ISAAC PRNG

ISAAC (Indirection, Shift, Accumulate, Add, Count) is a pseudorandom number generator (PRNG) specifically designed to deliver high performance while maintaining suitability for cryptographic applications.

Developed by Robert J. Jenkins Jr. in 1993, the ISAAC algorithm is renowned for its efficiency and robustness.
This Node.js implementation is a direct adaptation of Jenkins' original C implementation, which can be found in his official documentation at https://www.burtleburtle.net/bob/rand/isaacafa.html. The implementation faithfully preserves the core principles and functionality of the algorithm, ensuring both speed and reliability in generating pseudorandom numbers.

Installation

npm install isaac-js

Usage

const ISAAC = require('isaac-js');

const isaac = new ISAAC();

//
// Random int
//

const randomInt = isaac.random();

console.log('Random 32-bit unsigned integer between 0 and 2147483648 (exclusive):', randomInt);

//
// Random float
//

// Get the two 32-bit numbers
const random1 = isaac.random();
const random2 = isaac.random(); // Corrected typo from 'isacc.random' to 'isaac.random'
// Combine the numbers to get a 53-bit number
// The first number occupies the 21 most significant bits, the second the 32 least significant bits
const combined = (random1 * 2 ** 21) + (random2 >>> 11);
// Divide by 2^53 to get a number between 0 and 1
const randomFloat = combined / 2 ** 53;

console.log('Random 53-bit float between 0 and 1 (exclusive):', randomFloat);

API

seed(seedInput)

Initializes the internal state using the provided seed.
If no seed is provided, it initializes with a default value.

• seedInput (optional): A positive number or an array of positive numbers.

Important

To ensure an additional level of security, it is advisable to invoke this method at regular intervals (for example after 1 hour) or after a pre-set number of extractions.

random(iterations = 1)

Generates a 32-bit random integer number.

• iterations (optional): A positive integer specifying the number of iterations for the pseudo-random number generation (PRNG) process. Defaults to 1.
• Returns: A 32-bit random integer between 0 (0x00000000) and 2147483647 (0xFFFFFFFF).

NOTE: The following methods can be used to have detailed control over the generation process.

generate(iterations = 1)

Populates the internal buffer with pseudo-random values.

• iterations (optional): A positive integer specifying the number of iterations for the pseudo-random number generation (PRNG) process. Defaults to 1.

reset()

Resets the internal state by zeroing out variables and clearing buffers.

License

This project is licensed under the MIT License - see the LICENSE file for details.