Re: Issue 1612 in lilypond: Change staff produces long stems

[lilypond]

Comment #1 on issue 1612 by address@hidden: Change staff produces  
long stems
http://code.google.com/p/lilypond/issues/detail?id=1612

This seems like it needs to be fixed in the Beam_collision_engraver such  
that the notes on the lower staff never enter into the equation.

After putting some pacifier prints in beam-collision-engraver.cc, the beam  
that gets placed too low yields `true' for Beam::is_cross_staff.  I think  
the problem, then, stems from the timestep during which staff crossing is  
triggered.

I'm guessing that if we can get LilyPond to cleanly ID the moment at which  
staff crossing actually happens, this problem will go away.  Any ideas?