LilyIssues / Issues / #2527 Finger-flag collision

Google Importer - 2012-05-12

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

Adds Flag to Fingering side support list

http://codereview.appspot.com/6201079

Labels: Patch-new

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

Google Importer - 2012-05-12

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

The new patch exposes an area of the code base where the new skyline work will be key. Treating flags and stems as all-or-nothing objects pushes fingerings away too much, but if we used their vertical and horizontal skylines, the fingerings would only shift enough to avoid the part that they'd actually intersect with. So this patch would be a sort of stop gate solution before the skyline work.

I'm ambivalent about the patch. Normally I'd be for non-collision over collision, but this results in much wider default spacing. I'd be interested what pianists have to say. Specifically, la raison d'être of the patch comes from perusing Phil Hézaine's work on Schumann w/ fingerings.

What would Elaine Gould do?

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

Google Importer - 2012-05-12

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

(No comment was entered for this change.)

noMoreCollision.png

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

Google Importer - 2012-05-12

Originally posted by: dak@gnu.org

Elaine Gould would beat sense into the engraver given either choice. But then this is easier to do given human engravers.

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

Google Importer - 2012-05-12

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

The flag side is the last resort for fingering, so we only need do it when the music is very tight. Moving the numbers away from the stem would make them apply to the next chord. When it gets this tight I whiteout.

> Specifically, la raison d'être of the patch comes from perusing Phil Hézaine's work on Schumann
Could you be a bit more specific? In Album for the Young, the music does not get tight enough to put fingerings on the side.

Do you really want the patchy run for this ?

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

Google Importer - 2012-05-12

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

I've already run the regtests - it results in a few changed files.

Check out "Le cavalier sauvage" in http://superbonus.project.free.fr/IMG/pdf/Schumann-Album-pour-la-Jeunesse-avec-doigtes.pdf, left hand.

I'd be interested to see this work run through the patch to see if it improves the result. The downward pointing flagged stems are fine but the upward pointing flagged stems look meh with the fingering collisions.

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

Google Importer - 2012-05-12

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

Album for the Young is in German, Mike. You mean 8. Wilder Reiter (I looked first at ReiterStück)

We raise these: \relative c' { <f a>8-1-4-"ja" <f^1 a^4>_"nein" }

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

Google Importer - 2012-05-12

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

When you say "we raise these", do you mean "conventional engraving dictates that these be raised" or "these should be raised in Phil's version" or both?

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

Google Importer - 2012-05-13

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

LilyPond drivers raise the fingering (add-stem-support does it if you put them in chords).

This was actually my favorite piece when I was eight, and I still have the music I learned it from. Human engravers move the fingering a bit off-center, it seems, so the fingering is not quite centered over the head, if that helps to clear the stem.

imslp.org has the B&H edition, so you can look and tell me if I'm imagining the shift.

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

Google Importer - 2012-05-13

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

Patchy tests pass. Reg test diffs (significant ones) attached.

They look fine mostly, although it might be subjective if 'finger chords' and 'repetition chords' are now too 'high'.

Labels: -Patch-new Patch-review

Screenshot1.png

Screenshot2.png

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

Google Importer - 2012-05-13

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

Gould says to place the fingerings above or below the notes, to the left or even to the right. So she wouldn't generally put them where they could collide with the flag. Given that, moving them "too far" away in extremis would seem OK.

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

Google Importer - 2012-05-13

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

(No comment was entered for this change.)

Owner: mts...@gmail.com

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

Google Importer - 2012-05-13

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

Version 2.14 had an option add-stem-support that controlled this behavior. I tried searching mutopiaproject, but found only me using the option.

\set fingeringOrientations = #'(right)
\override Fingering #'add-stem-support = ##t
<f a>8-1-4 <f^1 a^4> <c-5 e-3> r

The patch (1) restores its effectiveness on Flags, and (2) makes add-stem-support the default.

I suggest leaving the default add-stem-support off. (Note that docs and reg-test 'finger-chords' would need changes if the default changes.)

2527.png

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

Google Importer - 2012-05-13

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

Adds Flag to Fingering side support list

http://codereview.appspot.com/6201079

Labels: -Patch-review Patch-new

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

Google Importer - 2012-05-13

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

Patchy the autobot says: LGTM.

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 - 2012-05-14

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

The latest patch restores the behavior of 2.14.
When the Flag was made separate from the Stem in LilyPond internals, the add-stem-support option no longer made scripts clear flags, so there was no easy way to make fingerings clear flags in several 2.15.x

I think this feature was an unintentional side effect, but we seem happy to get it back.

Labels: Regression

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

Google Importer - 2012-05-15

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

(No comment was entered for this change.)

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 - 2012-05-17

Originally posted by: dak@gnu.org

I think there is something wrong with the patch (or its characterization). If you take a look at comment #10, you'll find that most of the visual differences appear with quarter notes which don't even _have_ a flag.

It does not particularly help that apparently this has already been pushed without waiting for the end of countdown as [rcfe2e355b0630322f7d2f43934243af63844268f] if I am not mistaken.

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

Google Importer - 2012-05-17

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

Comment 10 showed the effects of an earlier patch-set
(which turned add-stem-support option on by default).

The pushed patch looks like the last one reviewed, which restores version 2.14 behavior.

Labels: -Patch-countdown fixed2_15_39
Status: Fixed

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

Google Importer - 2012-05-17

Originally posted by: dak@gnu.org

Ah, ok. Sorry for the confusion.

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

Google Importer - 2012-05-17

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

The original example (and the regression test) needs an override in order for fingering to avoid the flag.
\relative c' {
   \override Fingering #'add-stem-support
   \set fingeringOrientations = #'(right)
   <a-5 c-3>8
}
Fingerings still collide with each other.

Labels: -Started
Status: Started

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

Google Importer - 2012-05-17

Originally posted by: googlegr...@ensemble101.fr

I think you can push this override directly to staging.

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

Google Importer - 2012-05-18

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

Regtest combined with 'finger-chords.ly'.

Labels: -Regression

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

Google Importer - 2012-08-26

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

Could somebody confirm if this has been resolved?

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

Google Importer - 2012-08-27

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

Looking at it now, Mike.

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

Finger-flag collision

Issue Tracker for LilyPond

Searches

Help

#2527 Finger-flag collision

Related

Discussion