Updates:
Status: Verified
Comment #26 on issue 1031 by v.villenave: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Hi guys,
I'm closing this issue now; I'm not sure exactly what this cyclic
dependency was
Comment #24 on issue 1031 by joeneeman: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Does it help if you change rcol in rest-collision.cc:214 to rest?
___
bug-lilypond
Comment #25 on issue 1031 by n.puttock: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Yes, thanks.
It's confusingly named, don't you think? I originally read it as short
for rest collision, but it's actually rest
Comment #19 on issue 1031 by n.puttock: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Joe, have you seen what this does to lyric-combine-switch-voice.ly?
programming error: cyclic dependency: calculation-in-progress
Comment #20 on issue 1031 by joeneeman: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
hm, I can't reproduce it. Do you have a command line that gets the problem
consistently? If not, could you get me a gdb backtrace
Comment #21 on issue 1031 by n.puttock: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
It happens consistently via the command line or running inside Frescobaldi
(naturally, with --disable-optimising for the binary.)
Comment #22 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
I am getting the same results as Neil, running with
$ lilypond lyric-combine-switch-voice.ly
Attached is the backtrace I am
Comment #23 on issue 1031 by n.puttock: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Cheers, Patrick. Yours looks the same as mine. :)
___
bug-lilypond mailing list
Updates:
Status: Fixed
Labels: fixed_2_13_25
Comment #18 on issue 1031 by joeneeman: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
(No comment was entered for this change.)
Updates:
Owner: joeneeman
Comment #17 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
(No comment was entered for this change.)
___
Updates:
Labels: Patch
Comment #15 on issue 1031 by joeneeman: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
http://codereview.appspot.com/1669042/show
Thanks to Patrick's investigations, here is a patch that
Comment #16 on issue 1031 by joeneeman: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Oops, try this instead:
http://codereview.appspot.com/1694041
___
bug-lilypond
Comment #13 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
New info:
If I remove these two lines from rest-collision.cc, the bug seems to go
away.
67 if (scm_is_number (offset))
68
Comment #14 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
After checking the regtests, my suggested change from comment 13 creates
other
unwanted regressions.
I'm very close to figuring
Updates:
Status: Started
Owner: pnorcks
Comment #11 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Update:
I have discovered the culprit function call. See below. The result
Comment #12 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
I also wanted to mention that I *do* have a reproducible test case,
discovered
through a guess-and-check approach, but it might not
Comment #7 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Thank you for the helpful suggestions, Han-Wen.
Your intuition is correct: the problem is with the sorting vectors of
pointers to
Comment #8 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Hmm... in this case, I guess we are not actually sorting by pointer
address, so the
problem might be elsewhere.
Comment #9 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
(Ignore my last comment.)
There are certain cases where pointer addresses are compared in
staff_priority_less.
If I comment out
Comment #10 on issue 1031 by hanwenn: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
We should try to remove all of these comparisons for the sake of
reproducibility. It
does not fix the actual problem though, since
Comment #6 on issue 1031 by hanwenn: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
have you tried running it through valgrind? It might turn up a subtly
uninitialized
variable issue.
Another thing to look for is
Comment #4 on issue 1031 by pnorcks: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
IIRC, in the past I've seen this happen with
input/regression/dot-rest-beam-trigger.ly too.
The latest example I can find is the
Comment #3 on issue 1031 by percival.music.ca: constantly-changing
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031
Issue 1047 has been merged into this issue.
--
You received this message because you are listed in the owner
or CC fields
Updates:
Summary: constantly-changing
input/regression/rest-collision-beam-note.ly
Labels: -Priority-High Priority-Critical
Comment #2 on issue 1031 by percival.music.ca: constantly-changing
input/regression/rest-collision-beam-note.ly
24 matches
Mail list logo
