Comment #1 on issue 1817 by d...@gnu.org: Bug is autobeaming 3/4 http://code.google.com/p/lilypond/issues/detail?id=1817
After mulling this over and actually looking at this specific example, I am no longer all too sure I agree: I don't actually like pulling the upbeat phrase in the first measure apart in that manner as it makes no musical sense (which is the reason to put 6 quavers to one beam in the first place). Interestingly, splitting the beam in the second measure makes more sense. Where is the difference? In the first measure, a non-beat note has been given its own flag, lending it too much importance. In the second measure, it is a beat note: giving it its own flag seems reasonable in contrast, unless we are talking an actual hemiolic pattern.
So from my personal gut feeling, I'd leave the first measure alone, but fix the second measure.
Which seems even more awful to cast into actual computer code/rules than your proposal.
_______________________________________________ bug-lilypond mailing list bug-lilypond@gnu.org https://lists.gnu.org/mailman/listinfo/bug-lilypond
