Merge branch 'master' into dev/texi2html

[lilypond.git] / input / regression / beam-concave.ly

diff --git a/input/regression/beam-concave.ly b/input/regression/beam-concave.ly
index 52d66c107d9841844e1fb7ed5fc19e57b42433bf..5d15aa480e6c06f3110b45fc09973cd9671b03c2 100644 (file)
--- a/input/regression/beam-concave.ly
+++ b/input/regression/beam-concave.ly
@@ -1,84 +1,68 @@
-\version "1.7.18"
+\version "2.11.51"
 \header{
 
-texidoc = "Concave beams should be horizontal. informally spoken,
+  texidoc = "Fully concave beams should be horizontal. Informally spoken,
   concave refers to the shape of the notes that are opposite a
   beam. If an up-beam has high notes on its center stems, then we call
-  it concave.  This example shows borderline cases. Only the beams
-  that are marked `horiz' should be printed horizontally.  " }
+  it concave.
 
+  If a beam is fails a test, the desired slope is printed next to it.
 
-%{
- However, what exactly
-it is that makes a beam concave is still unclear.
+" 
+}
 
-Beams 1 and 3 should be sloped, 2 and 4 should be horizontal.  Two
-sane attempts of calculating concaveness of a beam fail to distinguish
-beams this way."
-%}
 
+resetMeasure = \set Score.measurePosition = #(ly:make-moment 0 1)
 
-\score{
-  \notes\relative c'{
+%% examples from Ross.
+rossFourBeams =\relative c'' {
+  \time 2/4
+  c8[ e b c]
+  b[ c d b]
+  c[ f b, c]
 
-%% This case seems easy: second beam should be horizontal.
-    
-    %% SCS-I Menuet I, m15
-    %% sloped
-    %% slope = -0.5ss
-    %% concaveness: 0.06
-    \clef bass
-    \time 3/4
-    \key g\major
-    a8 g fis e b dis
-    
-    %% SCS-I Menuet II, m20
-    %% horizontal
-    %% slope = 0
-    %% concaveness: 0.09
-    \key f\major
-    fis,^"horiz." a c es d c
+  a[ f a f] a[ a' a, a']
+  a[ d, e g] a[ f f g]
+  c,[ b f' e]
+  b[ e g, e']
+  c'[ c, c c]
+  c[ c c c']
+  f,,[ b a g]
+  f[g g e]
+  a[ d, b' g]
+}
+rossThreeBeams = \relative c'' {
+  \time 6/8
+  e[ a, c] f[ d g]
+  f,[ b g] a[ b f]
+  d'[ c b a b e]
+  c[ b a c b a]
+  d,[ g a g f e ]
+  a[ g f a g f]
+  d'[ g, a b c d]
+  f[ e d c b f']
+}
 
-%%% Sarabande: the first beam, obviously more concave, is not horizontal,
-%%% but is matched with the next beam in the piece: context.
-    
-    %% Sarabande: m24
-    %% sloped
-    %% concaveness: 0.00
-    \stemUp
-     d,16-[ a' b cis]
+rossBeams = \relative c'' {
+  \rossFourBeams
+  \rossThreeBeams
+}
 
-    %% Sarabande: m25
-    %% horizontal
-    %% concaveness:a: 0.12
-     a'16-[^"horiz." b c b]
-    
-% Hmm.  Concaveness of both: 1.75
-%     %% SCS-VI Prelude, m81
-%     %% slope = 0.0
-%     \stemBoth
-%     \key d\major
-%      e,8-[ cis a']
-    
-%     %% SCS-VI Prelude, m82
-%     %% slope = 0.1ss (possibly b.o. context?)
-%      g,-[ e' cis]
+\layout{
+  ragged-right = ##t
+  #(define debug-beam-quanting #t)
+}
 
-    
-%%% Han-Wen: this should be concave
-     a,16-[^"horiz." a' a a]
 
-    \clef treble
 
-%%%% This should not be concave (hwn)
-    [\stemUp bes8  \stemDown d'8 bes8]  
-  }
-  \paper{
-    raggedright = ##t
-  }
-}
 
-%% Local variables:
-%% LilyPond-indent-level:2
-%% End:
-%% new-chords-done %%
\ No newline at end of file
+#(define (<> x y) (not (= x  y)))
+\new Voice {
+  \override Beam #'positions = #(ly:make-simple-closure
+                                (ly:make-simple-closure
+                                 (append
+                                  (list chain-grob-member-functions `(,cons 0 0))
+                                  (check-slope-callbacks =))))
+  \rossBeams
+  
+}