Re: one more tuplet problem

[Werner LEMBERG]

> > % . Looking at the third tuplet I believe that the beam spanner
> > %   code is not appropriate for tuplet lines at all.  It seems
> > %   best to follow the note head positions only, ignoring stems
> > %   for computing the slope.
> 
> I think this is wrong, and will lead to large gaps. In the case
> shown, a horizontal bracket is better.

You are right.  I think I've found a better algorithm (see below).


    Werner

======================================================================

%
% This file shows problems with tuplet brackets in lilypond
% CVS 2004-01-19 18:06 MET.
%
% . We probably need a different algorithm to compute the slope of tuplet
%   brackets.  What about the following addition:
%
%   + Compare the number of stems `up' and `down'.  If we have more `down'
%     stems, put the tuplet bracket `down'.  In the example below, the
%     third tuplet bracket should be put on the other side.

\version "2.1.12"
\include "deutsch.ly"

\score {
  \notes \relative c' {
    \times 4/5 { fis4 e es d des }
    \times 4/5 { fis,4 e es d des }
    \times 4/5 { b'''4 b,, h c cis }
    \times 4/5 { b4 b'' a as g }
  }

  \paper {
    linewidth = -1
  }
}

% EOF