Re: Beams over rests

[Paul Hodges]

Thanks - adjusting the Stem.details.beamed-length as required did the job, though as you say it requires adjustment for each individual case.  So I gritted my teeth and did it - there were 54 instances (less than I would have guessed actually), so it was quite manageable.

To my surprise, mostly it looks quite good; there are a couple of bars in which no note has less than four leger lines, and these look more attenuated than I think is sensible - but the publishers will have to say.  Of course, if they agree a compromise in a few cases, the adjustment is very easy now the commands are in place.  Also, I've left beams which don't need this adjustment untouched - in a couple of cases tweaking them to match the extended beams either side of them might look better.

Thanks for the tip.

Paul

On 28/09/2021 10:14:26, "N. Andrew Walsh" <n.andrew.walsh@gmail.com> wrote:

Hi Paul,

I'm sure you know of the override pair:

\override Stem.stemlet-length = #0.5
\override Stem.details.beamed-lengths = #'(4.75)

( music with beamed rests here )

\revert Stem.stemlet-length
\revert Stem.details.beamed-lengths

I'm not aware of any way to set this globally. Well, the first one can be set once. You'll still have to use "[" and "]" to force beaming, and you should, for the sake of Best Practices™, set the stemlet length manually for each one. 

But this is the way I've had to do it; and while it's tedious to do that for every affected beam, it ensures that they look good and work properly.

Cheers,

A

On Tue, Sep 28, 2021 at 11:08 AM Paul Hodges <pwh@cassland.org> wrote:

The composer I'm working on is very fond of having beams over rests -
and indeed it helps a lot in reading his more complex rhythms.  However,
LilyPond is treating rests differently from notes, in that beam
positions are adjusted to suit the notes, whereas rests are then
adjusted to suit the beams.  So I get this kind of layout:
However, the publisher wants rests to remain in their standard position,
like this:

with occasional compromises to this extent:


I can obviously force this in individual cases, but again, is there a
global adjustment I can make to get nearer to what I'm being asked for?

Thanks,
Paul