Boy, am I ashamed now.
I bug#578, I reported a regression in the way lyrics were handled when
attached to music typeset using the Completion_heads_engraver. There
was indeed an ugly regression, but my description of the way the
lyrics *should* be typeset was wrong.
No regression this time, just
>8 cut here -- I'm not top posting...
\version "2.11.45"
%% -dpreview should only include first system, not post-music markup
#(ly:set-option 'preview)
\score {
\context Score <<
\relative c'' {
c-\markup "this line is in the preview" c c c |\break
c-\markup "this line should
[EMAIL PROTECTED] (Arvid Grøtting) writes:
> % Bug exists in both 2.11.39 and 2.10.33
> % This is a regression; it worked as it should in 2.8.8 and in 2.9.27.
2.11.11 also worked; 2.11.12 is broken (in this regard).
-- Arvid
___
bug-li
> I'm not top-posting.
% \version "2.11.39"
% lyrics are misaligned when using completion heads.
% They should be in the same place below both staves.
% Bug exists in both 2.11.39 and 2.10.33
% This is a regression; it worked as it should in 2.8.8 and in 2.9.27.
mel = \relative c'' {
c2 c1 c1
Han-Wen Nienhuys lilypond.org> writes:
> I have fixed this bug (I think), and am rolling a .25 to amend
> this brown-paper-bag bug
Much better. :-)
Now, I'm still wondering why "system-count = 9" gives 3 pages while specifying
the same breaks manually and not setting system-count gives 4 pag
Arvid Grøtting gmail.com> writes:
> I'll try cooking up a minimal example.
No luck there, I'm afraid. It looks like I've stumbled upon a
combination of lyrics and music that trigger a new bug. This
seems spacing-related, and removing stuff makes the bug
go into h
quot;;
\fill-line {
\smaller \forlagsinfo
\smaller \printismn
}
}
dnslogo = \markup{\epsfile #0 #10 #"../include/dns-logo.eps"}
\header {
copyright = \dnscopyright
}
\paper {
%% tagline = \markup\fill-line{\smaller \smaller \tagline
%%\smaller \printism
[Joe Neeman]
With the optimisations, it scales like (left column is the number of scores)
24 2m0.017s
19 1m23.650s
14 0m52.611s
9 0m27.961s
4 0m11.139s
1 0m4.402s
as you can see, it isn't O(n) but it isn't much worse either. The further
optimisations I have planned should make it even closer to
2007/5/2, Joe Neeman <[EMAIL PROTECTED]>:
With my working copy of lilypond (on a 1.6 GHz Opteron), I have this down to
3m22.321s. That's still about 3 times slower than 2.8 but it's progress.
FWIW, I don't think 2.10 (and above) will ever be as fast as 2.8 when there
are many scores -- the newer
Joe Neeman gmail.com> writes:
>
> I recently discovered that the new page-breaking algorithm is slow when there
> are many small scores.
FWIW, lilypond finished processing this score in 26m37.359s on a (fairly new)
laptop I use, using version 2.11.23 on Linux (64-bit Ubuntu 7.04).
The "conver
David Feuer gmail.com> writes:
> % If this is changed to \addlyrics { one __ two } then the extender is short.
Workaround:
\addlyrics { one __ "" }
(Untested as such, but I've used this construct in real life quite recently.)
--
Arvid
__
2007/2/28, Graham Percival <[EMAIL PROTECTED]>:
I can't reproduce this (even after uncommenting that line) with 2.11.20
on OSX. Could you check that you attached the right file?
I did. Are you sure you're not seeing a difference? I've attached
PNGs of the two different results I get...
I k
Arvid Grøtting gmail.com> writes:
> I haven't cooked up a minimal example yet, but is vertical stretching supposed
> to disappear if you set system-count to a value in the \paper block?
\version "2.11.20"
\paper {
%% uncomment next line, and stretching disappear
I haven't cooked up a minimal example yet, but is vertical stretching supposed
to disappear if you set system-count to a value in the \paper block?
--
Arvid
___
bug-lilypond mailing list
bug-lilypond@gnu.org
http://lists.gnu.org/mailman/listinfo/bu
Joe Neeman gmail.com> writes:
> No, it's still a bug. The fact that there's a workaround just means that we
> can give it a lower priority.
OK, as long as it stays manually overridable.
In some (eh, make that "many") songs, all (or the middle) voices sing the same
lyrics, and in that case I'll
Joe Neeman gmail.com> writes:
> This is easy to workaround, but I don't know how to make lilypond detect it
> by
> itself. [...]
Ah, OK. I'll update my templates. Consider the bug report retracted. ;-)
I might even clean up my TTBB template once I get this in and post it somewhere.
Cheer
>:-)
Hi,
let me first say that 2.11.20 does seem to resolve most of the bugs in the
new vertical spacing engine. Right now I only have one small nit to pick:
Lyrics contexts aligned above a given staff should stay close to that
staff when stretching. That is, in the case below, you should have:
Hi,
if a bug has been too hastily marked as a duplicate in the bug database, how
does it get its status changed?
Do I re-report the bug? Nag someone in private email? Give up on that bug?
Give up on bug reporting altogether?
I'd suggest commenting on the status change here, but someone alread
>> [This line intentionally not left quite blank]
As of 2.11.17, any kind of grace note cancels \voiceOne etc. from that point
onwards, even across voices or (not shown here) staffs.
This works fine in 2.10, and I think it worked in 2.11.5 as well. It definitely
failed in 2.11.16; haven't tried
>> (Please disregard this line. No, Gmane; I'm not top-posting.)
\version "2.11.5"
\paper {
ragged-last-bottom = ##f
texidoc = "
Two-pass page layout sometimes stretches too much.
When run without a tweak file, three pages of 3 + 3 + 2 systems are created.
After two-pass space tweaking, I g
Han-Wen Nienhuys lilypond.org> writes:
> Arvid Grøtting escreveu:
>
> > I don't see the undertie touching the vowels, however. What's going on?
> > Are we all getting different results here?
>
> The code checks for the width of the char. Maybe you can che
Eduardo Vieira click21.com.br> writes:
> Well, I think the lyric tie shouldn't touch the vowel, too. That is it would
> be
> good if the tip of the tie would be lowered a tid bit.
> It looks like that the example you provided shows that the tie is shifted more
> to the right in one case and more
Han-Wen Nienhuys lilypond.org> writes:
> The tie has a negative space, so it is starts below the syllables
> This was by request of Nancho, and he had scans to confirm that this is
> typesetting practice.
I agree that the tie should start -- and end -- below the tied
syllables, but there shoul
\version "2.10.0"
\header {
title = "Spacing with lyric tie"
texidoc = "There should be some space when there's a lyric tie.
A hair space or thin space on either side of the undertie is good
workaround. A regular space on either side also works, with one
side giving the same output as in 2.8
Joe Neeman gmail.com> writes:
> > What will happen if you keep the current behaviour of the
> > line-breaker, but apply it to each section after splitting on forced
> > page breaks instead of to the whole \book? (My guess is we'll get the
> > expected behaviour, quite possibly with a speedup as
Joe Neeman gmail.com> writes:
> This is because the page breaking algorithm tries to ensure that adjacent
> pages have similar vertical spacing (removing this code gives the results
> you expected). I'll see what happens if I just remove this behaviour (the
> line breaker does it, but the old pag
The spacing is much better now, but it still gets strange in some
not-too-contrived cases:
\version "2.9.19"
#(set-global-staff-size 18)
\header {
title = "breakbefore spacing test"
subtitle = "breakbefore = ##t still causes problems"
piece = "Test"
texidoc = "Score blocks seem to be ba
I'm trying to use bookTitleMarkup to produce front pages using
data from the header block. This works, kind of, but I've
stumbled upon a bug:
When I set breakbefore = ##t, the horisontal spacing of the score
gets very loose. Instead of getting one additional page, as
expected, a score may jump f
The following snippet:
\version "2.9.26"
\score {
\context Score <<
\new Staff <<
\new Voice \relative c'' { \voiceOne c c r c }
\new Voice \relative c'' { \voiceTwo c g r a } >> >>
\layout {}
}
...still produces the following bogus warnings as of 2.9.26:
simultan-rests
Might as well get A5 and A6 right while we're at it: Second try:
Index: paper.scm
===
RCS file: /sources/lilypond/lilypond/scm/paper.scm,v
retrieving revision 1.70
diff -u -r1.70 paper.scm
--- paper.scm 4 Oct 2006 10:51:57 -
> [big pages]
The following patch seems to help. Even if ISO216 has 2mm tolerances
in the paper size, something on my mac (Preview.app or the printing
system, I guess) doesn't.
It's 1 mm, and I'd love not to get that "crop or scale" question each
time I print... Any chance this could make it in
Marcus Macauley riseup.net> writes:
> So, Joe, whatever it is that you fixed, there still seems to be a problem
> in 2.9.22.
I think this was fixed *after* 2.9.22, so that makes sense. ;-)
--
Arvid,
waiting for 2.9.23 (and 2.10)
___
bug-lily
Arvid Grøtting gmail.com> writes:
>
> For some reason, 2.9.21 seems to give me slightly bigger pages
> than 2.8 did. This means I get asked if I want to crop or scale when
> printing.
Problem still present in 2.9.22.
--
Arvid
Arvid Grøtting gmail.com> writes:
> For some reasons, 2.9.21 has three systems per page on all pages,
> while 2.8.1 has three on the first page and four thereafter, but I
> don't know if that's a symptom or the problem.
Looks like it's related to the problem. Here
Hi!
I'm completing a score that I think shows a bit of regression with
newer 2.9 versions as opposed to 2.8. The problem is that the spacing
gets a bit weird, giving 7 pages with 2.9.21 when only 5 is needed
with 2.8.
The .ly file (and two include files) and resulting output can be
viewed at
For some reason, 2.9.21 seems to give me slightly bigger pages
than 2.8 did. This means I get asked if I want to crop or scale when printing.
\version "2.9.19"
#(set-default-paper-size "a4")
\score {
\context Score <<
\relative c'' { c c c c } >>
\layout {}
}
--
Arvid
_
Mats Bengtsson ee.kth.se> writes:
>
> I hope that you also have tried
> http://lists.gnu.org/archive/html/lilypond-user/2005-03/msg00035.html
That helps in the case where all the rests are staff-wide.
I'll guess an example says more than words:
\version "2.9.19"
\score {
\context Staff {
Mats Bengtsson ee.kth.se> writes:
> This is why many people try to use \partcombine for choral scores.
> However, as you can see from the mailing list archives, there are
> some problems with the current implementation of \partcombine.
Indeed there are. Partcombine may be useful for hymns -- I
Graham Percival gmail.com> writes:
> I'm working on a viola trio right now, and these warning messages
> completely conceal any other info. I shudder to think of all the
> messages if I were doing an orchestral score.
I'm typesetting some TTBB choral music now (as always), and I get
similar
Alan Shields alan.shields.name> writes:
>
> My apologies if this has previously been reported.
>
> Notice the bar numbers and their interaction with the staff grouping on
> bad versus good. This is true for both StaffGroup and ChoirGroup.
I think I've seen that before. No, I know I have. The
\version "2.8.1"
%% Here, the first two lyric extenders extend way too far. Visually,
%% it looks like the extender thinks the acciaccatura for the
%% following note is part of its syllable.
%% The third extender shows a workaround using a hidden note and a
%% blank syllable.
\score {
\relativ
Dynamic (e.g. \pp) markings aren't aligned properly
when stacking lots of markings and using, e.g.
\pp\> as well.
Using MacOS X 10.3 and Lilypond 2.8.1:
\version "2.8.1"
\score {
{
% fails to align correctly:
g'8\pp\>-.\( g'\!-.\)
% workaround:
<< g'8\pp-.\( { s16 s16\> } >> g
http://lilypond.org/doc/v2.8/input/test/collated-files.html
...contains a lot of links bearing the name of the snippets, leading to e.g.
http://lilypond.org/doc/v2.8/input/test/lily-1718542889
...where the code should probably be displayed. Instead, all these snippets
simply contain the followi
Mats Bengtsson <[EMAIL PROTECTED]> writes:
> However, I don't understand your problems with \skip, see the following
> version of your example:
I don't get \skip to play nicely with \lyricsto in 2.2.1, though. Is
this something that's been fixed later?
--
Arvid
44 matches
Mail list logo
