Skip to content

avoidwork/tiny-lru

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Tiny LRU

What is Tiny LRU?

Tiny LRU is a tool that helps programs remember things for a short time, so they can work faster and use less memory. It's useful for both websites and apps.

What is a "Cache"?

A cache is like a small, quick-access box where a program stores things it might need again soon. Instead of looking up information from scratch every time (which can be slow), it checks the cache first.

What does "Least Recently Used" (LRU) mean?

Imagine your cache is a box that can only fit a certain number of items. When the box is full and you want to add something new, you remove the item you haven't used in the longest time. This keeps the most recently used things handy, and clears out the old things you don't need anymore.

Why use Tiny LRU?

  • Speeds things up: By remembering recent information, programs can respond faster.
  • Saves resources: Limits how much memory is used by only keeping the most important items.
  • Works anywhere: Can be used in many kinds of apps, big or small.

When is it helpful?

  • Websites that show the same info to many people
  • Apps that look up data from the internet
  • Any program that wants to avoid repeating slow or expensive work

How Does Tiny LRU Work?

  1. You set a limit for how many things the cache can remember.
  2. When the program needs to remember something, it puts it in the cache.
  3. If the cache is full, it removes the oldest unused item to make space.
  4. If the program needs something, it checks the cache first before doing extra work.

Using the factory

import {lru} from "tiny-lru";
const cache = lru(max, ttl = 0, resetTtl = false);

Using the Class

import {LRU} from "tiny-lru";
const cache = new LRU(max, ttl = 0, resetTtl = false);
import {LRU} from "tiny-lru";
class MyCache extends LRU {}

Interoperability

Lodash provides a memoize function with a cache that can be swapped out as long as it implements the right interface. See the lodash docs for more on memoize.

_.memoize.Cache = lru().constructor;
const memoized = _.memoize(myFunc);
memoized.cache.max = 10;

Testing

Tiny-LRU has 100% code coverage with its tests.

--------------|---------|----------|---------|---------|-------------------
File          | % Stmts | % Branch | % Funcs | % Lines | Uncovered Line #s 
--------------|---------|----------|---------|---------|-------------------
All files     |     100 |    91.46 |     100 |     100 |                   
 tiny-lru.cjs |     100 |    91.46 |     100 |     100 | 11-31,134,181,215 
--------------|---------|----------|---------|---------|-------------------

API Reference

Properties

first

Item in "first" or "bottom" position; default is null

const cache = lru();

cache.first; // null - it's a new cache!

last

Item in "last" or "top" position; default is null

const cache = lru();

cache.last; // null - it's a new cache!

max

Max items to hold in cache; default is 1000

const cache = lru(500);

cache.max; // 500

resetTtl

Resets item.expiry with each set() if true; default is false

const cache = lru(500, 5*6e4, true);

cache.resetTtl; // true

size

Number of items in cache

const cache = lru();

cache.size; // 0 - it's a new cache!

ttl

Milliseconds an item will remain in cache; lazy expiration upon next get() of an item

const cache = lru(100, 3e4);

cache.ttl; // 30000;

Methods

clear

Clears the contents of the cache

return {Object} LRU instance
cache.clear();

delete

Removes item from cache

param  {String} key Item key
return {Object}     LRU instance
cache.delete("myKey");

entries(["key1", "key2"])

Returns an Array cache items

param  {Array} keys (Optional) Cache item keys to get, defaults to `this.keys()` if not provided
return {Object} LRU instance
cache.entries(['myKey1', 'myKey2']);

evict

Evicts the least recently used item from cache

return {Object} LRU instance
cache.evict();

expiresAt

Gets expiration time for cached item

param  {String} key Item key
return {Mixed}      Undefined or number (epoch time)
const item = cache.expiresAt("myKey");

get

Gets cached item and moves it to the front

param  {String} key Item key
return {Mixed}      Undefined or Item value
const item = cache.get("myKey");

has

Returns a Boolean indicating if key is in cache

return {Object} LRU instance
cache.has('myKey');

keys

Returns an Array of cache item keys (first to last)

return {Array} Array of keys
console.log(cache.keys());

set

Sets item in cache as first

param  {String} key   Item key
param  {Mixed}  value Item value
return {Object}       LRU instance
cache.set("myKey", {prop: true});

setWithEvicted

Sets an item in the cache and returns the evicted item if the cache was full and an eviction occurred. If no eviction occurs, returns null.

param  {String} key   Item key
param  {Mixed}  value Item value
param  {Boolean} [resetTtl] Optionally reset the TTL for the item
return {Object|null}  The evicted item (shallow clone) or null
const evicted = cache.setWithEvicted("myKey", {prop: true});
if (evicted) {
  console.log("Evicted item:", evicted.key, evicted.value);
}

values(["key1", "key2"])

Returns an Array cache items

param  {Array} keys (Optional) Cache item keys to get
return {Array} Cache items
cache.values(['abc', 'def']);

License

Copyright (c) 2025 Jason Mulligan Licensed under the BSD-3 license.

About

Tiny LRU cache for Client or Server

Resources

License

Stars

Watchers

Forks

Sponsor this project

 

Packages

No packages published