Re: Issue 879 in lilypond: RemoveEmptyStaffContext doesn't need to disable auto-knee-gap

[Keith OHara]

 <lilypond <at> googlecode.com> writes:
> Not sure what a regtest would look like.  What we're trying to do is show  
> that removing \override Beam #'auto-knee-gap = #'() from the definition of  
> RemoveEmptyStaves doesn't break something else. 

Hi Phil,
I see the point that the existing regression suite is not likely to have many 
examples that use both auto-knee with RemoveEmptyStaves.  So if passing `make 
test` leaves you unsatisfied, you could try to break things by adding some 
kneed 
beams to a couple vulnerable-looking tests of the RemoveEmpty.  I would play, 
deviously, for a while before writing a new test.

The thread Valentin referenced has a tiny example for the problem that required 
the override in the first place
http://lists.gnu.org/archive/html/lilypond-user/2009-09/msg00542.html
This score makes a Voice moves to/away from the Staff that is eventually 
removed.