Skip to main content

@edwinm/memoize-decorator@2.0.0

Add @memoize() to your class methods to have the result cached for future method calls.

Works with
It is unknown whether this package works with Bun
It is unknown whether this package works with Cloudflare Workers
This package works with Node.js
It is unknown whether this package works with Deno
This package works with Browsers
JSR Score
64%
Published
3 months ago

Test Coverage Status Quality Gate Status Socket Badge Snyk test results CodeFactor npm version GitHub

memoize-cache-decorator

Add the memoize decorator to your class methods to have the results cached for future calls.

This is an easy, clean and reliable way to prevent repeating unnecessary resource intensive tasks and improve the performance of your code.

Examples of resource intensive tasks that can be cached are: heavy calculations, network requests, file system operations and database operations.

With support for:

  • Both Node.js and browsers
  • Methods and getter functions
  • Async functions
  • Static functions
  • Cache expiration
  • Clearing the cache on two levels
  • Custom resolver function
  • TypeScript

Since TypeScript decorators are used, the source has to be TypeScript. Also, decorators can only be used for class methods and getters. Plain JavaScript decorators are planned for the future.

Installation

npm install --save-dev memoize-cache-decorator

Usage

class Example {
	@memoize()
	myFunction() {
		// …
	}
}

Simple example:

import { memoize } from "memoize-cache-decorator";

class Example {
	@memoize()
	myFunction() {
		// Heavy function getting data from disk, database or a server
		// For this example we return a random number
		return Math.random();
	}
}

const example = new Example();

// Instead of a different random number for each call, the first,
// cached number is returned each time.

console.log(example.myFunction());
//=> 0.7649863352328616
console.log(example.myFunction());
//=> 0.7649863352328616
console.log(example.myFunction());
//=> 0.7649863352328616

In practice, the function would probably do a fetch, read a file or do a database call. Here's another, more realistic example:

import { memoize } from "memoize-cache-decorator";

class Example {
	@memoize({ ttl: 5 * 60 * 1000 })
	async getData(path: string) {
		try {
			const response = await fetch(path, {
				headers: {
					Accept: "application/json",
				},
			});
			return response.json();
		} catch (error) {
			console.error(
				`While fetching ${path}, the following error occured`,
				error
			);
			return error;
		}
	}
}

const example = new Example();

const data = await example.getData("/path-to-data");

Now, every time getData is called with this path, it returns the data without fetching it over the network every time. It will do a fetch over the network again after 5 minutes or when clearFunction(example.getData) is called.

API

@memoize(config)

Memoize the class method or getter below it.

Type: [optional] Config

interface Config {
	resolver?: (...args: any[]) => string | number;
	ttl?: number;
}
resolver [optional] function

Function to convert function arguments to a unique key.

Without a resolver function, the arguments are converted to a key with a save version of JSON stringify. This works fine when the arguments are primitives like strings, numbers and booleans. This is undesirable when passing in objects with irrelevant data, like DOM elements. Use resolver to provide a function to calculate a unique key yourself.

Example:

import { memoize } from "memoize-cache-decorator";

class Example {
	@memoize({ resolver: (el) => el.id })
	myFunction(el) {
		// el is some complex object
		return fetch(`/rest/example/${el.id}`);
	}
}
ttl [optional] number

With ttl (time to live), the cache will never live longer than the given number of milliseconds.

import { memoize } from "memoize-cache-decorator";

class Example {
	// The result is cached for at most 10 minutes
	@memoize({ ttl: 10 * 60 * 1000 })
	getComments() {
		return fetch(`/rest/example/comments`);
	}
}

clear(instance, fn, arguments)

instance object
fn function
arguments [optional] arguments of fn

Clears the cache belonging to a memoized function for a specific instance and specific arguments.

Call clear with as arguments the instance, memoized function and memoized function arguments.

import { memoize, clear } from "memoize-cache-decorator";

class Example {
	@memoize()
	getDirection(direction: string) {
		return fetch(`/rest/example/direction/${direction}`);
	}

	southUpdated() {
		// The next time getComments("south") is called in this instance, data will
		// be fetched from the server again. But only for this instance.
		clear(this, this.getDirection, "south");
	}
}

clearFunction(fn)

fn function

Clears all caches belonging to a memoized function. All caches are cleared for the given function for all instances and for all arguments.

Call clearFunction with as argument the memoized function.

import { memoize, clearFunction } from "memoize-cache-decorator";

class Example {
	@memoize()
	getComments() {
		return fetch(`/rest/example/comments`);
	}

	commentsUpdated() {
		// The next time getComments() is called, comments will
		// be fetched from the server again.
		clearFunction(this.getComments);
	}
}

Tests

npm test

License

MIT © 2023 Edwin Martin

Add Package

deno add @edwinm/memoize-decorator

Import symbol

import * as mod from "@edwinm/memoize-decorator";

Add Package

npx jsr add @edwinm/memoize-decorator

Import symbol

import * as mod from "@edwinm/memoize-decorator";

Add Package

yarn dlx jsr add @edwinm/memoize-decorator

Import symbol

import * as mod from "@edwinm/memoize-decorator";

Add Package

pnpm dlx jsr add @edwinm/memoize-decorator

Import symbol

import * as mod from "@edwinm/memoize-decorator";

Add Package

bunx jsr add @edwinm/memoize-decorator

Import symbol

import * as mod from "@edwinm/memoize-decorator";