LilyIssues / Issues / #2462 unnecessarily wide spacing caused by Lyrics

Google Importer - 2012-03-31

Originally posted by: PhilEHol...@googlemail.com

There's too much in Gould to type right now, but in summary she says that instrumental spacing is often impossible with vocal music because of the need to accommodate the words. In this case, don't space all the notes out in proportion to the newly streatched spacing, but do your best to fit them in sensibly.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-03-31

Originally posted by: janek.li...@gmail.com

Phil,
i don't quite understand what do you mean: is the current Lily spacing correct in your opinion?
For clarity, here's roughly how these examples should be spaced in my opinion:
http://lilypond-stuff.1065243.n5.nabble.com/examples-for-Lyric-report-spacing-bug-td5608830.html

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-03-31

Originally posted by: PhilEHol...@googlemail.com

They look OK to me. However, I think I'd try this out with more than one line of music - I suspect the right-flush/ragged-right may be spacing the notes and lyrics differently from how it would be spaced in the middle of the music.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-03-31

Originally posted by: janek.li...@gmail.com

Hmm? The real-life examples come from middle of multi-system pieces, see here:
http://lilypond-stuff.1065243.n5.nabble.com/examples-for-Lyric-report-spacing-bug-td5608830.html#a5608970
Try compiling these without lyrics (in Gabriel's Oboe: edit lines 65-72 of choir.ly) or with ragged-right on.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-03-31

Originally posted by: PhilEHol...@googlemail.com

I was thinking about the la la straight straight example. That illustrates the problem quite well, but as shown looks like less than a real full line of music.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-03-31

Originally posted by: janek.li...@gmail.com

The point is that a "real full line of music" is quite often less than a full line. In other words, what constitutes a full line is a result of stretching something that would naturally be spaced tighter. The measures don't fit line width perfectly; usually some gap is left which is eliminated by stretching the line.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-03-31

Originally posted by: k-ohara5...@oco.net

Why no example LilyPond input ?
It took a lot of trial-and-error to get a line-spacing that showed the problem.

\version "2.14"
\paper { #(set-paper-size "a5") ragged-right=##f }
<< \relative c'' \new Voice {a2.( b4) a8 g2..
    b2.( d,4) e8 g4. g4 b4 |\break
    [r2]. f4 d' c c b b1 [r1]*5 }
\addlyrics { Oh, free -- dom! Oh, free -- dom o -- ver
    me ti -- bi com -- mis -- sum, } >>

Probably the desired change, in my guessed example, would remove the hyphens in 'freedom' and 'commissum'. In general, it might be wise to try to keep the ideal spacing between note-columns based on the rhythm alone, rather than have lyrics and other bulky decorations influence the 'ideal' spacing (note-spacing.cc:107).

2462.png

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-03-31

Originally posted by: janek.li...@gmail.com

Keith:
i posted two source files together with pdf output on nabble (since we're running out of diskspace on tracker) and linked to them in comment #4. Sorry for not making this more clear.

Also, here is the code of the la la straight straight example - it's really straightforward:

\score {
\relative f' { \repeat unfold 8 a4 }
\addlyrics { \repeat unfold 4 la \repeat unfold 4 straight }

\layout {
ragged-right = ##f
}
}

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-04-16

Originally posted by: k-ohara5...@oco.net

The guilty code seems to be spring.cc:103 inside merge()
avg_distance = max (min_distance + 0.3, avg_distance);
avg_distance is then used as the ideal spacing between columns.

An interesting experiment to see how the regtests look after removing that line.

The discussion of the patch to issue 1779 shows a previous attempt to move the 0.3 somewhere more appropriate; it should probably be a property of spaccing-spanner.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2012-04-18

Originally posted by: janek.li...@gmail.com

Thanks for suggestion, Keith!
I did a quick test, but removing that line doesn't seem to change anything in this lyrics issue. There are changes to the regtests - generally everything gets closer to previous note (sometimes it's good, sometimes bad).
More details will follow.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2013-12-22

Originally posted by: janek.li...@gmail.com

Keith,

somehow i forgot to post an update to this issue...

Turns out you were quite right, but it was necessary to remove another similar line. In branch dev/janek/experimental-fix-for-issue-2462 you can find a commit that does this: http://git.savannah.gnu.org/gitweb/?p=lilypond.git;a=commit;h=2eed825e869901991f5068a9d737ffdff5f45bc6

It turns out that removing these two lines fixes this issue completely (i have since tested it on many scores and i can confirm this), but there are some unwanted side-effects (basically, sometimes elements get too close) - that's why i haven't put it up for review. But since you're currently looking at spacing engine, you might be able to come up with something.

Cc: k-ohara5...@oco.net

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2013-12-22

Originally posted by: k-ohara5...@oco.net

Probably the 'padding and similar options were tweaked with this strange 0.3 in place, so some will need to be adjusted when it is removed.
I have no motivation to guess which side-effects you consider unwanted.

Cc: -k-ohara5...@oco.net

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-06-30

Originally posted by: janek.li...@gmail.com

fix Issue 2462.

there's no description yet, i'd just like to see what Patchy says about this patch.

http://codereview.appspot.com/108280044

Labels: Patch-new
Status: Started

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-06-30

Originally posted by: pkx1...@gmail.com

Patchy the autobot says: passes make, make check and a full make docs. reg test diffs here - https://www.hightail.com/download/ZUczTGs3TEQ4aVB2WnNUQw

Labels: -Patch-new Patch-review

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-06-30

Originally posted by: lemniska...@gmail.com

Nice! All changes look good, except input/regression/autobeam-show-defaults.ly (which seems to be not very problematic) and input/regression/unset-once.ly (which probably is really a different problem that this patch just triggered accidentally).

I'll look into these two in detail (and post a better description of what i'm doing) later this week. Thanks!

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-02

Originally posted by: pkx1...@gmail.com

Patch on countdown for July 5th 2014

If this does get pushed can we make sure we have tracker items for the two things that are listed in comment #15 above so we don't forget them?

Labels: -Patch-review Patch-countdown

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-02

Originally posted by: janek.li...@gmail.com

Actually, it turns out that these two files are not really a problem. My patch introduces slight differencies in spacing, and in case of these two tests thsese spacing changes (which are quite small by themselves) happen to trigger different line breaking.

With different page layout - for example, when i compile these files separately by hand - output with and without my patch is virtually identical.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-02

Originally posted by: janek.li...@gmail.com

In fact, change in autobeam-show-defaults.ly is for the better.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-02

Originally posted by: k-ohara5...@oco.net

The patch is just one line, but changes something rather fundamental about spacing.
I am surprised that it has the desired effect, and Janek has not yet explained why it works, so we should take a bit more time to think about it.

Labels: -Patch-countdown Patch-review

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-04

Originally posted by: pkx1...@gmail.com

Leaving Patch on review for now. If there is no change at all (no comments or no update to this Tracker to tell me otherwise) I'll set this back to Needs-work on the next countdown.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-07

Originally posted by: pkx1...@gmail.com

Patch on countdown for July 11th 2014

Labels: -Patch-review Patch-countdown

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-10

Originally posted by: pkx1...@gmail.com

Patch counted down - please push

Labels: -Patch-countdown Patch-push

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-10

Originally posted by: janek.li...@gmail.com

Pushed as

commit [rab6842155a003ba7d9243507594e3e973ebbb3e4]
Author: Janek Warchoł <lemniskata.bernoullego@gmail.com>
Date:   Mon Jun 30 22:14:16 2014 +0200

    Issue 2462: don't change ideal spacing when adding a rod

    When objects like lyrics are added to the PaperColumns, LilyPond inserts
    rods between these columns to ensure that the objects won't overlap.
    However, the ideal distance should remain unchanged. For example, in

    \layout {
      ragged-right = ##f
    }
    \relative f' { \repeat unfold 8 a4 }
    \addlyrics { \repeat unfold 4 la \repeat unfold 4 straight }

    notes in 2nd measure have long lyrics attached to them - min_distance_
    of the springs between these columns should be adjusted, but distance_
    (the ideal distance) should not: it should be the same as for notes
    with short lyrics. When the line is stretched so much that the minimum
    distances are not involved, the springs between notes in both measures
    should behave (almost) identical.

    Note that there are more such problems in the springs code, for example
    in merge_springs. The code should be rewritten, but that's not trivial.
    Since this one-liner fixes some instances of the problem without any bad
    side-effects, it makes sense to include it separately.

Woohoo! First GSoC issue is officially fixed :)

Labels: -Patch-push Fixed_2_19_10
Status: Fixed

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-10

Originally posted by: janek.li...@gmail.com

I have opened issue 4006 for the remainder of the spring code problems.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Google Importer - 2014-07-18

Originally posted by: fedel...@gmail.com

(No comment was entered for this change.)

Status: Verified

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

unnecessarily wide spacing caused by Lyrics

Issue Tracker for LilyPond

Searches

Help

#2462 unnecessarily wide spacing caused by Lyrics

Related

Discussion