1//Javascript Punycode converter derived from example in RFC3492. 2//This implementation is created by [email protected] and released into public domain 3var punycode = new function Punycode() { 4 // This object converts to and from puny-code used in IDN 5 // 6 // punycode.ToASCII ( domain ) 7 // 8 // Returns a puny coded representation of "domain". 9 // It only converts the part of the domain name that 10 // has non ASCII characters. I.e. it dosent matter if 11 // you call it with a domain that already is in ASCII. 12 // 13 // punycode.ToUnicode (domain) 14 // 15 // Converts a puny-coded domain name to unicode. 16 // It only converts the puny-coded parts of the domain name. 17 // I.e. it dosent matter if you call it on a string 18 // that already has been converted to unicode. 19 // 20 // 21 this.utf16 = { 22 // The utf16-class is necessary to convert from javascripts internal character representation to unicode and back. 23 decode:function(input){ 24 var output = [], i=0, len=input.length,value,extra; 25 while (i < len) { 26 value = input.charCodeAt(i++); 27 if ((value & 0xF800) === 0xD800) { 28 extra = input.charCodeAt(i++); 29 if ( ((value & 0xFC00) !== 0xD800) || ((extra & 0xFC00) !== 0xDC00) ) { 30 throw new RangeError("UTF-16(decode): Illegal UTF-16 sequence"); 31 } 32 value = ((value & 0x3FF) << 10) + (extra & 0x3FF) + 0x10000; 33 } 34 output.push(value); 35 } 36 return output; 37 }, 38 encode:function(input){ 39 var output = [], i=0, len=input.length,value; 40 while (i < len) { 41 value = input[i++]; 42 if ( (value & 0xF800) === 0xD800 ) { 43 throw new RangeError("UTF-16(encode): Illegal UTF-16 value"); 44 } 45 if (value > 0xFFFF) { 46 value -= 0x10000; 47 output.push(String.fromCharCode(((value >>>10) & 0x3FF) | 0xD800)); 48 value = 0xDC00 | (value & 0x3FF); 49 } 50 output.push(String.fromCharCode(value)); 51 } 52 return output.join(""); 53 } 54 } 55 56 //Default parameters 57 var initial_n = 0x80; 58 var initial_bias = 72; 59 var delimiter = "\x2D"; 60 var base = 36; 61 var damp = 700; 62 var tmin=1; 63 var tmax=26; 64 var skew=38; 65 var maxint = 0x7FFFFFFF; 66 67 // decode_digit(cp) returns the numeric value of a basic code 68 // point (for use in representing integers) in the range 0 to 69 // base-1, or base if cp is does not represent a value. 70 71 function decode_digit(cp) { 72 return cp - 48 < 10 ? cp - 22 : cp - 65 < 26 ? cp - 65 : cp - 97 < 26 ? cp - 97 : base; 73 } 74 75 // encode_digit(d,flag) returns the basic code point whose value 76 // (when used for representing integers) is d, which needs to be in 77 // the range 0 to base-1. The lowercase form is used unless flag is 78 // nonzero, in which case the uppercase form is used. The behavior 79 // is undefined if flag is nonzero and digit d has no uppercase form. 80 81 function encode_digit(d, flag) { 82 return d + 22 + 75 * (d < 26) - ((flag != 0) << 5); 83 // 0..25 map to ASCII a..z or A..Z 84 // 26..35 map to ASCII 0..9 85 } 86 //** Bias adaptation function ** 87 function adapt(delta, numpoints, firsttime ) { 88 var k; 89 delta = firsttime ? Math.floor(delta / damp) : (delta >> 1); 90 delta += Math.floor(delta / numpoints); 91 92 for (k = 0; delta > (((base - tmin) * tmax) >> 1); k += base) { 93 delta = Math.floor(delta / ( base - tmin )); 94 } 95 return Math.floor(k + (base - tmin + 1) * delta / (delta + skew)); 96 } 97 98 // encode_basic(bcp,flag) forces a basic code point to lowercase if flag is zero, 99 // uppercase if flag is nonzero, and returns the resulting code point. 100 // The code point is unchanged if it is caseless. 101 // The behavior is undefined if bcp is not a basic code point. 102 103 function encode_basic(bcp, flag) { 104 bcp -= (bcp - 97 < 26) << 5; 105 return bcp + ((!flag && (bcp - 65 < 26)) << 5); 106 } 107 108 // Main decode 109 this.decode=function(input,preserveCase) { 110 // Dont use utf16 111 var output=[]; 112 var case_flags=[]; 113 var input_length = input.length; 114 115 var n, out, i, bias, basic, j, ic, oldi, w, k, digit, t, len; 116 117 // Initialize the state: 118 119 n = initial_n; 120 i = 0; 121 bias = initial_bias; 122 123 // Handle the basic code points: Let basic be the number of input code 124 // points before the last delimiter, or 0 if there is none, then 125 // copy the first basic code points to the output. 126 127 basic = input.lastIndexOf(delimiter); 128 if (basic < 0) basic = 0; 129 130 for (j = 0; j < basic; ++j) { 131 if(preserveCase) case_flags[output.length] = ( input.charCodeAt(j) -65 < 26); 132 if ( input.charCodeAt(j) >= 0x80) { 133 throw new RangeError("Illegal input >= 0x80"); 134 } 135 output.push( input.charCodeAt(j) ); 136 } 137 138 // Main decoding loop: Start just after the last delimiter if any 139 // basic code points were copied; start at the beginning otherwise. 140 141 for (ic = basic > 0 ? basic + 1 : 0; ic < input_length; ) { 142 143 // ic is the index of the next character to be consumed, 144 145 // Decode a generalized variable-length integer into delta, 146 // which gets added to i. The overflow checking is easier 147 // if we increase i as we go, then subtract off its starting 148 // value at the end to obtain delta.
149 for (oldi = i, w = 1, k = base; ; k += base) { 150 if (ic >= input_length) { 151 throw RangeError ("punycode_bad_input(1)"); 152 } 153 digit = decode_digit(input.charCodeAt(ic++)); 154 155 if (digit >= base) { 156 throw RangeError("punycode_bad_input(2)"); 157 } 158 if (digit > Math.floor((maxint - i) / w)) { 159 throw RangeError ("punycode_overflow(1)"); 160 } 161 i += digit * w; 162 t = k <= bias ? tmin : k >= bias + tmax ? tmax : k - bias; 163 if (digit < t) { break; } 164 if (w > Math.floor(maxint / (base - t))) { 165 throw RangeError("punycode_overflow(2)"); 166 } 167 w *= (base - t); 168 } 169 170 out = output.length + 1; 171 bias = adapt(i - oldi, out, oldi === 0); 172 173 // i was supposed to wrap around from out to 0, 174 // incrementing n each time, so we'll fix that now: 175 if ( Math.floor(i / out) > maxint - n) { 176 throw RangeError("punycode_overflow(3)"); 177 } 178 n += Math.floor( i / out ) ; 179 i %= out; 180 181 // Insert n at position i of the output: 182 // Case of last character determines uppercase flag: 183 if (preserveCase) { case_flags.splice(i, 0, input.charCodeAt(ic -1) -65 < 26);} 184 185 output.splice(i, 0, n); 186 i++; 187 } 188 if (preserveCase) { 189 for (i = 0, len = output.length; i < len; i++) { 190 if (case_flags[i]) { 191 output[i] = (String.fromCharCode(output[i]).toUpperCase()).charCodeAt(0); 192 } 193 } 194 } 195 return this.utf16.encode(output); 196 }; 197 198 //** Main encode function ** 199 200 this.encode = function (input,preserveCase) { 201 //** Bias adaptation function ** 202 203 var n, delta, h, b, bias, j, m, q, k, t, ijv, case_flags; 204 205 if (preserveCase) { 206 // Preserve case, step1 of 2: Get a list of the unaltered string 207 case_flags = this.utf16.decode(input); 208 } 209 // Converts the input in UTF-16 to Unicode 210 input = this.utf16.decode(input.toLowerCase()); 211 212 var input_length = input.length; // Cache the length 213 214 if (preserveCase) { 215 // Preserve case, step2 of 2: Modify the list to true/false 216 for (j=0; j < input_length; j++) { 217 case_flags[j] = input[j] != case_flags[j]; 218 } 219 } 220 221 var output=[]; 222 223 224 // Initialize the state: 225 n = initial_n; 226 delta = 0; 227 bias = initial_bias; 228 229 // Handle the basic code points: 230 for (j = 0; j < input_length; ++j) { 231 if ( input[j] < 0x80) { 232 output.push( 233 String.fromCharCode( 234 case_flags ? encode_basic(input[j], case_flags[j]) : input[j] 235 ) 236 ); 237 } 238 } 239 240 h = b = output.length; 241 242 // h is the number of code points that have been handled, b is the 243 // number of basic code points 244 245 if (b > 0) output.push(delimiter); 246 247 // Main encoding loop: 248 // 249 while (h < input_length) { 250 // All non-basic code points < n have been 251 // handled already. Find the next larger one: 252 253 for (m = maxint, j = 0; j < input_length; ++j) { 254 ijv = input[j]; 255 if (ijv >= n && ijv < m) m = ijv; 256 } 257 258 // Increase delta enough to advance the decoder's 259 // <n,i> state to <m,0>, but guard against overflow: 260 261 if (m - n > Math.floor((maxint - delta) / (h + 1))) { 262 throw RangeError("punycode_overflow (1)"); 263 } 264 delta += (m - n) * (h + 1); 265 n = m; 266 267 for (j = 0; j < input_length; ++j) { 268 ijv = input[j]; 269 270 if (ijv < n ) { 271 if (++delta > maxint) return Error("punycode_overflow(2)"); 272 } 273 274 if (ijv == n) { 275 // Represent delta as a generalized variable-length integer: 276 for (q = delta, k = base; ; k += base) { 277 t = k <= bias ? tmin : k >= bias + tmax ? tmax : k - bias; 278 if (q < t) break; 279 output.push( String.fromCharCode(encode_digit(t + (q - t) % (base - t), 0)) ); 280 q = Math.floor( (q - t) / (base - t) ); 281 } 282 output.push( String.fromCharCode(encode_digit(q, preserveCase && case_flags[j] ? 1:0 ))); 283 bias = adapt(delta, h + 1, h == b); 284 delta = 0; 285 ++h; 286 } 287 } 288 289 ++delta, ++n; 290 } 291 return output.join(""); 292 } 293 294 this.ToASCII = function ( domain ) { 295 var domain_array = domain.split("."); 296 var out = []; 297 for (var i=0; i < domain_array.length; ++i) { 298 var s = domain_array[i]; 299 out.push(
300 s.match(/[^A-Za-z0-9-]/) ? 301 "xn--" + punycode.encode(s) : 302 s 303 ); 304 } 305 return out.join("."); 306 } 307 this.ToUnicode = function ( domain ) { 308 var domain_array = domain.split("."); 309 var out = []; 310 for (var i=0; i < domain_array.length; ++i) { 311 var s = domain_array[i]; 312 out.push( 313 s.match(/^xn--/) ? 314 punycode.decode(s.slice(4)) : 315 s 316 ); 317 } 318 return out.join("."); 319 } 320}();
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.