Re: Uses single algorithm for side-position spacing. (issue 6827072)

[Keith OHara]

On Fri, 30 Nov 2012 01:37:48 -0800, <address@hidden> wrote:

> Ugh...this is a really nasty cyclic dependency.

There is sometimes a cyclic dependency message, but that is not the problem.  
I'm pretty sure I used \stemUp to determine the stem directions and break this 
dependency chain.

> 1) note_head_x_shift
> 2) stem direction
> 3) beam direction

There is a problem whenever a cross-staff beam /would/ auto-knee, even if the 
knee is prevented by overridden stem directions.  The beam is drawn as if the 
note-heads were all on one staff.