Merge pull request #22 from lsegal/unicode-support

Add support for hashing Unicode data by using Buffers

Author: dominictarr

helpers.js

@@ -0,0 +1,35 @@
+var Buffer = require('buffer').Buffer;
+var intSize = 4;
+var zeroBuffer = new Buffer(intSize); zeroBuffer.fill(0);
+var chrsz = 8;
+
+function toArray(buf, bigEndian) {
+  if ((buf.length % intSize) !== 0) {
+    var len = buf.length + (intSize - (buf.length % intSize));
+    buf = Buffer.concat([buf, zeroBuffer], len);
+  }
+
+  var arr = [];
+  var fn = bigEndian ? buf.readInt32BE : buf.readInt32LE;
+  for (var i = 0; i < buf.length; i += intSize) {
+    arr.push(fn.call(buf, i));
+  }
+  return arr;
+}
+
+function toBuffer(arr, size, bigEndian) {
+  var buf = new Buffer(size);
+  var fn = bigEndian ? buf.writeInt32BE : buf.writeInt32LE;
+  for (var i = 0; i < arr.length; i++) {
+    fn.call(buf, arr[i], i * 4, true);
+  }
+  return buf;
+}
+
+function hash(buf, fn, hashSize, bigEndian) {
+  if (!Buffer.isBuffer(buf)) buf = new Buffer(buf);
+  var arr = fn(toArray(buf, bigEndian), buf.length * chrsz);
+  return toBuffer(arr, hashSize, bigEndian);
+}
+
+module.exports = { hash: hash };

index.js

@@ -5,105 +5,72 @@ var rng = require('./rng')
 var md5 = require('./md5')
 
 var algorithms = {
-  sha1: {
-    hex: sha.hex_sha1,
-    base64: sha.b64_sha1,
-    binary: sha.str_sha1
-  },
-  sha256: {
-    hex: sha256.hex_sha256,
-    base64: sha256.b64_sha256,
-    binary: sha256.str_sha256
-  },
-  md5: {
-    hex: md5.hex_md5,
-    base64: md5.b64_md5,
-    binary: md5.bin_md5
-  }
+  sha1: sha,
+  sha256: sha256,
+  md5: md5
 }
 
-var algorithmsHmac = {
-  sha1: {
-    hex: sha.hex_hmac_sha1,
-    base64: sha.b64_hmac_sha1,
-    binary: sha.str_hmac_sha1
-  },
-  sha256: {
-    hex: sha256.hex_hmac_sha256,
-    base64: sha256.b64_hmac_sha256,
-    binary: sha256.str_hmac_sha256
-  },
-  md5: {
-    hex: md5.hex_hmac_md5,
-    base64: md5.b64_hmac_md5,
-    binary: md5.bin_hmac_md5
+var blocksize = 64
+var zeroBuffer = new Buffer(blocksize); zeroBuffer.fill(0)
+function hmac(fn, key, data) {
+  if(!Buffer.isBuffer(key)) key = new Buffer(key)
+  if(!Buffer.isBuffer(data)) data = new Buffer(data)
+
+  if(key.length > blocksize) {
+    key = fn(key)
+  } else if(key.length < blocksize) {
+    key = Buffer.concat([key, zeroBuffer], blocksize)
   }
-}
 
+  var ipad = new Buffer(blocksize), opad = new Buffer(blocksize)
+  for(var i = 0; i < blocksize; i++) {
+    ipad[i] = key[i] ^ 0x36
+    opad[i] = key[i] ^ 0x5C
+  }
 
-function error () {
-  var m = [].slice.call(arguments).join(' ')
-  throw new Error([
-    m,
-    'we accept pull requests',
-    'http://github.com/dominictarr/crypto-browserify'
-    ].join('\n'))
+  var hash = fn(Buffer.concat([ipad, data]))
+  return fn(Buffer.concat([opad, hash]))
 }
 
-exports.createHash = function (alg) {
+function hash(alg, key) {
   alg = alg || 'sha1'
-  if(!algorithms[alg])
-    error('algorithm:', alg, 'is not yet supported')
-  var s = ''
-  var _alg = algorithms[alg]
+  var fn = algorithms[alg]
+  var bufs = []
+  var length = 0
+  if(!fn) error('algorithm:', alg, 'is not yet supported')
   return {
     update: function (data) {
-      s += data
+      bufs.push(data)
+      length += data.length
       return this
     },
     digest: function (enc) {
-      enc = enc || 'binary'
-      var fn
-      if(!(fn = _alg[enc]))
-        error('encoding:', enc , 'is not yet supported for algorithm', alg)
-      var r = fn(s)
-      s = null //not meant to use the hash after you've called digest.
-      return r
+      var buf = Buffer.concat(bufs)
+      var r = key ? hmac(fn, key, buf) : fn(buf)
+      bufs = null
+      return enc ? r.toString(enc) : r
     }
   }
 }
 
-exports.createHmac = function (alg, key) {
-  if (!algorithmsHmac[alg])
-    error('algorithm:', alg, 'is not yet supported')
-  if (typeof key != 'string')
-    key = key.toString('binary')
-  var s = ''
-  var _alg = algorithmsHmac[alg]
-  return {
-    update: function (data) {
-      s += data
-      return this
-    },
-    digest: function (enc) {
-      enc = enc || 'binary'
-      var fn
-      if (!(fn = _alg[enc]))
-        error('encoding:', enc, 'is not yet support for algorithm', alg)
-      var r = fn(key, s)
-      s = null
-      return r
-    }
-  }
+function error () {
+  var m = [].slice.call(arguments).join(' ')
+  throw new Error([
+    m,
+    'we accept pull requests',
+    'http://github.com/dominictarr/crypto-browserify'
+    ].join('\n'))
 }
 
+exports.createHash = function (alg) { return hash(alg) }
+exports.createHmac = function (alg, key) { return hash(alg, key) }
 exports.randomBytes = function(size, callback) {
   if (callback && callback.call) {
     try {
-      callback.call(this, undefined, new Buffer(rng(size)));
-    } catch (err) { callback(err); }
+      callback.call(this, undefined, new Buffer(rng(size)))
+    } catch (err) { callback(err) }
   } else {
-    return new Buffer(rng(size));
+    return new Buffer(rng(size))
   }
 }
 

md5.js

@@ -7,24 +7,7 @@
  * See http://pajhome.org.uk/crypt/md5 for more info.
  */
 
-/*
- * Configurable variables. You may need to tweak these to be compatible with
- * the server-side, but the defaults work in most cases.
- */
-var hexcase = 0;   /* hex output format. 0 - lowercase; 1 - uppercase        */
-var b64pad  = "="; /* base-64 pad character. "=" for strict RFC compliance   */
-var chrsz   = 8;   /* bits per input character. 8 - ASCII; 16 - Unicode      */
-
-/*
- * These are the functions you'll usually want to call
- * They take string arguments and return either hex or base-64 encoded strings
- */
-function hex_md5(s){ return binl2hex(core_md5(str2binl(s), s.length * chrsz));}
-function b64_md5(s){ return binl2b64(core_md5(str2binl(s), s.length * chrsz));}
-function str_md5(s){ return binl2str(core_md5(str2binl(s), s.length * chrsz));}
-function hex_hmac_md5(key, data) { return binl2hex(core_hmac_md5(key, data)); }
-function b64_hmac_md5(key, data) { return binl2b64(core_hmac_md5(key, data)); }
-function str_hmac_md5(key, data) { return binl2str(core_hmac_md5(key, data)); }
+var helpers = require('./helpers');
 
 /*
  * Perform a simple self-test to see if the VM is working
@@ -156,25 +139,6 @@ function md5_ii(a, b, c, d, x, s, t)
   return md5_cmn(c ^ (b | (~d)), a, b, x, s, t);
 }
 
-/*
- * Calculate the HMAC-MD5, of a key and some data
- */
-function core_hmac_md5(key, data)
-{
-  var bkey = str2binl(key);
-  if(bkey.length > 16) bkey = core_md5(bkey, key.length * chrsz);
-
-  var ipad = Array(16), opad = Array(16);
-  for(var i = 0; i < 16; i++)
-  {
-    ipad[i] = bkey[i] ^ 0x36363636;
-    opad[i] = bkey[i] ^ 0x5C5C5C5C;
-  }
-
-  var hash = core_md5(ipad.concat(str2binl(data)), 512 + data.length * chrsz);
-  return core_md5(opad.concat(hash), 512 + 128);
-}
-
 /*
  * Add integers, wrapping at 2^32. This uses 16-bit operations internally
  * to work around bugs in some JS interpreters.
@@ -194,70 +158,6 @@ function bit_rol(num, cnt)
   return (num << cnt) | (num >>> (32 - cnt));
 }
 
-/*
- * Convert a string to an array of little-endian words
- * If chrsz is ASCII, characters >255 have their hi-byte silently ignored.
- */
-function str2binl(str)
-{
-  var bin = Array();
-  var mask = (1 << chrsz) - 1;
-  for(var i = 0; i < str.length * chrsz; i += chrsz)
-    bin[i>>5] |= (str.charCodeAt(i / chrsz) & mask) << (i%32);
-  return bin;
-}
-
-/*
- * Convert an array of little-endian words to a string
- */
-function binl2str(bin)
-{
-  var str = "";
-  var mask = (1 << chrsz) - 1;
-  for(var i = 0; i < bin.length * 32; i += chrsz)
-    str += String.fromCharCode((bin[i>>5] >>> (i % 32)) & mask);
-  return str;
-}
-
-/*
- * Convert an array of little-endian words to a hex string.
- */
-function binl2hex(binarray)
-{
-  var hex_tab = hexcase ? "0123456789ABCDEF" : "0123456789abcdef";
-  var str = "";
-  for(var i = 0; i < binarray.length * 4; i++)
-  {
-    str += hex_tab.charAt((binarray[i>>2] >> ((i%4)*8+4)) & 0xF) +
-           hex_tab.charAt((binarray[i>>2] >> ((i%4)*8  )) & 0xF);
-  }
-  return str;
-}
-
-/*
- * Convert an array of little-endian words to a base-64 string
- */
-function binl2b64(binarray)
-{
-  var tab = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
-  var str = "";
-  for(var i = 0; i < binarray.length * 4; i += 3)
-  {
-    var triplet = (((binarray[i   >> 2] >> 8 * ( i   %4)) & 0xFF) << 16)
-                | (((binarray[i+1 >> 2] >> 8 * ((i+1)%4)) & 0xFF) << 8 )
-                |  ((binarray[i+2 >> 2] >> 8 * ((i+2)%4)) & 0xFF);
-    for(var j = 0; j < 4; j++)
-    {
-      if(i * 8 + j * 6 > binarray.length * 32) str += b64pad;
-      else str += tab.charAt((triplet >> 6*(3-j)) & 0x3F);
-    }
-  }
-  return str;
-}
-
-exports.hex_md5 = hex_md5;
-exports.b64_md5 = b64_md5;
-exports.bin_md5 = str_md5;
-exports.hex_hmac_md5 = hex_hmac_md5;
-exports.b64_hmac_md5 = b64_hmac_md5;
-exports.bin_hmac_md5 = str_hmac_md5;
+module.exports = function md5(buf) {
+  return helpers.hash(buf, core_md5, 16);
+};