> On Sat, Nov 10, 2012 at 12:42 AM, David Kastrup <d...@gnu.org> wrote:
>> That regression test tampered with internals of LilyPond in a manner >> that broke session integrity (any changes propagated to further files >> processed on the command line). It likely still does. The internals >> changed in the mean time. As a demonstration of the bleed-over between files processed on the command line, the following file (let's say named `frameEngraver4test.ly') will process without error if run with 'frameEngraver4.ly' on the command line: lilypond frameEngraver4 frameEngraver4test.ly The file: \version "2.17.6" frameStart = #(make-span-event 'FrameEvent START) frameEnd = #(make-span-event 'FrameEvent STOP) \relative c'' { \override Stem #'transparent = ##t \override Beam #'transparent = ##t \once \override Frame #'extender-length = #8 \frameStart dis'8[ e f \frameEnd ges] s2 \once \override Frame #'extender-length = #11 \frameStart d,8[ e f \frameEnd g] s2 \once \override Staff.BarLine #'X-extent = #'(+inf.0 . -inf.0) % line can cross bar line s2 \once \override Frame #'extender-length = #3.5 \frameStart fis'8[ bes,, aes, \frameEnd e''] } \layout { \context { \Global \grobdescriptions #my-grob-descriptions #my-event-classes } \context { \Voice \consists \frameEngraver } } _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user