Movable Type Home Page

Movable Type Scripts

SHA-3 Cryptographic Hash Algorithms

Sorry, this page doesn’t function in Internet Explorer 11.

A cryptographic hash (sometimes called ‘digest’) is a kind of ‘signature’ for a text or a data file. The SHA-3 family of hash routines generate almost-unique 224-bit, 256-bit, 384-bit, or 512-bit (28-/32-/48-/64-byte) signatures for a text. See below for the source code.

Enter any message to check its SHA-3 hash

JavaScript is not an obvious candidate for an SHA-3 (Keccak) implementation, as it is far from ‘close to the metal’, and also has no native 64-bit words.

Nonetheless, it is a good candidate for a ‘reference implementation’ to aid understanding of the algorithms; it is widely used and understood, its untyped C-style syntax is easily followed, and it has acquired convenient functional-style elements.

Even the SHA-3 ‘reference’ implementations I found left me scratching my head a bit, so I thought to complement my existing implementations of SHA1 and SHA-2 (SHA256 & SHA512), I would make this SHA-3 implementation. SHA-3 / Keccak lends itself to a wide variety of implementations / optimisations – I have deviated as little as possible from the reference; I am anything but a cryptography expert, and have no inclination to ‘improve’ on the algorithms documented in the original references/standard, just to help understanding them. Note that while there are no ‘magic numbers’ in Keccak, it is conventional to pre-calculate the round constants for the ι (iota) step. It is also conventional to combine the ρ and π steps, as explained in the code. §ection numbers relate the code back to sections in the Keccak reference.

Using Chrome on a low-to-middling Core i5 PC, in timing tests this script will hash a short message in around 0.4 – 0.6 ms; longer messages will be hashed at a speed of around 0.2 – 0.4 MB/sec.

As a trial, I implemented the ‘bit-interleaved’ variant of Keccak to enable use of (more native) 32-bit integer operations, but (at least in Chrome) the interleaving operation makes it slower than the Long version (a tribute to how well optimised constructors are in JavaScript!).

Source code (below) is also available on GitHub. §ection numbers relate the code back to sections in the reference.

With its untyped C-style syntax, JavaScript reads remarkably close to pseudo-code: exposing the algorithms with a minimum of syntactic distractions. This code is to aid understanding of the algorithms, though can also be used as-is in browsers and Node.js, where performance is not a prime consideration.

OSI MIT License I offer these scripts for free use and adaptation to balance my debt to the open-source info-verse. You are welcome to re-use these scripts [under an MIT licence, without any warranty express or implied] provided solely that you retain my copyright notice and a link to this page.

Paypal donation If you would like to show your appreciation and support continued development of these scripts, I would most gratefully accept donations.

If you have any queries or find any problems, contact me at ku.oc.epyt-elbavom@cne-stpircs.

© 2016 Chris Veness