simple-transduce v1.0.0

simple-transduce

A really simple transducer module to easily convert map-filter-reduce chains to single pass transducers.

Install

npm install simple-transduce

Concept

Take some existing code like this:

const collection = [1,2,3,4,5,6,7,8,9,10];

const multiplyBy5 = (value) => value + 5;
const notMultipleOf3 = (value) => value % 3 !== 0;
const square = (value) => value * value;
const notMultipleOf4 = (value) => value % 4 !== 0;

const newCollection = collection
  .map(multiplyBy5)
  .filter(notMultipleOf3)
  .map(square)
  .filter(notMultipleOf4);

// Result = [ 49, 121, 169 ]
// Total elements of all intermediate collections, including original: 34

This works fine, and the original list is transformed from 10 members to just 3 in the end, as expected. However this transformation happens over multiple intermediate collections, meaning while the original array has 10 elements, you actually iterate over 4 collections of 10, and thus (potentially) 40 elements. A transducer allows for a generic method that lets you do just a single pass of the 10 elements without sacrificing the clarity of having those 4 distinct transformation functions.

Converting map-filter to transduce

const st = require('simple-transduce');

const collection = [1,2,3,4,5,6,7,8,9,10];

const multiplyBy5 = st.mapTransducer((value) => value + 5);
const notMultipleOf3 = st.filterTransducer((value) => value % 3 !== 0);
const square = st.mapTransducer((value) => value * value);
const notMultipleOf4 = st.filterTransducer((value) => value % 4 !== 0);

const transducer = st.compose(multiplyBy5, notMultipleOf3, square, notMultipleOf4);

const newCollection = st.reduce(transducer(st.concat), [], collection);

// Result = [ 49, 121, 169 ]
// Total elements of all intermediate collections, including original: 10

Note that compose, reduce and concat are not special functions - they are found in many common functional libraries (and lodash) and are thus interchangable with those.

Functions

• mapTransducer(mapFunction) -> creates a composable transducer from the map function
• filterTransducer(filter) -> creates a composable transducer from the filter function
• reduce(reducerFunction, start, collection) -> generic reduce
• compose(...fns) -> generic compose. Example: compose(f, g) = (x) => f(g(x))
• pipe(...fns) -> generic pipe (compose inverse). Example: pipe(f, g) = (x) => g(f(x))
• concat(acc, item) -> generic concat reducer
• sum(acc, item -> generic sum reducer