Choosing the key and determining how to detect duplicates

In the next example, we index the states of a [finite-state machine][1]. Each state has a unique id that we'd like to use as key. There is only one instance of each unique FSMState, so we can instruct HashIndex to use the much simpler strict equality check instead of the more advanced isEqual. Likewise, keys are just numbers, so we can save come CPU cycles by directly comparing them.

import { HashIndex } from "scl";

interface FSMState {
  id: string;
  isFinal: boolean;
  edgeCount: number;
  nextStates: Array<[string, FSMState]>;
}

const s0: FSMState = {
  id: 0,
  isFinal: false,
  edgeCount: 1,
  nextStates: [
    ['a', s1],
  ]
}

const s1: FSMState = {
  id: 1,
  isFinal: true,
  edgeCount 0,
  nextStates: [],
}

const allStates = new HashIndex<({
  getKey: state => state.id,
  compareKeys: (a, b) => a < b,
  isEqual: (a, b) => a === b,
  elements: [s0, s1],
});

assert(allStates.getKey(1) === s1);

see

AVLTreeIndex for when the elements have to be sorted and there are frequent lookups

see

RBTreeIndex for when the elements have to be sorted and there are frequent mutations