ts-memoize-decorator

Simple memoization decorator for TypeScript.

In computing, memoization or memoisation is an optimization technique used primarily to speed up computer programs by storing the results of expensive function calls and returning the cached result when the same inputs occur again.

• https://en.wikipedia.org/wiki/Memoization

Installation

npm install --save ts-memoize-decorator

Usage:

type Primitive = string | number | boolean

@Memoize(hashFunction?: (...args: any[]) => Primitive)

or

interface Cache {
  get(key: Primitive): any | undefined;
  has(key: Primitive): boolean;
  set(key: Primitive, value: any);
}

@Memoize({ 
  hashFunction?: (...args: any[]) => Primitive,
  cacheFactory?: () => Cache
}) 

hashFunction accepts the same arguments as original method, and should return a value of a primitive type (a string or a number, and not an object of an array) that identifies that set of arguments deterministically (i.e. subsequent calls with an equal arguments will produce the same result).

cacheFactory parameter allows you to override default cache (new Map()). See tests for example of integration with node-lfu-cache

hashFunction may be omitted in the following cases:

1. decorated method is a get accessor

class MyClass {
    @Memoize()
    public get piSquared(): number {
        return Math.PI * Math.PI;
    }
}

2. decorated method is a method of zero arguments

class MyClass {
    @Memoize()
    public getHalfPI(): number {
        return Math.PI / 2;
    }
}

3. decorated method is a method of one argument, and the argument is of primitive value itself

class MyClass {
    @Memoize()
    public getAreaOfACircle(radius: number): number {
        return Math.PI * Math.pow(radius, 2);
    }
}

When using this decorator on a method that accepts an argument of a complex type, or a method of multiple arguments, hashFunction MUST be provided:

class MyClass {
    @Memoize((u: User) => u.id)
    public getUserBalance(user: User): number {
        return user.calculateBalance()
    }
}

This rule is enforced by throwing an exception at runtime. This is done to ensure that usage of @Memoize does not cause memory leaks by referencing potentially heavy objects as a cache keys, and also reduces cache size by providing smallest keys possible (contrary to implementations that choose to construct cache keys using JSON.stringify)