Skip to content

A standalone, pure JavaScript implementation of the Mersenne Twister pseudo random number generator. Compatible with Node.js, requirejs and browser environments. Packages are available for npm, Jam and Bower.

License

Notifications You must be signed in to change notification settings

pigulla/mersennetwister

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

49 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

mersennetwister

The Mersenne Twister is a pseudo-random number generator invented by Makoto Matsumoto in 1997. Details can be found on the Wikipedia page and on Matsumoto's website.

This implementation is based on Sean McCullough's port of the original C code written by Makato Matsumoto and Takuji Nishimura.

Improvements over Sean's version are

  • more idiomatic, jshint-compliant and jsdoc-annotated code
  • compatible with Node.js, requirejs and browser environments
  • available as a module for npm, Jam and Bower
  • (somewhat) unit tested ;-)

Please note that the mersenne twister is not cryptographically secure.

Installation and setup

Node.js

Simply run npm install mersennetwister (or npm install --save mersennetwister of you want to directly add it to your package.json file). Import as usual: var MersenneTwister = require('mersennetwister');

Jam

Use the Jam command line tool: jam install mersennetwister and import as usual require(['mersennetwister'], function (MersenneTwister) { ...

Bower

Via the Bower tool: bower install mersennetwister

requirejs

Tools like Jam will usually configure requirejs so that it can be accessed via its package name (i.e., mersennetwister). If you use requirejs without such a customized configuration you need to import it via its camelcased filename: requirejs(['MersenneTwister'], function (MersenneTwister) { ...

Standalone

Download and include the src/MersenneTwister.js file: <script src="path/to/MersenneTwister.js"></script>. It is now available as the global variable MersenneTwister.

Usage

You can either just use the static random method of the module, which will return a random float just like Math.random does. If desired you can also instantiate your own instance of the mersenne twister and use its methods:

var mt = new MersenneTwister(seed); // if no seed is defined, seed randomly

mt.int();    // random 32-bit integer
mt.int31();  // random 31-bit integer

mt.rnd();       // random float in the interval [0;1[ with 32-bit resolution
mt.random();    // random float in the interval [0;1[ (same as mt.rnd() above)
mt.rndHiRes();  // random float in the interval [0;1[ with 53-bit resolution
mt.real();      // random float in the interval [0;1]
mt.realx();     // random float in the interval ]0;1[

mt.seed(seed);      // (re)seed the generator with an unsigned 32-bit integer
mt.seedArray(key);  // (re)seed using a state vector of unsigned 32-bit integers

Take a look at the inventor´s website if more detailed information is required.

Licensing

As indicated here, the Mersenne Twister algorithm is free to be used for any purpose, including commercial use. The license file of this module contains the BSD 3-clause license found in the C implementation on which it is based.

Changelog

0.2.3 (12/18/2015)
  • updated license used to BSD-3-Clause (@sibartlett)
0.2.2 (10/28/2015)
  • fixed typo readme
  • fixed license in package.json
  • bump some dependencies
0.2.1 (10/16/2014)
  • added bower.json (with ignore section) and .editorconfig
0.2.0 (07/13/2014)
  • added .random() alias to .rnd()
0.1.1 (06/19/2013)
  • published as a Jam module
  • registered as a Bower component
  • added installation instructions
  • completed jsdoc annotations and added build target
  • changelog added ;)
0.1.0 (06/16/2013)
  • initial release
  • published as an npm module

About

A standalone, pure JavaScript implementation of the Mersenne Twister pseudo random number generator. Compatible with Node.js, requirejs and browser environments. Packages are available for npm, Jam and Bower.

Resources

License

Stars

Watchers

Forks

Packages

No packages published