Issue 2059 in lilypond: Bar numbering of alternatives

[lilypond]

Status: Accepted
Owner: ----
Labels: Type-Enhancement

New issue 2059 by address@hidden: Bar numbering of alternatives
http://code.google.com/p/lilypond/issues/detail?id=2059

Requested by Xavier Scheuer address@hidden

This is by far the most popular enhancement request from the French
users mailing list: Bar numbering of repeat volta _alternatives_.

Currently LilyPond counts the measures of each "alternative" as
normal, unrepeated, measures hence incrementing the currentBarNumber
in both alternative "1." and "2.".  Several references supports what is
actually asked by many users: bar numbering of alternatives should not
be counted twice.

%%%%

\version "2.15.20"

\relative c' {
 \override Score.BarNumber #'break-visibility = #'#(#t #t #t)
 \repeat volta 2 {
   c1 |  % 1
   c1 |  % 2
   c1 |  % 3
 }
 \alternative {
   {
     c1 |  % currenlty: 4  % request: 4 or 4a
   }
   {
     c1 |  % currently: 5  % request: 4 or 4b !!
   }
 }
 \textLengthOn
 c1^"currently: 6  request: 5 !!" |  % currently: 6  % request: 5 !!
 \textLengthOff
 c1 |  % currently: 7  % request: 6
 % etc.
}

%%%%

Please do not discard this (popular) request again.
I'd suggest to have _at least_ the possibility to switch on this
feature, for example via a variable setting such as
 \set barNumberAlternativeOnce = ##t

Well-know music publishing houses that use this numbering includes
Simrock, Breitkopf, Bärenreiter, Henle.

E.g. Henle et Breitkopf: |1. 34 | 35 :||2. 34 | 35 etc.,
Bärenreiter: |1. 34a | 35a :|| 2. 34b | 35b.

I do not know what say music engraving books like Ross, Read or Gould
but here is what says Jean-Pierre Coulon in his "Repository of
music-notation mistakes" or "Essay on the true art of music engraving":

 "When first ending-second ending measures are present, measure numbers
 of only the first ending measure act on the numbering. If necessary,
 corresponding measure numbers have the subscripts a and b."

Section 5.4 page 9.
http://www.icking-music-archive.org/lists/sottisier/sottieng.pdf
http://www.icking-music-archive.org/lists/sottisier/sottisier.html

And here are some messages of users supporting this request.

lilypond-user-fr (in French):
http://lilypond-french-users.1298960.n2.nabble.com/Repetitions-volta-et-numeros-de-mesure-td7036772.html
http://lilypond-french-users.1298960.n2.nabble.com/numerotation-des-mesures-en-cas-de-reprise-avec-alternatives-td6936029.html
http://lilypond-french-users.1298960.n2.nabble.com/numeros-de-mesures-et-reprises-td5565647.html
http://lilypond-french-users.1298960.n2.nabble.com/Numeros-de-mesure-dans-1-et-2-td5227683.html
http://lilypond-french-users.1298960.n2.nabble.com/Numerotation-de-mesures-et-reprises-td4474721.html

lilypond-user:
http://lists.gnu.org/archive/html/lilypond-user/2009-07/msg00522.html
http://lists.gnu.org/archive/html/lilypond-user/2007-09/msg00096.html
http://lists.gnu.org/archive/html/lilypond-user/2006-09/msg00113.html

The implementation and what to do in the case of alternatives of
different measures duration are open to discussion.
But keep in mind that in 90% of the cases there are only 2 alternatives
which have the same measures duration.  In the remaining cases
the first alternative is often the shortest.

Thanks for your attention.

Cheers,
Xavier (and many French users)