Haitian Creole for Hash

Regle is a collection of utilities that generate ids.

Regle currently contains:

• An Hashids port that generates short, unique, non-sequential ids from numbers.
• An Optimus port that obfuscates ids based on Knuth's multiplicative hashing method.
• An Nano ID port that is a tiny, platform dependant secure, URL friendly, unique string ID generator.

You need to install the following libraries from HaxeLib and GitHub.

1. regle - haxelib git regle https://github.com/skial/regle master src
2. hashids - haxelib git hashids https://github.com/kevinresol/hashids master src
   • This is optional, but it is more widely used.

Then in your .hxml file, add -lib regle and you're set.

package ;

import uhx.uid.Hashids;

class Main {
	
	public static function main() {
		var hids = new Hashids();
		var id = hids.encode( [1, 2, 3] ); // `id` is now `o2fXhV`.
		var values = hids.decode( id ); // `values` is now `[1, 2, 3]`.
	}
	
}

• If you already have the hashids library included, uhx.uid.Hashids points to that instead of the one included.
• A bytes based implementation, which should be faster, is available via -D hashids_bytes. Compile bench.hxml and run one of the target outputs to see the speed differences.

package ;

import uhx.uid.Optimus;

class Main {

	public static function main() {
		var optimus = new Optimus(1580030173, 59260789, 1163945558);
		var id = optimus.encode(15); // 1103647397
		var value = optimus.decode(id); // 15
	}

}

To generate project specific values, call Optimus.make() which will generate an .optimus file containing a Json string. The only dependency at the moment to use Opimus.make() is to have the unzip command available. See OptimusSpec for an example.

package ;

import uhx.uid.Nanoid;

class Main {

	public static function main() {
		var nanoid = new Nanoid();
		var id = nanoid.toString(); // `nanoid.toString() == nanoid.toString()` == true.
		// Using a custom alphabet and length
		var id = Nanoid.generate(Nanoid.Url, 50);
	}

}

• The HashLink & Neko targets uses Math.random. They pass the flat distribution test, but wont be classed as secure.