Re: "Mensurstriche" (barlines between systems) and Repeat signs.

[Thomas Morley]

Am Fr., 1. Feb. 2019 um 20:19 Uhr schrieb Malte Meyn <address@hidden>:
>
> Am 01.02.19 um 18:42 schrieb Aaron Hill:
> > On 2019-02-01 9:04 am, Malte Meyn wrote:
> >> Am 01.02.19 um 16:53 schrieb Aaron Hill:
> >>> "" is already a valid bar line that has an empty glyph, so just
> >>> extend that one:
> >>>
> >>> %%%%
> >>> \version "2.19.82"
> >>>
> >>> \defineBarLine "-span" #'("|" "|" "|")
> >>> \layout { \context { \Score defaultBarType = "-span" } }
> >>> \new StaffGroup << \new Staff \repeat unfold 12 b'4
> >>>                     \new Staff \repeat unfold 12 b'4 >>
> >>> %%%%
> >>
> >> That’s not enough: #'("|" "|" "|") will make full bar lines at line
> >> breaks. When I try #'("" "" "|") instead, no span bar is printed at
> >> line end. But the following seems to work:
> >
> > I goofed when I posted my solution, as I had intended it to be #'("" ""
> > "|"), since it was meant to be only the span bar.
> >
> >> • in the middle of a line, the staff bar is taken from the name "-m"
> >> and the span bar from the third argument "|"
> >> • at the begin of a line, the second argument is used: "" is an empty
> >> staff bar *with empty span bar* (default for "" bar lines)
> >> • at the end of a line, the first argument is used: "" is not
> >> sufficient here, because the span bar is taken from the "" definition
> >> instead of the third argument; setting the first argument to "-m"
> >> looks into the third argument of this definition.
> >
> > That very much sounds like a bug and should be reported.  The final
> > argument is the span setting which should be agnostic of start and end
> > styles and vice versa.
> Hm … I don’t think that this is correct: How about
>         \defineBarLine ".|:-||" #'("||" ".|:" ".|")
> ? If the span bar ".|" would be used at all positions, the output at
> line breaks wouldn’t be correct: the "||" bar at the line end would have
> a ".|" span bar too …

I don't see any bug, neither I understand:

> But I don’t know why we have the span bar argument at all then …

Anyway, if a new barline is wished (with a new glyph, like "m"), it
needs to be defined from scratch.
The NR says:
"If additional elements are needed, LilyPond provides a simple way to
define them. For more information on modifying or adding bar lines,
see file ‘scm/bar-line.scm’."

Well, this is not very verbose ... lol ...  but to explain it, would
beyond the Notation _Reference_, rather the Extending Manual could be
amended.
We have ‘bar-line-define-bar-glyph.ly’ in the regtests, though.
lilypond.org/doc/v2.19/input/regression/8d/lily-e521e794.ly
(This example avoids caring about span-bars, though, setting span-bars to "")

In short:
A barline-printing-procedure, with the required arguments 'grob' and
'extent' needs to be assigned to a glyph (of string-length 1) via
'add-bar-glyph-print-procedure'.
Then the new barline could be defined via '\defineBarLine' or it's
scheme-equivalent 'define-bar-line' An example:

\version "2.19.82"

#(define (make-foo-bar-line grob extent)
  (let* ((stil
           (box-stencil
             (grob-interpret-markup grob (make-vcenter-markup "FOO"))
             0.2 0.1))
         (stil-x-extent (ly:stencil-extent stil X)))
;; To get span-bars correct, some moving is needed
    (ly:stencil-translate-axis
      (ly:make-stencil (ly:stencil-expr stil) stil-x-extent extent)
      (interval-center extent)
      Y)))

#(add-bar-glyph-print-procedure "f" make-foo-bar-line)
#(define-bar-line "f" "f" "f" "f")


\score {
  \new StaffGroup \with { \override SpanBar.color = #red }
    <<
      \repeat unfold 10 R1
      \repeat unfold 10 R1
      {
        R1
        \bar "f"
        r1^\markup \raise #20 \null
        \repeat unfold 8 R1
      }
    >>
  \layout {
    \context {
      \Score
      defaultBarType = "f"
    }
  }
}

For the topic of Mensurstriche, Malte already pointed out the howto:
\defineBarLine "-m" #'("-m" "" "|")

Which does not need a new printing procedure, but is an alternative to ""-bar.


Cheers,
  Harm