On Tue, 4 Mar 2008, Hans Aberg wrote:
On 4 Mar 2008, at 13:20, Hugo Flordal wrote:
The problem is that you want to beam differently than
implied by the time signature you want to write. So you
want to beam in what might be described as 8 =
(1+1)+(1+1)+(1+1)+(1+1), whereas the common interpretation
of 4 is (1+1)+(1+1) or perhaps 1+1+1+1. If taken
literally, your beaming implies that the 4/4 has one
primary accent, and on the 1/4s have secondary accents.
Then each 1/4 has a yet subordinate accent on the second
1/8 of it. And LilyPond does not a dedicated function
doing just that, though it might be possible to do write
one.
Thank you for the reply, but I'm not sure I get it... Are
you saying (indirectly) that when I set the beatLength to
1/8, I implicitly change the timing to 8/8? Should I then
use 8/8 beaming rules? Is that why my 4/4 auto-beam rules do
not work but the */* rules do?
No. LilyPond currently only has ready-made high-level functions for
some standard situations, where time signature and beaming agree. In
other cases one will have to use low level functions, as I showed you.
Hans Aberg
Well, believe it or not, it _does_ work "as it should" when
I change the rules to
#(override-auto-beam-setting '(end * * 8 8) 1 4 'Voice)
#(override-auto-beam-setting '(end * * 8 8) 1 2 'Voice)
#(override-auto-beam-setting '(end * * 8 8) 3 4 'Voice)
That is to say, 8/8 instead of 4/4. But the timing is still
unchanged, i.e. 4/4, just like I wanted it. (I have only
tried it for the example that I posted originally, perhaps
this is a freak example, I'll experiment more later.)
Clearly, this is not a bug---it is an accidental feature!