Carl D. Sorensenwrote Thursday, March 19, 2009 2:42 PM
On 3/19/09 8:34 AM, "Mats Bengtsson" <mats.bengts...@ee.kth.se>
wrote:
Carl D. Sorensen wrote:
On 3/19/09 5:13 AM, "Mats Bengtsson" <mats.bengts...@ee.kth.se>
wrote:
Should we turn this into feature request to make the automatic
subdivision of beams even more flexible, with separate rules
for
different note lengths, or would the resulting scheme get too
messy to
use and implement?
It seems to me that this should be a feature request.
It also seems to me that we could make this quite easy to use as
a default.
We currently have beatGrouping used to end beams and beatLength
used to
subdivide beams.
If we want more control over ending beams, we use
override-auto-beam-setting
to add beam endings for specific beam types (16th, 32ned, etc.)
We should be able to add a functionality for
override-beam-subdivision that
is beam type specific, just like we've done for beam endings.
Although it
may be somewhat hard to get just right, the documentation is in
*much*
better shape than it used to be (thanks, Trevor), and we can use
a
corresponding syntax so it must only be learned once.
You mean, using
#(override-auto-beam-setting '(subdivide 1 32 3 4) 1 8)
to get a subdivision after the first four 32nd notes in 3/4
meter? Good
idea!
Actually, that's not what I meant. But it's what I should have
meant. So I
guess I should have meant that. I'm glad people smarter than me
are around
to pick up the pieces!
I'd go along with this. If the new subdivide rules didn't
inhibit the beatGrouping rules it would make it easier for
users to override the defaults, since we could remove some
of the beam-ending rules and replace them with subdivide
rules and still get acceptable beaming out of the box, with
fewer reverts being needed.
BTW, did we ever file a feature request for a function to
revert all the beam-ending rules in a particular time
signature? This would also be very useful.
Trevor
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user
