Re: Issue 1031 in lilypond: constantly-changing input/regression/rest-collision-beam-note.ly

[lilypond]

Comment #9 on issue 1031 by pnorcks: constantly-changing  
input/regression/rest-collision-beam-note.ly
http://code.google.com/p/lilypond/issues/detail?id=1031

(Ignore my last comment.)

There are certain cases where pointer addresses are compared in  
staff_priority_less.
 If I comment out these two lines of code, the problem seems to be fixed,  
though I
haven't run `make check' to verify.

  /* if neither grob has an outside-staff priority, the ordering will have  
no
     effect -- we just need to choose a consistent ordering. We do this to
     avoid the side-effect of calculating extents. */
  if (isinf (priority_1))
    return g1 < g2;


Is this the likely cause of the inconsistency?