LilyPond: Subdivided beam grouping for compound time signatures

Question

I'm trying to write a 12/8 drum score in Lilypond, with a lot of 32nd note phrasing, and the default beam grouping looks a bit ugly to me:

(Here's the above example in LilyBin: http://lilybin.com/w0s4nh/2)

I'd like it to be subdivided into 8th note groups, like this:

As can be seen on the LilyBin link I pasted above (line 36), I tried modifying the baseMoment parameter as per the documentation, but it doesn't matter what I set, the beaming always stays the same.

Is there anything else I should be doing here?

Answer 1

Here's a good start.

\layout {
  \context {
    \DrumStaff
      drumStyleTable = #(alist->hash-table drum-table)
      subdivideBeams = ##t
      baseMoment = #(ly:make-moment 1/8)
      beatStructure = #'(3 3 3 3)
  }
}
top = \drummode {
  cymc8 \repeat unfold 5 { cymr }
  cymr16 \repeat unfold 8 { sn32 }
  \repeat unfold 6 { toml }
  \repeat unfold 8 { tomfl }
}
bot = \drummode {
  bd4 bd16 bd
  sn bd sn bd sn bd
  bd8 \repeat unfold 5 { hhp }
}
\score {
  \new DrumStaff <<
    \new DrumVoice {
      \voiceOne \time 12/8 \top
    }
    \new DrumVoice {
      \voiceTwo \time 12/8 \bot
    }
  >>
  \layout {}
}

What this does end up doing is separating your code into more manageable parts: music is just music, layout is just layout, and score is just score. (Well, for the most part). So, when you need to make a change, you can pinpoint a lot more quickly where the change needs to happen.

In a case like this, you might need to temporarily intermix the music and the layout for only one Voice:

\layout {
  \context {
    \DrumStaff
      drumStyleTable = #(alist->hash-table drum-table)
  }
}
top = \drummode {
  \set Voice.subdivideBeams = ##t
  \set Voice.baseMoment = #(ly:make-moment 1/8)
  \set Voice.beatStructure = #'(3 3 3 3)
  cymc8 \repeat unfold 5 { cymr }
  cymr16 \repeat unfold 8 { sn32 }
  \repeat unfold 6 { toml }
  \repeat unfold 8 { tomfl }
}
bot = \drummode {
  bd4 bd16 bd
  sn bd sn bd sn bd
  bd8 \repeat unfold 5 { hhp }
}
\score {
  \new DrumStaff <<
    \new DrumVoice {
      \voiceOne \time 12/8 \top
    }
    \new DrumVoice {
      \voiceTwo \time 12/8 \bot
    }
  >>
  \layout {}
}

In my opinion, where you can achieve something "programmatically" like the first example, that is preferable. But, in a pinch, or where necessary, the second example can work.