lattice_sets

Grow-only, two-phase, and observed-remove CRDT sets for Gleam.

Use this package when replicas need set membership that can be merged without coordination. Choose the set type based on whether elements can be removed and later re-added.

Installation

gleam add lattice_sets

Quick example

import lattice_core/replica_id
import lattice_sets/or_set

pub fn main() {
  let node_a =
    or_set.new(replica_id.new("node-a"))
    |> or_set.add("alice")

  let node_b =
    or_set.new(replica_id.new("node-b"))
    |> or_set.add("bob")

  let merged = or_set.merge(node_a, node_b)

  or_set.contains(merged, "alice")
  // -> True
}

Modules

Module	Purpose
`lattice_sets/g_set`	Grow-only set. Elements can be added but never removed.
`lattice_sets/two_p_set`	Two-phase set. Elements can be removed once and cannot be re-added.
`lattice_sets/or_set`	Observed-remove set. Elements can be added, removed, and re-added.

Notes

All set modules expose new, add, contains, value, merge, to_json, and from_json.
two_p_set and or_set also expose remove operations.
Delta-state operations are available with add_with_delta and remove_with_delta.
or_set also supports diff, merge_with_diff, remove_all, remove_where, remove_with_bound, and prune.

Typed serialization

Use to_json_with and from_json_with with a payload encoder and decoder to serialize sets of integers, records, or tagged unions. The String to_json and from_json entry points retain their existing formats.

The generic ORSet format stores values with their causal tags in entries. It does not convert arbitrary values into JSON object keys. Serialization preserves tombstones, allocation state, and pruning metadata as well as visible members.

License

MIT

lattice_sets - v1.2.0