ioredis-memoize

A lightweight Redis-powered caching layer for async functions.

Install

$ npm i ioredis-memoize --save

Usage

const Redis = require('ioredis')
const Memoize = require('ioredis-memoize')

const redis = new Redis()

// create a memoizer with global defaults
const memoize = Memoize({
  client: redis,
  prefix: 'cache:',
  ttl: 10 * 60 * 1000 // default ttl in ms
})

// memoize(fn[, fnOptions]) => AsyncFunction with .raw/.get/.set/.clear
const cached = memoize(async function someAsyncFn (number) {
  console.log(`someAsyncFn: ${number}`)
  return number
}, {
  // optional per-function overrides
  // key can be string or function; return false to skip caching
  key (number) {
    if (number >= 4) {
      return false // only cache when number < 4
    }
    return this.name + ':' + number
  },
  ttl: 1000
})

Options

• client {RedisClient}: redis client instance (e.g. new Redis()). Required (globally or per function).
• prefix {string}: prefix for redis cache keys, default ''.
• key {string | Function}: default key or key generator. If function returns false, skip get/set cache. Default is fn.name.
• ttl {number | Function}: time to live in milliseconds, or a function (...args) => number that returns TTL dynamically. Required (globally or per function).
• get {Function}: custom getter (redis, key) => value | undefined.
• set {Function}: custom setter (redis, key, value, ttl) => void.

Example

import Redis from 'ioredis'
import Memoize from 'ioredis-memoize'

const redis = new Redis()

const memoize = Memoize({
  client: redis,
  prefix: 'cache:',
  ttl: 10 * 60 * 1000 // default ttl in ms
})

;(async function () {
  const someAsyncFn = memoize(async function someAsyncFn (number) {
    console.log(`someAsyncFn: ${number}`)
    return number
  }, {
    key: function (number) {
      if (number >= 4) {
        return false // only cache when number < 4
      }
      return this.name + ':' + number
    }
  })

  console.log(await someAsyncFn(1))
  console.log(await someAsyncFn(2))
  console.log(await someAsyncFn(2)) // get from cache
  console.log('---')

  console.log(await someAsyncFn.get(3))
  await someAsyncFn.set(3, 'some value') // manually set cache
  console.log(await someAsyncFn.get(3)) // get from cache
  console.log('---')

  console.log(await someAsyncFn(4)) // not cache
  console.log(await someAsyncFn.get(4))
  console.log('---')

  console.log(await someAsyncFn.raw(5)) // skip cache
  console.log(await someAsyncFn.get(5))
  console.log('---')

  console.log(await someAsyncFn.clear(1)) // clear cache
  console.log(await someAsyncFn.clear(2)) // clear cache
  console.log(await someAsyncFn.clear(3)) // clear cache
  console.log(await someAsyncFn.clear(4)) // no effect, because there is no cache
})().catch(console.error)

Test (100% coverage)

$ npm run test

License

MIT