Functional Programming example: Implementing `map` in terms of `reduce`

map-reduce.js

// implementing map in terms of reduce

function count(coll) {
  return (typeof coll === "undefined") ? 0 : coll.length;
}

function isEmpty(coll) {
  return (0 === count(coll));
}

function first(coll) {
  return isEmpty(coll) ? null : coll[0];
}

function rest(coll) {
  return isEmpty(coll) ? [] : coll.slice(1);
}

function conj(coll, item) {
  var copy = Array.from(coll);
  copy.push(item);
  return copy;
}

function reduce(f, init, coll) {
  if (isEmpty(coll)) {
    return init;
  }
  
  return reduce(f, f(init, first(coll)), rest(coll));
}

// the point of this gist

function map(f, coll) {
  return reduce((acc, i) => conj(acc, f(i)), [], coll);
}

// some functions to play around with

function add(a, b) {
  return a + b;
}

function inc(x) {
  return x + 1;
}

// use it

reduce(add, 0, [1, 2, 3, 4, 5]);
map(inc, [0, 1, 2, 3]);

Note that reduce is implemented recursively here. This will overflow the stack without tail-call optimization for large enough collections, but it is the most straightforward implementation of the algorithm. The point of this gist is to illustrate the idea, not to produce production-quality code (besides, map and reduce are native now).