by: Elu...@gmail.com
(No comment was entered for this change.)
Status: Verified