Sstable.bloom_pick: def of mylsm-lsm-store@0.6.0.1
Import: import mylsm-lsm-store@0.6.0.1/src/Sstable.bend as Sstable
Lines 343-357 of mylsm-lsm-store@0.6.0.1/src/Sstable.bend: https://jend.leeeet.dev/read.txt?path=mylsm-lsm-store@0.6.0.1/src/Sstable.bend&start_line=343&end_line=357

import Base
import mylsm-lsm-store@0.6.0.1/src/MemTable.bend as MemTable

# Size-gated entry point: sequential on both sides of the gate.
# MEASURED VERDICT (2026-09-26, 1M interleaved A/B, Darwin arm64, Bend 2.0.28):
# sequential median 225105 ms (4442 ops/s) vs 4-chunk 249150 ms (4014 ops/s)
# vs hybrid 261746 ms — the parallel builds cost ~10% (fork/join plus
# allocation overhead dominates; invisible at 100k, compounds over 244
# flushes). Confirmed under two machine states (all arms shifted +15k ms on
# the slow day; relative order unchanged). The parallel builds stay as
# law-covered spikes; the `!` variants stay opt-in behind the GPU recovery
# crash documented below.
def bloom_pick(small: Bool, +entries: List<&2, MemTable.Entry>, +nb: Nat) -> Bloom:
  match small:
    case True{}:
      bloom_of(entries, bloom_new(nb))
    case False{}:
      bloom_of(entries, bloom_new(nb))

Uses (7):
  https://jend.leeeet.dev/definition.txt?path=Base/base.bend&name=Bool
  https://jend.leeeet.dev/definition.txt?path=Base/base.bend&name=List
  https://jend.leeeet.dev/definition.txt?path=Base/base.bend&name=Nat
  https://jend.leeeet.dev/definition.txt?path=mylsm-lsm-store@0.6.0.1/src/MemTable.bend&name=Entry
  https://jend.leeeet.dev/definition.txt?path=mylsm-lsm-store@0.6.0.1/src/Sstable.bend&name=Bloom
  https://jend.leeeet.dev/definition.txt?path=mylsm-lsm-store@0.6.0.1/src/Sstable.bend&name=bloom_new
  https://jend.leeeet.dev/definition.txt?path=mylsm-lsm-store@0.6.0.1/src/Sstable.bend&name=bloom_of