[2537] | 1 | <?php |
---|
| 2 | /* 9 April 2008. version 1.1 |
---|
| 3 | * |
---|
| 4 | * This is the php version of the Dean Edwards JavaScript's Packer, |
---|
| 5 | * Based on : |
---|
| 6 | * |
---|
| 7 | * ParseMaster, version 1.0.2 (2005-08-19) Copyright 2005, Dean Edwards |
---|
| 8 | * a multi-pattern parser. |
---|
| 9 | * KNOWN BUG: erroneous behavior when using escapeChar with a replacement |
---|
| 10 | * value that is a function |
---|
| 11 | * |
---|
| 12 | * packer, version 2.0.2 (2005-08-19) Copyright 2004-2005, Dean Edwards |
---|
| 13 | * |
---|
| 14 | * License: http://creativecommons.org/licenses/LGPL/2.1/ |
---|
| 15 | * |
---|
| 16 | * Ported to PHP by Nicolas Martin. |
---|
| 17 | * |
---|
| 18 | * ---------------------------------------------------------------------- |
---|
| 19 | * changelog: |
---|
| 20 | * 1.1 : correct a bug, '\0' packed then unpacked becomes '\'. |
---|
| 21 | * ---------------------------------------------------------------------- |
---|
| 22 | * |
---|
| 23 | * examples of usage : |
---|
| 24 | * $myPacker = new JavaScriptPacker($script, 62, true, false); |
---|
| 25 | * $packed = $myPacker->pack(); |
---|
| 26 | * |
---|
| 27 | * or |
---|
| 28 | * |
---|
| 29 | * $myPacker = new JavaScriptPacker($script, 'Normal', true, false); |
---|
| 30 | * $packed = $myPacker->pack(); |
---|
| 31 | * |
---|
| 32 | * or (default values) |
---|
| 33 | * |
---|
| 34 | * $myPacker = new JavaScriptPacker($script); |
---|
| 35 | * $packed = $myPacker->pack(); |
---|
| 36 | * |
---|
| 37 | * |
---|
| 38 | * params of the constructor : |
---|
| 39 | * $script: the JavaScript to pack, string. |
---|
| 40 | * $encoding: level of encoding, int or string : |
---|
| 41 | * 0,10,62,95 or 'None', 'Numeric', 'Normal', 'High ASCII'. |
---|
| 42 | * default: 62. |
---|
| 43 | * $fastDecode: include the fast decoder in the packed result, boolean. |
---|
| 44 | * default : true. |
---|
| 45 | * $specialChars: if you are flagged your private and local variables |
---|
| 46 | * in the script, boolean. |
---|
| 47 | * default: false. |
---|
| 48 | * |
---|
| 49 | * The pack() method return the compressed JavasScript, as a string. |
---|
| 50 | * |
---|
| 51 | * see http://dean.edwards.name/packer/usage/ for more information. |
---|
| 52 | * |
---|
| 53 | * Notes : |
---|
| 54 | * # need PHP 5 . Tested with PHP 5.1.2, 5.1.3, 5.1.4, 5.2.3 |
---|
| 55 | * |
---|
| 56 | * # The packed result may be different than with the Dean Edwards |
---|
| 57 | * version, but with the same length. The reason is that the PHP |
---|
| 58 | * function usort to sort array don't necessarily preserve the |
---|
| 59 | * original order of two equal member. The Javascript sort function |
---|
| 60 | * in fact preserve this order (but that's not require by the |
---|
| 61 | * ECMAScript standard). So the encoded keywords order can be |
---|
| 62 | * different in the two results. |
---|
| 63 | * |
---|
| 64 | * # Be careful with the 'High ASCII' Level encoding if you use |
---|
| 65 | * UTF-8 in your files... |
---|
| 66 | */ |
---|
| 67 | |
---|
| 68 | |
---|
| 69 | class JavaScriptPacker { |
---|
| 70 | // constants |
---|
| 71 | const IGNORE = '$1'; |
---|
| 72 | |
---|
| 73 | // validate parameters |
---|
| 74 | private $_script = ''; |
---|
| 75 | private $_encoding = 62; |
---|
| 76 | private $_fastDecode = true; |
---|
| 77 | private $_specialChars = false; |
---|
| 78 | |
---|
| 79 | private $LITERAL_ENCODING = array( |
---|
| 80 | 'None' => 0, |
---|
| 81 | 'Numeric' => 10, |
---|
| 82 | 'Normal' => 62, |
---|
| 83 | 'High ASCII' => 95 |
---|
| 84 | ); |
---|
| 85 | |
---|
| 86 | public function __construct($_script, $_encoding = 62, $_fastDecode = true, $_specialChars = false) |
---|
| 87 | { |
---|
| 88 | $this->_script = $_script . "\n"; |
---|
| 89 | if (array_key_exists($_encoding, $this->LITERAL_ENCODING)) |
---|
| 90 | $_encoding = $this->LITERAL_ENCODING[$_encoding]; |
---|
| 91 | $this->_encoding = min((int)$_encoding, 95); |
---|
| 92 | $this->_fastDecode = $_fastDecode; |
---|
| 93 | $this->_specialChars = $_specialChars; |
---|
| 94 | } |
---|
| 95 | |
---|
| 96 | public function pack() { |
---|
| 97 | $this->_addParser('_basicCompression'); |
---|
| 98 | if ($this->_specialChars) |
---|
| 99 | $this->_addParser('_encodeSpecialChars'); |
---|
| 100 | if ($this->_encoding) |
---|
| 101 | $this->_addParser('_encodeKeywords'); |
---|
| 102 | |
---|
| 103 | // go! |
---|
| 104 | return $this->_pack($this->_script); |
---|
| 105 | } |
---|
| 106 | |
---|
| 107 | // apply all parsing routines |
---|
| 108 | private function _pack($script) { |
---|
| 109 | for ($i = 0; isset($this->_parsers[$i]); $i++) { |
---|
| 110 | $script = call_user_func(array(&$this,$this->_parsers[$i]), $script); |
---|
| 111 | } |
---|
| 112 | return $script; |
---|
| 113 | } |
---|
| 114 | |
---|
| 115 | // keep a list of parsing functions, they'll be executed all at once |
---|
| 116 | private $_parsers = array(); |
---|
| 117 | private function _addParser($parser) { |
---|
| 118 | $this->_parsers[] = $parser; |
---|
| 119 | } |
---|
| 120 | |
---|
| 121 | // zero encoding - just removal of white space and comments |
---|
| 122 | private function _basicCompression($script) { |
---|
| 123 | $parser = new ParseMaster(); |
---|
| 124 | // make safe |
---|
| 125 | $parser->escapeChar = '\\'; |
---|
| 126 | // protect strings |
---|
| 127 | $parser->add('/\'[^\'\\n\\r]*\'/', self::IGNORE); |
---|
| 128 | $parser->add('/"[^"\\n\\r]*"/', self::IGNORE); |
---|
| 129 | // remove comments |
---|
| 130 | $parser->add('/\\/\\/[^\\n\\r]*[\\n\\r]/', ' '); |
---|
| 131 | $parser->add('/\\/\\*[^*]*\\*+([^\\/][^*]*\\*+)*\\//', ' '); |
---|
| 132 | // protect regular expressions |
---|
| 133 | $parser->add('/\\s+(\\/[^\\/\\n\\r\\*][^\\/\\n\\r]*\\/g?i?)/', '$2'); // IGNORE |
---|
| 134 | $parser->add('/[^\\w\\x24\\/\'"*)\\?:]\\/[^\\/\\n\\r\\*][^\\/\\n\\r]*\\/g?i?/', self::IGNORE); |
---|
| 135 | // remove: ;;; doSomething(); |
---|
| 136 | if ($this->_specialChars) $parser->add('/;;;[^\\n\\r]+[\\n\\r]/'); |
---|
| 137 | // remove redundant semi-colons |
---|
| 138 | $parser->add('/\\(;;\\)/', self::IGNORE); // protect for (;;) loops |
---|
| 139 | $parser->add('/;+\\s*([};])/', '$2'); |
---|
| 140 | // apply the above |
---|
| 141 | $script = $parser->exec($script); |
---|
| 142 | |
---|
| 143 | // remove white-space |
---|
| 144 | $parser->add('/(\\b|\\x24)\\s+(\\b|\\x24)/', '$2 $3'); |
---|
| 145 | $parser->add('/([+\\-])\\s+([+\\-])/', '$2 $3'); |
---|
| 146 | $parser->add('/\\s+/', ''); |
---|
| 147 | // done |
---|
| 148 | return $parser->exec($script); |
---|
| 149 | } |
---|
| 150 | |
---|
| 151 | private function _encodeSpecialChars($script) { |
---|
| 152 | $parser = new ParseMaster(); |
---|
| 153 | // replace: $name -> n, $$name -> na |
---|
| 154 | $parser->add('/((\\x24+)([a-zA-Z$_]+))(\\d*)/', |
---|
| 155 | array('fn' => '_replace_name') |
---|
| 156 | ); |
---|
| 157 | // replace: _name -> _0, double-underscore (__name) is ignored |
---|
| 158 | $regexp = '/\\b_[A-Za-z\\d]\\w*/'; |
---|
| 159 | // build the word list |
---|
| 160 | $keywords = $this->_analyze($script, $regexp, '_encodePrivate'); |
---|
| 161 | // quick ref |
---|
| 162 | $encoded = $keywords['encoded']; |
---|
| 163 | |
---|
| 164 | $parser->add($regexp, |
---|
| 165 | array( |
---|
| 166 | 'fn' => '_replace_encoded', |
---|
| 167 | 'data' => $encoded |
---|
| 168 | ) |
---|
| 169 | ); |
---|
| 170 | return $parser->exec($script); |
---|
| 171 | } |
---|
| 172 | |
---|
| 173 | private function _encodeKeywords($script) { |
---|
| 174 | // escape high-ascii values already in the script (i.e. in strings) |
---|
| 175 | if ($this->_encoding > 62) |
---|
| 176 | $script = $this->_escape95($script); |
---|
| 177 | // create the parser |
---|
| 178 | $parser = new ParseMaster(); |
---|
| 179 | $encode = $this->_getEncoder($this->_encoding); |
---|
| 180 | // for high-ascii, don't encode single character low-ascii |
---|
| 181 | $regexp = ($this->_encoding > 62) ? '/\\w\\w+/' : '/\\w+/'; |
---|
| 182 | // build the word list |
---|
| 183 | $keywords = $this->_analyze($script, $regexp, $encode); |
---|
| 184 | $encoded = $keywords['encoded']; |
---|
| 185 | |
---|
| 186 | // encode |
---|
| 187 | $parser->add($regexp, |
---|
| 188 | array( |
---|
| 189 | 'fn' => '_replace_encoded', |
---|
| 190 | 'data' => $encoded |
---|
| 191 | ) |
---|
| 192 | ); |
---|
| 193 | if (empty($script)) return $script; |
---|
| 194 | else { |
---|
| 195 | //$res = $parser->exec($script); |
---|
| 196 | //$res = $this->_bootStrap($res, $keywords); |
---|
| 197 | //return $res; |
---|
| 198 | return $this->_bootStrap($parser->exec($script), $keywords); |
---|
| 199 | } |
---|
| 200 | } |
---|
| 201 | |
---|
| 202 | private function _analyze($script, $regexp, $encode) { |
---|
| 203 | // analyse |
---|
| 204 | // retreive all words in the script |
---|
| 205 | $all = array(); |
---|
| 206 | preg_match_all($regexp, $script, $all); |
---|
| 207 | $_sorted = array(); // list of words sorted by frequency |
---|
| 208 | $_encoded = array(); // dictionary of word->encoding |
---|
| 209 | $_protected = array(); // instances of "protected" words |
---|
| 210 | $all = $all[0]; // simulate the javascript comportement of global match |
---|
| 211 | if (!empty($all)) { |
---|
| 212 | $unsorted = array(); // same list, not sorted |
---|
| 213 | $protected = array(); // "protected" words (dictionary of word->"word") |
---|
| 214 | $value = array(); // dictionary of charCode->encoding (eg. 256->ff) |
---|
| 215 | $this->_count = array(); // word->count |
---|
| 216 | $i = count($all); $j = 0; //$word = null; |
---|
| 217 | // count the occurrences - used for sorting later |
---|
| 218 | do { |
---|
| 219 | --$i; |
---|
| 220 | $word = '$' . $all[$i]; |
---|
| 221 | if (!isset($this->_count[$word])) { |
---|
| 222 | $this->_count[$word] = 0; |
---|
| 223 | $unsorted[$j] = $word; |
---|
| 224 | // make a dictionary of all of the protected words in this script |
---|
| 225 | // these are words that might be mistaken for encoding |
---|
| 226 | //if (is_string($encode) && method_exists($this, $encode)) |
---|
| 227 | $values[$j] = call_user_func(array(&$this, $encode), $j); |
---|
| 228 | $protected['$' . $values[$j]] = $j++; |
---|
| 229 | } |
---|
| 230 | // increment the word counter |
---|
| 231 | $this->_count[$word]++; |
---|
| 232 | } while ($i > 0); |
---|
| 233 | // prepare to sort the word list, first we must protect |
---|
| 234 | // words that are also used as codes. we assign them a code |
---|
| 235 | // equivalent to the word itself. |
---|
| 236 | // e.g. if "do" falls within our encoding range |
---|
| 237 | // then we store keywords["do"] = "do"; |
---|
| 238 | // this avoids problems when decoding |
---|
| 239 | $i = count($unsorted); |
---|
| 240 | do { |
---|
| 241 | $word = $unsorted[--$i]; |
---|
| 242 | if (isset($protected[$word]) /*!= null*/) { |
---|
| 243 | $_sorted[$protected[$word]] = substr($word, 1); |
---|
| 244 | $_protected[$protected[$word]] = true; |
---|
| 245 | $this->_count[$word] = 0; |
---|
| 246 | } |
---|
| 247 | } while ($i); |
---|
| 248 | |
---|
| 249 | // sort the words by frequency |
---|
| 250 | // Note: the javascript and php version of sort can be different : |
---|
| 251 | // in php manual, usort : |
---|
| 252 | // " If two members compare as equal, |
---|
| 253 | // their order in the sorted array is undefined." |
---|
| 254 | // so the final packed script is different of the Dean's javascript version |
---|
| 255 | // but equivalent. |
---|
| 256 | // the ECMAscript standard does not guarantee this behaviour, |
---|
| 257 | // and thus not all browsers (e.g. Mozilla versions dating back to at |
---|
| 258 | // least 2003) respect this. |
---|
| 259 | usort($unsorted, array(&$this, '_sortWords')); |
---|
| 260 | $j = 0; |
---|
| 261 | // because there are "protected" words in the list |
---|
| 262 | // we must add the sorted words around them |
---|
| 263 | do { |
---|
| 264 | if (!isset($_sorted[$i])) |
---|
| 265 | $_sorted[$i] = substr($unsorted[$j++], 1); |
---|
| 266 | $_encoded[$_sorted[$i]] = $values[$i]; |
---|
| 267 | } while (++$i < count($unsorted)); |
---|
| 268 | } |
---|
| 269 | return array( |
---|
| 270 | 'sorted' => $_sorted, |
---|
| 271 | 'encoded' => $_encoded, |
---|
| 272 | 'protected' => $_protected); |
---|
| 273 | } |
---|
| 274 | |
---|
| 275 | private $_count = array(); |
---|
| 276 | private function _sortWords($match1, $match2) { |
---|
| 277 | return $this->_count[$match2] - $this->_count[$match1]; |
---|
| 278 | } |
---|
| 279 | |
---|
| 280 | // build the boot function used for loading and decoding |
---|
| 281 | private function _bootStrap($packed, $keywords) { |
---|
| 282 | $ENCODE = $this->_safeRegExp('$encode\\($count\\)'); |
---|
| 283 | |
---|
| 284 | // $packed: the packed script |
---|
| 285 | $packed = "'" . $this->_escape($packed) . "'"; |
---|
| 286 | |
---|
| 287 | // $ascii: base for encoding |
---|
| 288 | $ascii = min(count($keywords['sorted']), $this->_encoding); |
---|
| 289 | if ($ascii == 0) $ascii = 1; |
---|
| 290 | |
---|
| 291 | // $count: number of words contained in the script |
---|
| 292 | $count = count($keywords['sorted']); |
---|
| 293 | |
---|
| 294 | // $keywords: list of words contained in the script |
---|
| 295 | foreach ($keywords['protected'] as $i=>$value) { |
---|
| 296 | $keywords['sorted'][$i] = ''; |
---|
| 297 | } |
---|
| 298 | // convert from a string to an array |
---|
| 299 | ksort($keywords['sorted']); |
---|
| 300 | $keywords = "'" . implode('|',$keywords['sorted']) . "'.split('|')"; |
---|
| 301 | |
---|
| 302 | $encode = ($this->_encoding > 62) ? '_encode95' : $this->_getEncoder($ascii); |
---|
| 303 | $encode = $this->_getJSFunction($encode); |
---|
| 304 | $encode = preg_replace('/_encoding/','$ascii', $encode); |
---|
| 305 | $encode = preg_replace('/arguments\\.callee/','$encode', $encode); |
---|
| 306 | $inline = '\\$count' . ($ascii > 10 ? '.toString(\\$ascii)' : ''); |
---|
| 307 | |
---|
| 308 | // $decode: code snippet to speed up decoding |
---|
| 309 | if ($this->_fastDecode) { |
---|
| 310 | // create the decoder |
---|
| 311 | $decode = $this->_getJSFunction('_decodeBody'); |
---|
| 312 | if ($this->_encoding > 62) |
---|
| 313 | $decode = preg_replace('/\\\\w/', '[\\xa1-\\xff]', $decode); |
---|
| 314 | // perform the encoding inline for lower ascii values |
---|
| 315 | elseif ($ascii < 36) |
---|
| 316 | $decode = preg_replace($ENCODE, $inline, $decode); |
---|
| 317 | // special case: when $count==0 there are no keywords. I want to keep |
---|
| 318 | // the basic shape of the unpacking funcion so i'll frig the code... |
---|
| 319 | if ($count == 0) |
---|
| 320 | $decode = preg_replace($this->_safeRegExp('($count)\\s*=\\s*1'), '$1=0', $decode, 1); |
---|
| 321 | } |
---|
| 322 | |
---|
| 323 | // boot function |
---|
| 324 | $unpack = $this->_getJSFunction('_unpack'); |
---|
| 325 | if ($this->_fastDecode) { |
---|
| 326 | // insert the decoder |
---|
| 327 | $this->buffer = $decode; |
---|
| 328 | $unpack = preg_replace_callback('/\\{/', array(&$this, '_insertFastDecode'), $unpack, 1); |
---|
| 329 | } |
---|
| 330 | $unpack = preg_replace('/"/', "'", $unpack); |
---|
| 331 | if ($this->_encoding > 62) { // high-ascii |
---|
| 332 | // get rid of the word-boundaries for regexp matches |
---|
| 333 | $unpack = preg_replace('/\'\\\\\\\\b\'\s*\\+|\\+\s*\'\\\\\\\\b\'/', '', $unpack); |
---|
| 334 | } |
---|
| 335 | if ($ascii > 36 || $this->_encoding > 62 || $this->_fastDecode) { |
---|
| 336 | // insert the encode function |
---|
| 337 | $this->buffer = $encode; |
---|
| 338 | $unpack = preg_replace_callback('/\\{/', array(&$this, '_insertFastEncode'), $unpack, 1); |
---|
| 339 | } else { |
---|
| 340 | // perform the encoding inline |
---|
| 341 | $unpack = preg_replace($ENCODE, $inline, $unpack); |
---|
| 342 | } |
---|
| 343 | // pack the boot function too |
---|
| 344 | $unpackPacker = new JavaScriptPacker($unpack, 0, false, true); |
---|
| 345 | $unpack = $unpackPacker->pack(); |
---|
| 346 | |
---|
| 347 | // arguments |
---|
| 348 | $params = array($packed, $ascii, $count, $keywords); |
---|
| 349 | if ($this->_fastDecode) { |
---|
| 350 | $params[] = 0; |
---|
| 351 | $params[] = '{}'; |
---|
| 352 | } |
---|
| 353 | $params = implode(',', $params); |
---|
| 354 | |
---|
| 355 | // the whole thing |
---|
| 356 | return 'eval(' . $unpack . '(' . $params . "))\n"; |
---|
| 357 | } |
---|
| 358 | |
---|
| 359 | private $buffer; |
---|
| 360 | private function _insertFastDecode($match) { |
---|
| 361 | return '{' . $this->buffer . ';'; |
---|
| 362 | } |
---|
| 363 | private function _insertFastEncode($match) { |
---|
| 364 | return '{$encode=' . $this->buffer . ';'; |
---|
| 365 | } |
---|
| 366 | |
---|
| 367 | // mmm.. ..which one do i need ?? |
---|
| 368 | private function _getEncoder($ascii) { |
---|
| 369 | return $ascii > 10 ? $ascii > 36 ? $ascii > 62 ? |
---|
| 370 | '_encode95' : '_encode62' : '_encode36' : '_encode10'; |
---|
| 371 | } |
---|
| 372 | |
---|
| 373 | // zero encoding |
---|
| 374 | // characters: 0123456789 |
---|
| 375 | private function _encode10($charCode) { |
---|
| 376 | return $charCode; |
---|
| 377 | } |
---|
| 378 | |
---|
| 379 | // inherent base36 support |
---|
| 380 | // characters: 0123456789abcdefghijklmnopqrstuvwxyz |
---|
| 381 | private function _encode36($charCode) { |
---|
| 382 | return base_convert($charCode, 10, 36); |
---|
| 383 | } |
---|
| 384 | |
---|
| 385 | // hitch a ride on base36 and add the upper case alpha characters |
---|
| 386 | // characters: 0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ |
---|
| 387 | private function _encode62($charCode) { |
---|
| 388 | $res = ''; |
---|
| 389 | if ($charCode >= $this->_encoding) { |
---|
| 390 | $res = $this->_encode62((int)($charCode / $this->_encoding)); |
---|
| 391 | } |
---|
| 392 | $charCode = $charCode % $this->_encoding; |
---|
| 393 | |
---|
| 394 | if ($charCode > 35) |
---|
| 395 | return $res . chr($charCode + 29); |
---|
| 396 | else |
---|
| 397 | return $res . base_convert($charCode, 10, 36); |
---|
| 398 | } |
---|
| 399 | |
---|
| 400 | // use high-ascii values |
---|
| 401 | // characters: ¡¢£€¥Š§š©ª«¬Â®¯°±²³Žµ¶·ž¹º»ŒœŸ¿ÃÃÃÃÃÃ
ÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃà áâãÀåÊçÚéêëìÃîïðñòóÎõö÷ÞùúûÌÜß |
---|
| 402 | private function _encode95($charCode) { |
---|
| 403 | $res = ''; |
---|
| 404 | if ($charCode >= $this->_encoding) |
---|
| 405 | $res = $this->_encode95($charCode / $this->_encoding); |
---|
| 406 | |
---|
| 407 | return $res . chr(($charCode % $this->_encoding) + 161); |
---|
| 408 | } |
---|
| 409 | |
---|
| 410 | private function _safeRegExp($string) { |
---|
| 411 | return '/'.preg_replace('/\$/', '\\\$', $string).'/'; |
---|
| 412 | } |
---|
| 413 | |
---|
| 414 | private function _encodePrivate($charCode) { |
---|
| 415 | return "_" . $charCode; |
---|
| 416 | } |
---|
| 417 | |
---|
| 418 | // protect characters used by the parser |
---|
| 419 | private function _escape($script) { |
---|
| 420 | return preg_replace('/([\\\\\'])/', '\\\$1', $script); |
---|
| 421 | } |
---|
| 422 | |
---|
| 423 | // protect high-ascii characters already in the script |
---|
| 424 | private function _escape95($script) { |
---|
| 425 | return preg_replace_callback( |
---|
| 426 | '/[\\xa1-\\xff]/', |
---|
| 427 | array(&$this, '_escape95Bis'), |
---|
| 428 | $script |
---|
| 429 | ); |
---|
| 430 | } |
---|
| 431 | private function _escape95Bis($match) { |
---|
| 432 | return '\x'.((string)dechex(ord($match))); |
---|
| 433 | } |
---|
| 434 | |
---|
| 435 | |
---|
| 436 | private function _getJSFunction($aName) { |
---|
| 437 | if (defined('self::JSFUNCTION'.$aName)) |
---|
| 438 | return constant('self::JSFUNCTION'.$aName); |
---|
| 439 | else |
---|
| 440 | return ''; |
---|
| 441 | } |
---|
| 442 | |
---|
| 443 | // JavaScript Functions used. |
---|
| 444 | // Note : In Dean's version, these functions are converted |
---|
| 445 | // with 'String(aFunctionName);'. |
---|
| 446 | // This internal conversion complete the original code, ex : |
---|
| 447 | // 'while (aBool) anAction();' is converted to |
---|
| 448 | // 'while (aBool) { anAction(); }'. |
---|
| 449 | // The JavaScript functions below are corrected. |
---|
| 450 | |
---|
| 451 | // unpacking function - this is the boot strap function |
---|
| 452 | // data extracted from this packing routine is passed to |
---|
| 453 | // this function when decoded in the target |
---|
| 454 | // NOTE ! : without the ';' final. |
---|
| 455 | const JSFUNCTION_unpack = |
---|
| 456 | |
---|
| 457 | 'function($packed, $ascii, $count, $keywords, $encode, $decode) { |
---|
| 458 | while ($count--) { |
---|
| 459 | if ($keywords[$count]) { |
---|
| 460 | $packed = $packed.replace(new RegExp(\'\\\\b\' + $encode($count) + \'\\\\b\', \'g\'), $keywords[$count]); |
---|
| 461 | } |
---|
| 462 | } |
---|
| 463 | return $packed; |
---|
| 464 | }'; |
---|
| 465 | /* |
---|
| 466 | 'function($packed, $ascii, $count, $keywords, $encode, $decode) { |
---|
| 467 | while ($count--) |
---|
| 468 | if ($keywords[$count]) |
---|
| 469 | $packed = $packed.replace(new RegExp(\'\\\\b\' + $encode($count) + \'\\\\b\', \'g\'), $keywords[$count]); |
---|
| 470 | return $packed; |
---|
| 471 | }'; |
---|
| 472 | */ |
---|
| 473 | |
---|
| 474 | // code-snippet inserted into the unpacker to speed up decoding |
---|
| 475 | const JSFUNCTION_decodeBody = |
---|
| 476 | //_decode = function() { |
---|
| 477 | // does the browser support String.replace where the |
---|
| 478 | // replacement value is a function? |
---|
| 479 | |
---|
| 480 | ' if (!\'\'.replace(/^/, String)) { |
---|
| 481 | // decode all the values we need |
---|
| 482 | while ($count--) { |
---|
| 483 | $decode[$encode($count)] = $keywords[$count] || $encode($count); |
---|
| 484 | } |
---|
| 485 | // global replacement function |
---|
| 486 | $keywords = [function ($encoded) {return $decode[$encoded]}]; |
---|
| 487 | // generic match |
---|
| 488 | $encode = function () {return \'\\\\w+\'}; |
---|
| 489 | // reset the loop counter - we are now doing a global replace |
---|
| 490 | $count = 1; |
---|
| 491 | } |
---|
| 492 | '; |
---|
| 493 | //}; |
---|
| 494 | /* |
---|
| 495 | ' if (!\'\'.replace(/^/, String)) { |
---|
| 496 | // decode all the values we need |
---|
| 497 | while ($count--) $decode[$encode($count)] = $keywords[$count] || $encode($count); |
---|
| 498 | // global replacement function |
---|
| 499 | $keywords = [function ($encoded) {return $decode[$encoded]}]; |
---|
| 500 | // generic match |
---|
| 501 | $encode = function () {return\'\\\\w+\'}; |
---|
| 502 | // reset the loop counter - we are now doing a global replace |
---|
| 503 | $count = 1; |
---|
| 504 | }'; |
---|
| 505 | */ |
---|
| 506 | |
---|
| 507 | // zero encoding |
---|
| 508 | // characters: 0123456789 |
---|
| 509 | const JSFUNCTION_encode10 = |
---|
| 510 | 'function($charCode) { |
---|
| 511 | return $charCode; |
---|
| 512 | }';//;'; |
---|
| 513 | |
---|
| 514 | // inherent base36 support |
---|
| 515 | // characters: 0123456789abcdefghijklmnopqrstuvwxyz |
---|
| 516 | const JSFUNCTION_encode36 = |
---|
| 517 | 'function($charCode) { |
---|
| 518 | return $charCode.toString(36); |
---|
| 519 | }';//;'; |
---|
| 520 | |
---|
| 521 | // hitch a ride on base36 and add the upper case alpha characters |
---|
| 522 | // characters: 0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ |
---|
| 523 | const JSFUNCTION_encode62 = |
---|
| 524 | 'function($charCode) { |
---|
| 525 | return ($charCode < _encoding ? \'\' : arguments.callee(parseInt($charCode / _encoding))) + |
---|
| 526 | (($charCode = $charCode % _encoding) > 35 ? String.fromCharCode($charCode + 29) : $charCode.toString(36)); |
---|
| 527 | }'; |
---|
| 528 | |
---|
| 529 | // use high-ascii values |
---|
| 530 | // characters: ¡¢£€¥Š§š©ª«¬Â®¯°±²³Žµ¶·ž¹º»ŒœŸ¿ÃÃÃÃÃÃ
ÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃÃà áâãÀåÊçÚéêëìÃîïðñòóÎõö÷ÞùúûÌÜß |
---|
| 531 | const JSFUNCTION_encode95 = |
---|
| 532 | 'function($charCode) { |
---|
| 533 | return ($charCode < _encoding ? \'\' : arguments.callee($charCode / _encoding)) + |
---|
| 534 | String.fromCharCode($charCode % _encoding + 161); |
---|
| 535 | }'; |
---|
| 536 | |
---|
| 537 | } |
---|
| 538 | |
---|
| 539 | |
---|
| 540 | class ParseMaster { |
---|
| 541 | public $ignoreCase = false; |
---|
| 542 | public $escapeChar = ''; |
---|
| 543 | |
---|
| 544 | // constants |
---|
| 545 | const EXPRESSION = 0; |
---|
| 546 | const REPLACEMENT = 1; |
---|
| 547 | const LENGTH = 2; |
---|
| 548 | |
---|
| 549 | // used to determine nesting levels |
---|
| 550 | private $GROUPS = '/\\(/';//g |
---|
| 551 | private $SUB_REPLACE = '/\\$\\d/'; |
---|
| 552 | private $INDEXED = '/^\\$\\d+$/'; |
---|
| 553 | private $TRIM = '/([\'"])\\1\\.(.*)\\.\\1\\1$/'; |
---|
| 554 | private $ESCAPE = '/\\\./';//g |
---|
| 555 | private $QUOTE = '/\'/'; |
---|
| 556 | private $DELETED = '/\\x01[^\\x01]*\\x01/';//g |
---|
| 557 | |
---|
| 558 | public function add($expression, $replacement = '') { |
---|
| 559 | // count the number of sub-expressions |
---|
| 560 | // - add one because each pattern is itself a sub-expression |
---|
| 561 | $length = 1 + preg_match_all($this->GROUPS, $this->_internalEscape((string)$expression), $out); |
---|
| 562 | |
---|
| 563 | // treat only strings $replacement |
---|
| 564 | if (is_string($replacement)) { |
---|
| 565 | // does the pattern deal with sub-expressions? |
---|
| 566 | if (preg_match($this->SUB_REPLACE, $replacement)) { |
---|
| 567 | // a simple lookup? (e.g. "$2") |
---|
| 568 | if (preg_match($this->INDEXED, $replacement)) { |
---|
| 569 | // store the index (used for fast retrieval of matched strings) |
---|
| 570 | $replacement = (int)(substr($replacement, 1)) - 1; |
---|
| 571 | } else { // a complicated lookup (e.g. "Hello $2 $1") |
---|
| 572 | // build a function to do the lookup |
---|
| 573 | $quote = preg_match($this->QUOTE, $this->_internalEscape($replacement)) |
---|
| 574 | ? '"' : "'"; |
---|
| 575 | $replacement = array( |
---|
| 576 | 'fn' => '_backReferences', |
---|
| 577 | 'data' => array( |
---|
| 578 | 'replacement' => $replacement, |
---|
| 579 | 'length' => $length, |
---|
| 580 | 'quote' => $quote |
---|
| 581 | ) |
---|
| 582 | ); |
---|
| 583 | } |
---|
| 584 | } |
---|
| 585 | } |
---|
| 586 | // pass the modified arguments |
---|
| 587 | if (!empty($expression)) $this->_add($expression, $replacement, $length); |
---|
| 588 | else $this->_add('/^$/', $replacement, $length); |
---|
| 589 | } |
---|
| 590 | |
---|
| 591 | public function exec($string) { |
---|
| 592 | // execute the global replacement |
---|
| 593 | $this->_escaped = array(); |
---|
| 594 | |
---|
| 595 | // simulate the _patterns.toSTring of Dean |
---|
| 596 | $regexp = '/'; |
---|
| 597 | foreach ($this->_patterns as $reg) { |
---|
| 598 | $regexp .= '(' . substr($reg[self::EXPRESSION], 1, -1) . ')|'; |
---|
| 599 | } |
---|
| 600 | $regexp = substr($regexp, 0, -1) . '/'; |
---|
| 601 | $regexp .= ($this->ignoreCase) ? 'i' : ''; |
---|
| 602 | |
---|
| 603 | $string = $this->_escape($string, $this->escapeChar); |
---|
| 604 | $string = preg_replace_callback( |
---|
| 605 | $regexp, |
---|
| 606 | array( |
---|
| 607 | &$this, |
---|
| 608 | '_replacement' |
---|
| 609 | ), |
---|
| 610 | $string |
---|
| 611 | ); |
---|
| 612 | $string = $this->_unescape($string, $this->escapeChar); |
---|
| 613 | |
---|
| 614 | return preg_replace($this->DELETED, '', $string); |
---|
| 615 | } |
---|
| 616 | |
---|
| 617 | public function reset() { |
---|
| 618 | // clear the patterns collection so that this object may be re-used |
---|
| 619 | $this->_patterns = array(); |
---|
| 620 | } |
---|
| 621 | |
---|
| 622 | // private |
---|
| 623 | private $_escaped = array(); // escaped characters |
---|
| 624 | private $_patterns = array(); // patterns stored by index |
---|
| 625 | |
---|
| 626 | // create and add a new pattern to the patterns collection |
---|
| 627 | private function _add() { |
---|
| 628 | $arguments = func_get_args(); |
---|
| 629 | $this->_patterns[] = $arguments; |
---|
| 630 | } |
---|
| 631 | |
---|
| 632 | // this is the global replace function (it's quite complicated) |
---|
| 633 | private function _replacement($arguments) { |
---|
| 634 | if (empty($arguments)) return ''; |
---|
| 635 | |
---|
| 636 | $i = 1; $j = 0; |
---|
| 637 | // loop through the patterns |
---|
| 638 | while (isset($this->_patterns[$j])) { |
---|
| 639 | $pattern = $this->_patterns[$j++]; |
---|
| 640 | // do we have a result? |
---|
| 641 | if (isset($arguments[$i]) && ($arguments[$i] != '')) { |
---|
| 642 | $replacement = $pattern[self::REPLACEMENT]; |
---|
| 643 | |
---|
| 644 | if (is_array($replacement) && isset($replacement['fn'])) { |
---|
| 645 | |
---|
| 646 | if (isset($replacement['data'])) $this->buffer = $replacement['data']; |
---|
| 647 | return call_user_func(array(&$this, $replacement['fn']), $arguments, $i); |
---|
| 648 | |
---|
| 649 | } elseif (is_int($replacement)) { |
---|
| 650 | return $arguments[$replacement + $i]; |
---|
| 651 | |
---|
| 652 | } |
---|
| 653 | $delete = ($this->escapeChar == '' || |
---|
| 654 | strpos($arguments[$i], $this->escapeChar) === false) |
---|
| 655 | ? '' : "\x01" . $arguments[$i] . "\x01"; |
---|
| 656 | return $delete . $replacement; |
---|
| 657 | |
---|
| 658 | // skip over references to sub-expressions |
---|
| 659 | } else { |
---|
| 660 | $i += $pattern[self::LENGTH]; |
---|
| 661 | } |
---|
| 662 | } |
---|
| 663 | } |
---|
| 664 | |
---|
| 665 | private function _backReferences($match, $offset) { |
---|
| 666 | $replacement = $this->buffer['replacement']; |
---|
| 667 | $quote = $this->buffer['quote']; |
---|
| 668 | $i = $this->buffer['length']; |
---|
| 669 | while ($i) { |
---|
| 670 | $replacement = str_replace('$'.$i--, $match[$offset + $i], $replacement); |
---|
| 671 | } |
---|
| 672 | return $replacement; |
---|
| 673 | } |
---|
| 674 | |
---|
| 675 | private function _replace_name($match, $offset){ |
---|
| 676 | $length = strlen($match[$offset + 2]); |
---|
| 677 | $start = $length - max($length - strlen($match[$offset + 3]), 0); |
---|
| 678 | return substr($match[$offset + 1], $start, $length) . $match[$offset + 4]; |
---|
| 679 | } |
---|
| 680 | |
---|
| 681 | private function _replace_encoded($match, $offset) { |
---|
| 682 | return $this->buffer[$match[$offset]]; |
---|
| 683 | } |
---|
| 684 | |
---|
| 685 | |
---|
| 686 | // php : we cannot pass additional data to preg_replace_callback, |
---|
| 687 | // and we cannot use &$this in create_function, so let's go to lower level |
---|
| 688 | private $buffer; |
---|
| 689 | |
---|
| 690 | // encode escaped characters |
---|
| 691 | private function _escape($string, $escapeChar) { |
---|
| 692 | if ($escapeChar) { |
---|
| 693 | $this->buffer = $escapeChar; |
---|
| 694 | return preg_replace_callback( |
---|
| 695 | '/\\' . $escapeChar . '(.)' .'/', |
---|
| 696 | array(&$this, '_escapeBis'), |
---|
| 697 | $string |
---|
| 698 | ); |
---|
| 699 | |
---|
| 700 | } else { |
---|
| 701 | return $string; |
---|
| 702 | } |
---|
| 703 | } |
---|
| 704 | private function _escapeBis($match) { |
---|
| 705 | $this->_escaped[] = $match[1]; |
---|
| 706 | return $this->buffer; |
---|
| 707 | } |
---|
| 708 | |
---|
| 709 | // decode escaped characters |
---|
| 710 | private function _unescape($string, $escapeChar) { |
---|
| 711 | if ($escapeChar) { |
---|
| 712 | $regexp = '/'.'\\'.$escapeChar.'/'; |
---|
| 713 | $this->buffer = array('escapeChar'=> $escapeChar, 'i' => 0); |
---|
| 714 | return preg_replace_callback |
---|
| 715 | ( |
---|
| 716 | $regexp, |
---|
| 717 | array(&$this, '_unescapeBis'), |
---|
| 718 | $string |
---|
| 719 | ); |
---|
| 720 | |
---|
| 721 | } else { |
---|
| 722 | return $string; |
---|
| 723 | } |
---|
| 724 | } |
---|
| 725 | private function _unescapeBis() { |
---|
| 726 | if (isset($this->_escaped[$this->buffer['i']]) |
---|
| 727 | && $this->_escaped[$this->buffer['i']] != '') |
---|
| 728 | { |
---|
| 729 | $temp = $this->_escaped[$this->buffer['i']]; |
---|
| 730 | } else { |
---|
| 731 | $temp = ''; |
---|
| 732 | } |
---|
| 733 | $this->buffer['i']++; |
---|
| 734 | return $this->buffer['escapeChar'] . $temp; |
---|
| 735 | } |
---|
| 736 | |
---|
| 737 | private function _internalEscape($string) { |
---|
| 738 | return preg_replace($this->ESCAPE, '', $string); |
---|
| 739 | } |
---|
| 740 | } |
---|
| 741 | ?> |
---|