Need edition-engraver-compatible extended glissandi, will pay for solution if necessary

[Mason Hock]

In my music, I frequently use
- glissandi during which the performer rearticulates with a particular rhythm, 
represented with headless steams
- glissandi that extend longer than a bar or otherwise require intermediate 
headless stems to clarify duration, with a slur indicating that the performer 
should not rearticulate

My current solution is based on the hack[1] recommended in the documentation. 
Here is an example:

------------------------------------------------------------------------
\version "2.19.81"
\language "english"

% starts an extended glissand
glon = {
  \override NoteColumn.glissando-skip = ##t
  \hide NoteHead
  \override NoteHead.no-ledgers = ##t
  \override Stem.thickness = #0.5
  \override Accidental #'stencil = ##f
}
% terminates an extended glissando
gloff = {
  \revert NoteColumn.glissando-skip
  \undo \hide NoteHead
  \revert Glissando #'stencil
  \revert NoteHead.no-ledgers
  \revert Stem.thickness
  \revert Accidental #'stencil
}
% slightly increase slope of glissando so that line is
% not perfectly horizontal when glissing between notes
% on same staff line/space
glup = { \once\override Glissando.extra-dy = #0.5 }
gldown = { \once\override Glissando.extra-dy = #-0.5 }

\relative c' {
  \glup c8 \glissando \glon c-_ d-_ e-_ \tuplet 5/4 { e-_ f-_ g g a } |
  \gloff a a \gldown a \glissando \glon a a a \gloff af4 |
  \gldown af2 \glissando \gldown g4( \glissando \glon f |
  e) \gloff e r2 |
}
------------------------------------------------------------------------

This hack has several flaws.
- The stems are all different distances from the glissando line.
- Articulations have a noticably inconsistent slope.
- Selecting the least-bad pitch for each hidden note is tedious.
- Even slight changes to horizontal spacing can ruin the vertical position of 
the hidden notes.
- The position of a stem is not musical content and should not be represented 
by a pitch.

These flaws add up to a serious problem: Not only is the stem placement 
regardless of which hidden pitches I choose, but because horizontal spacing 
affects which pitches are least-bad, there is no way to choose the hidden 
pitches to be optimal for multiple editions. In order to fix this with edition 
engraver, I would have to tweak each stem individually for each edition, which 
would not realistically be maintainable. 

A proper implementation of extended glissandi would place each intermediate 
stem a fixed distance from the gliss line. It would also be nice to tweak the 
slope of all gliss lines globally, something I can not do now because the 
extra-dy value is different for ascending and descending glissandi, but this 
would be a convenience, not a necessity.

The ideal syntax would resemble this:

   c8 \gliss-end c c c c \gliss-end f

and result in a gliss from c to f, where
- the intermediate eighth note stems are all exactly the same distance from the 
gliss line
- the slope of the gliss line is slightly increased like in my example above

However, the bare minimum I need is for the stems to be automatically 
positioned vertically, and (if I need to tweak the slope manually) for them to 
be positioned relative to the gliss line *after the slope tweak*.

I don't understand Scheme or Lilypond's internals well enough to do this 
myself. If the solution is non-trivial for someone who does know Scheme and and 
Lilypond well, enough, I am willing to pay them for their time spent on the 
solution. The solution could be in the form of an include, or a feature of 2.19.

If anyone can help me with this, please let me know approximately how many 
hours of work you would expect it to take and what you would consider fair 
compensation for the amount of time spent.

Thanks,

Mason 

[1] 
http://lilypond.org/doc/v2.19/Documentation/notation/expressive-marks-as-lines#glissando

signature.asc
Description: PGP signature