Releases: CodeByZach/jhash.js
Releases · CodeByZach/jhash.js
Version 3.1.0
Npm package ready
Version 3.0.0
Consolidated all the hash functions into a single class.
Version 2.2 (2009-06-30) Paul Johnston
- More algorithms - RIPEMD-160, SHA-256 and SHA-512 are now supported, based on contributed code. Support for MD4 has been dropped.
- Compressed versions of the scripts are available, processed by YUI Compressor. sha1-min.js is less than 3KB.
- Improved Unicode support. By default, the input string is now encoded as utf-8, before being hashed, which is what most people want. It is still possible to use utf-16, but the
chrsz
configuration variable is gone. - Arbitrary output encoding. You can specify a string of characters to use for output encoding; the string length does not need to be a power of two. This is helpful for password makers, which want to use as much unpredictability as possible, in a short password.
- Bug fixes:
- The scripts now work correctly if immediate code is not executed, so they support some unusual platforms, including GreaseMonkey.
rstr2binl
now preinitialises the array with zeros, to work around bugs in some browsers.md5_vm_test
now works ifhex_case
is set to1
Version 2.1 (2002-12-15) Paul Johnston
- Binary string output
- Backward-compatibility with version 1 is removed.
- Base-64 output
- Unicode input. Note: this uses utf-16 encoding, which is rarely what is desired. Version 2.2 has improved Unicode support, with utf-8 encoding.
- HMAC - keyed hashing
- New function interface, to support the new features.
Version 1.1 (2001-06-15) Paul Johnston
- Renamed the add function to safe_add because add is a reserved word in ActionScript. Reported by Amanuel Workneh.
- All variables are now prefixed with var to make them local. Suggested by Richard Mitchell and Ian Richardson.
- The code now works in IE for Macintosh, thanks to Greg Holt and Andrew Kepert.
- The code now returns a lower-case hex string, like most MD5 functions. Thanks to Brian Lozier for noticing this.
- Code changed to use charCodeAt function instead of the dodgy sAscii variable. Thanks to Erik Johnson.
- Calls to Math.pow removed, now I've discovered the >>> operator.
- The code now works in Opera, because Peter Valach fixed a precedence problem.