Menu
▾
▴
#3232 Inserting an additional column without notes breaks spacing
Accepted
nobody
None
Ugly
2013-04-28
2013-03-09
Anonymous
No
Originally created by: *anonymous
Originally created by: dak@gnu.org
Positioning subscripts/superscripts between notes requires attaching them to an intermediate time, like in the following example from the mailing list:
at =
#(define-music-function (parser location t e m)
(ly:duration? ly:event? ly:music?)
#{ << #m { \skip $t <>$e } >> #})
\new Voice \relative c'' {
\at 4 \turn f4. g16-. a-.
}
A similar use case is attaching dynamics like crescendo and decrescendo to intermediate positions of a whole note.
Now absent of actual note columns, the only visual cue about the positioning of some construct is its horizontal position, so to let visual and audible positioning (in case the event is audible) correspond, it would be important to have such note-less events positioned in a strictly proportional manner (though a barline or key change or similar would displace an _upcoming_ notecolumn as the reference point for the next time with an event).
Originally posted by: dak@gnu.org
I don't think you understood my proposal, so I'll try being more explicit and then we will see whether this is a misunderstanding or not. If you _did_ understand the proposal, I can't make sense of your objection, so it is clear that either you or me should have explained his point better. I guess it's me, so I'll try again:
I was not, repeat not, suggesting to change LilyPond's manner of placing notes. What I was instead talking about was LilyPond's manner of placing non-note material _between_ notes when the non-note material was produced at musical times where no note was produced. In that particular case, it would appear that LilyPond arrives at the position to place the non-note material by more or less pretending it is attached to a zero-width note playing at the same time, and then uses non-proportional spacing to find the location for this zero-width note.
However, since the note is not actually visible, its notehead does not help for reading the timing relations between the surrounding notes, and the non-note material that has been placed in-between.
So I want the location for the non-note material chosen by first placing the note material according to the current spacing algorithm, and then finding the location for the non-note material by picking its X position _strictly_ proportional to its timing between the surrounding note material. That way, non-note material can be positioned reliably by timing it appropriately.
Originally posted by: k-ohara5...@oco.net
I might understand now. When we put graphical objects in a column between regular note-columns, they affect the spacing. If these inserted columns contain no note-heads, LilyPond's note-spacing gives ugly spacing (especially when compressed).
\score { \relative c''' { g2 b | <<g2 {s4 s-| }>> b2 }
\layout {line-width = 5.5\cm } }
I suppose desired behavior is to space the half-notes as if there were no |, and center the | between centers (not left edges) of the half-notes.
(Hairpins indicating a swell {s4\< s\> <>\!} are asymmetric for the additional reason that the change from < to > is aligned to the left edge of a note-column, but the decrescendo finishes at the right edge.)
Originally posted by: k-ohara5...@oco.net
There is a mechanism to have a column ignored for note-spacing purposes, so the half-note spacing is unchanged by the | in between
\relative c''' { g2 b | <<g2 {s4 s-\tweak #'non-musical ##t -| }>> b2 }
but then we get the spacing rules for things like clef-changes, which are not what is desired for the articulation.
The workaround I have been using is to use a hidden note when placing articulations. (This also gives correct vertical placement of the articulation.)
Originally posted by: janek.li...@gmail.com
As far as i can see, there are two separate issues here:
1) wrong positioning of grobs (dynamics, articulations) attached to spacer rests,
2) the fact that attaching grobs to spacer rests affect spacing of ordinary note columns.
So, let's split this. I've created issue 3299 for 1).
Summary: objects attached to spacers shouldn't affect the spacing of the notes
Originally posted by: dak@gnu.org
I just noticed that you created a separate issue "for 1)" which was the _sole_ issue I reported. Then you gave this issue a title that is no longer related to the actual issue but rather to some point raised in the discussion. I don't even have a clue what this issue is now supposed to be about. The original issue was "LilyPond should use strict proportional spacing between note columns" and that is exactly what you have now placed in issue 3299.
Since you have not bothered providing anything but a title for the empty husk of an issue you have left lying here, could you either provide a _complete_ summary of what you think this issue is supposed to be about now, or close it as invalid or duplicate?
At the current point of time, this issue only has a title and nothing else.
Originally posted by: janek.li...@gmail.com
Hmm. Indeed; it's not clear enough.
My intent was that this issue would be about what Keith showed in comment #3: inserting an additional column without notes breaks the spacing of the notes, i.e. here
\score { \relative c''' { g2 b | <<g2 {s4 s-| }>> b2 }
\layout {line-width = 5.5\cm } }
1st measure is spaced differently than 2nd one.
Unfortunately, i cannot edit issue description. Should i create a new issue?
Originally posted by: PhilEHol...@googlemail.com
Description updated. Hope this is adequate.
Summary: Inserting an additional column without notes breaks spacing