Regression test different to normal compile.

[Matthew Rowles]

I was writing a regression test for the edge-height patch for the
OttavaBrackets (Thanks Joe!) and I came across an oddity.

The regression is producing incorrect output. The bracket heights are
limited to ~-1. The heights from a normal compile are correct.

This is the code:

\version "2.11.45"
\layout { ragged-right = ##t}
\relative c'''  {
   % large negative number
  #(set-octavation 1)
  \once \override Staff.OttavaBracket #'edge-height = #'(-12 . -5)
  a b c
  #(set-octavation 0)
  a
}

I've attached a png of the regression test and a pdf of the normal compile.



At the conclusion of the regression test, I exit with an error. I
don't know what it is, or if it's related or if it's because of my
computer:

<BLAHBLAHBLAH>
(./36/lily-74c369a7-systems.texi) (./8d/lily-642729cc.texidoc)
(./8d/lily-642729cc-systems.texi) (./cb/lily-60a13ce8-systems.texi)
(./58/lily-a4c1cd85.texidoc) (./58/lily-a4c1cd85-systems.texi) [33] )
(see the transcript file for additional information)</usr/share/texmf-texlive/f
onts/type1/bluesky/cm/cmbx12.pfb></usr/share/texmf-texlive/fonts/type1/bluesky/
cm/cmr10.pfb></usr/share/texmf-texlive/fonts/type1/bluesky/cm/cmsltt10.pfb></us
r/share/texmf-texlive/fonts/type1/bluesky/cm/cmsy10.pfb></usr/share/texmf-texli
ve/fonts/type1/bluesky/cm/cmtt10.pfb>
Output written on collated-files.pdf (33 pages, 140933 bytes).
Transcript written on collated-files.log.
/usr/local/bin/texi2dvi: pdfetex exited with bad status, quitting.
make[1]: *** [out-www/collated-files.pdf] Error 1
rm out-www/collated-files.texi out-www/version.itexi
make[1]: Leaving directory
`/home/matthew/lilypond/lilypond-2.11.45/input/regression'
make: *** [web] Error 2
make: Leaving directory
`/home/matthew/lilypond/lilypond-2.11.45/input/regression'
address@hidden:~/lilypond/lilypond-2.11.45$


Can anybody reproduce this?


-- 
Matthew Rowles

- Be Alert. Australia needs lerts.

lily-f4620cb2.png
Description: PNG image

ottava-edge.pdf
Description: Adobe PDF document