Most caching libraries don't do locking, meaning that >1 process can be calculating a cached value at the same time. Since you presumably cache things because they cost CPU, database reads, or money, doesn't it make sense to lock while caching?
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.babelrc
.gitignore
LICENSE
README.md
index.js
package-lock.json
package.json
test.js

README.md

lock_and_cache

Lock and cache using redis!

Most caching libraries don't do locking, meaning that >1 process can be calculating a cached value at the same time. Since you presumably cache things because they cost CPU, database reads, or money, doesn't it make sense to lock while caching?

Quickstart

const cache = require('lock_and_cache')

// standalone mode
function getStockQuote (symbol) {
  return cache(['stock', symbol], 60, async function () {
    // fetch stock price from remote source, cache for one minute
    // calling this multiple times in parallel will only run it once
  })
}

// wrap mode
const getStockQuote = cache.wrap(60, async function stock (symbol) {
  // fetch stock price from remote source, cache for one minute
  // calling this multiple times in parallel will only run it once
  // the cache key is based on the function name and arguments
})

Install

npm install --save lock_and_cache

Theory

lock_and_cache...

  1. returns cached value (if exists)
  2. acquires a lock
  3. returns cached value (just in case it was calculated while we were waiting for a lock)
  4. calculates and caches the value
  5. releases the lock
  6. returns the value

As you can see, most caching libraries only take care of (1) and (4) (well, and (5) of course).

Setup

Just setting REDIS_URL in your environment is enough.

REDIS_URL=redis://redis:6379/2

If you want to put the cache and the locks in different places (which is useful if you want to invalidate the cache without messing with any locks currently held), you can do:

CACHE_URL=redis://redis:6379/2 LOCK_URL=redis://redis:6379/3

Distributed locking

Distributed locking is supported and uses redlock.

const lockAndCache = require('lock_and_cache')
const cache = lockAndCache.configure([
  'redis://client-1',
  'redis://client-2',
  'redis://client-3'
])

All clients you specify will be used for locking, but only the first will be used for caching.

API

lockAndCache (mixed key, number ttl, function work)

  • returns Promise
  • key can be a number, string, boolean, object, etc. It will be passed to JSON.stringify.
  • ttl is the time-to-live for the cached value in seconds
  • work is the function that will be executed if the value is not cached. It can return a promise.

lockAndCache.configure (mixed client(s)[, object options])

  • returns a lockAndCache function with the given configuration
  • client(s) can be a connection url, a redis options object, or an array of connection urls and/or options objects
  • options get passed to redlock

Global configuration

The LOCK_DRIFT_FACTOR, LOCK_RETRY_COUNT, and LOCK_RETRY_DELAY environment variables can be set to configure the global lockAndCache function's locking behavior. You probably don't need this unless the jobs you are caching are either very fast or very slow. By default, the global lockAndCache function will retry every 100ms for one hour using the standard drift factor.

Contributing

Please send me your pull requests!