| 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 |
namespace float_menu_free; |
| 69 |
|
| 70 |
class JavaScriptPacker |
| 71 |
{ |
| 72 |
// constants |
| 73 |
const IGNORE = '$1'; |
| 74 |
// validate parameters |
| 75 |
const JSFUNCTION_unpack = |
| 76 |
'function($packed, $ascii, $count, $keywords, $encode, $decode) { |
| 77 |
while ($count--) { |
| 78 |
if ($keywords[$count]) { |
| 79 |
$packed = $packed.replace(new RegExp(\'\\\\b\' + $encode($count) + \'\\\\b\', \'g\'), $keywords[$count]); |
| 80 |
} |
| 81 |
} |
| 82 |
return $packed; |
| 83 |
}'; |
| 84 |
const JSFUNCTION_decodeBody = |
| 85 |
//_decode = function() { |
| 86 |
// does the browser support String.replace where the |
| 87 |
// replacement value is a function? |
| 88 |
' if (!\'\'.replace(/^/, String)) { |
| 89 |
// decode all the values we need |
| 90 |
while ($count--) { |
| 91 |
$decode[$encode($count)] = $keywords[$count] || $encode($count); |
| 92 |
} |
| 93 |
// global replacement function |
| 94 |
$keywords = [function ($encoded) {return $decode[$encoded]}]; |
| 95 |
// generic match |
| 96 |
$encode = function () {return \'\\\\w+\'}; |
| 97 |
// reset the loop counter - we are now doing a global replace |
| 98 |
$count = 1; |
| 99 |
} |
| 100 |
'; |
| 101 |
const JSFUNCTION_encode10 = |
| 102 |
'function($charCode) { |
| 103 |
return $charCode; |
| 104 |
}'; |
| 105 |
const JSFUNCTION_encode36 = |
| 106 |
'function($charCode) { |
| 107 |
return $charCode.toString(36); |
| 108 |
}'; |
| 109 |
const JSFUNCTION_encode62 = |
| 110 |
'function($charCode) { |
| 111 |
return ($charCode < _encoding ? \'\' : arguments.callee(parseInt($charCode / _encoding))) + |
| 112 |
(($charCode = $charCode % _encoding) > 35 ? String.fromCharCode($charCode + 29) : $charCode.toString(36)); |
| 113 |
}'; |
| 114 |
const JSFUNCTION_encode95 = |
| 115 |
'function($charCode) { |
| 116 |
return ($charCode < _encoding ? \'\' : arguments.callee($charCode / _encoding)) + |
| 117 |
String.fromCharCode($charCode % _encoding + 161); |
| 118 |
}'; |
| 119 |
private $_script = ''; |
| 120 |
|
| 121 |
// apply all parsing routines |
| 122 |
private $_encoding = 62; |
| 123 |
|
| 124 |
// keep a list of parsing functions, they'll be executed all at once |
| 125 |
private $_fastDecode = true; |
| 126 |
private $_specialChars = false; |
| 127 |
|
| 128 |
// zero encoding - just removal of white space and comments |
| 129 |
private $LITERAL_ENCODING = array( |
| 130 |
'None' => 0, |
| 131 |
'Numeric' => 10, |
| 132 |
'Normal' => 62, |
| 133 |
'High ASCII' => 95, |
| 134 |
); |
| 135 |
private $_parsers = array(); |
| 136 |
private $_count = array(); |
| 137 |
private $buffer; |
| 138 |
|
| 139 |
public function __construct( $_script, $_encoding = 62, $_fastDecode = true, $_specialChars = false ) |
| 140 |
{ |
| 141 |
$this->_script = $_script . "\n"; |
| 142 |
if ( array_key_exists( $_encoding, $this->LITERAL_ENCODING ) ) |
| 143 |
$_encoding = $this->LITERAL_ENCODING[ $_encoding ]; |
| 144 |
$this->_encoding = min( (int)$_encoding, 95 ); |
| 145 |
$this->_fastDecode = $_fastDecode; |
| 146 |
$this->_specialChars = $_specialChars; |
| 147 |
} |
| 148 |
|
| 149 |
private function _basicCompression( $script ) |
| 150 |
{ |
| 151 |
$parser = new ParseMaster(); |
| 152 |
// make safe |
| 153 |
$parser->escapeChar = '\\'; |
| 154 |
// protect strings |
| 155 |
$parser->add( '/\'[^\'\\n\\r]*\'/', self::IGNORE ); |
| 156 |
$parser->add( '/"[^"\\n\\r]*"/', self::IGNORE ); |
| 157 |
// remove comments |
| 158 |
$parser->add( '/\\/\\/[^\\n\\r]*[\\n\\r]/', ' ' ); |
| 159 |
$parser->add( '/\\/\\*[^*]*\\*+([^\\/][^*]*\\*+)*\\//', ' ' ); |
| 160 |
// protect regular expressions |
| 161 |
$parser->add( '/\\s+(\\/[^\\/\\n\\r\\*][^\\/\\n\\r]*\\/g?i?)/', '$2' ); // IGNORE |
| 162 |
$parser->add( '/[^\\w\\x24\\/\'"*)\\?:]\\/[^\\/\\n\\r\\*][^\\/\\n\\r]*\\/g?i?/', self::IGNORE ); |
| 163 |
// remove: ;;; doSomething(); |
| 164 |
if ( $this->_specialChars ) $parser->add( '/;;;[^\\n\\r]+[\\n\\r]/' ); |
| 165 |
// remove redundant semi-colons |
| 166 |
$parser->add( '/\\(;;\\)/', self::IGNORE ); // protect for (;;) loops |
| 167 |
$parser->add( '/;+\\s*([};])/', '$2' ); |
| 168 |
// apply the above |
| 169 |
$script = $parser->exec( $script ); |
| 170 |
// remove white-space |
| 171 |
$parser->add( '/(\\b|\\x24)\\s+(\\b|\\x24)/', '$2 $3' ); |
| 172 |
$parser->add( '/([+\\-])\\s+([+\\-])/', '$2 $3' ); |
| 173 |
$parser->add( '/\\s+/', '' ); |
| 174 |
// done |
| 175 |
return $parser->exec( $script ); |
| 176 |
} |
| 177 |
|
| 178 |
// build the boot function used for loading and decoding |
| 179 |
|
| 180 |
private function _encodeSpecialChars( $script ) |
| 181 |
{ |
| 182 |
$parser = new ParseMaster(); |
| 183 |
// replace: $name -> n, $$name -> na |
| 184 |
$parser->add( '/((\\x24+)([a-zA-Z$_]+))(\\d*)/', |
| 185 |
array( 'fn' => '_replace_name' ) |
| 186 |
); |
| 187 |
// replace: _name -> _0, double-underscore (__name) is ignored |
| 188 |
$regexp = '/\\b_[A-Za-z\\d]\\w*/'; |
| 189 |
// build the word list |
| 190 |
$keywords = $this->_analyze( $script, $regexp, '_encodePrivate' ); |
| 191 |
// quick ref |
| 192 |
$encoded = $keywords[ 'encoded' ]; |
| 193 |
$parser->add( $regexp, |
| 194 |
array( |
| 195 |
'fn' => '_replace_encoded', |
| 196 |
'data' => $encoded, |
| 197 |
) |
| 198 |
); |
| 199 |
return $parser->exec( $script ); |
| 200 |
} |
| 201 |
|
| 202 |
private function _analyze( $script, $regexp, $encode ) |
| 203 |
{ |
| 204 |
// analyse |
| 205 |
// retreive all words in the script |
| 206 |
$all = array(); |
| 207 |
preg_match_all( $regexp, $script, $all ); |
| 208 |
$_sorted = array(); // list of words sorted by frequency |
| 209 |
$_encoded = array(); // dictionary of word->encoding |
| 210 |
$_protected = array(); // instances of "protected" words |
| 211 |
$all = $all[ 0 ]; // simulate the javascript comportement of global match |
| 212 |
if ( !empty( $all ) ) { |
| 213 |
$unsorted = array(); // same list, not sorted |
| 214 |
$protected = array(); // "protected" words (dictionary of word->"word") |
| 215 |
$value = array(); // dictionary of charCode->encoding (eg. 256->ff) |
| 216 |
$this->_count = array(); // word->count |
| 217 |
$i = count( $all ); |
| 218 |
$j = 0; //$word = null; |
| 219 |
// count the occurrences - used for sorting later |
| 220 |
do { |
| 221 |
--$i; |
| 222 |
$word = '$' . $all[ $i ]; |
| 223 |
if ( !isset( $this->_count[ $word ] ) ) { |
| 224 |
$this->_count[ $word ] = 0; |
| 225 |
$unsorted[ $j ] = $word; |
| 226 |
// make a dictionary of all of the protected words in this script |
| 227 |
// these are words that might be mistaken for encoding |
| 228 |
//if (is_string($encode) && method_exists($this, $encode)) |
| 229 |
$values[ $j ] = call_user_func( array( &$this, $encode ), $j ); |
| 230 |
$protected[ '$' . $values[ $j ] ] = $j++; |
| 231 |
} |
| 232 |
// increment the word counter |
| 233 |
$this->_count[ $word ]++; |
| 234 |
} while ( $i > 0 ); |
| 235 |
// prepare to sort the word list, first we must protect |
| 236 |
// words that are also used as codes. we assign them a code |
| 237 |
// equivalent to the word itself. |
| 238 |
// e.g. if "do" falls within our encoding range |
| 239 |
// then we store keywords["do"] = "do"; |
| 240 |
// this avoids problems when decoding |
| 241 |
$i = count( $unsorted ); |
| 242 |
do { |
| 243 |
$word = $unsorted[ --$i ]; |
| 244 |
if ( isset( $protected[ $word ] ) /*!= null*/ ) { |
| 245 |
$_sorted[ $protected[ $word ] ] = substr( $word, 1 ); |
| 246 |
$_protected[ $protected[ $word ] ] = true; |
| 247 |
$this->_count[ $word ] = 0; |
| 248 |
} |
| 249 |
} while ( $i ); |
| 250 |
// sort the words by frequency |
| 251 |
// Note: the javascript and php version of sort can be different : |
| 252 |
// in php manual, usort : |
| 253 |
// " If two members compare as equal, |
| 254 |
// their order in the sorted array is undefined." |
| 255 |
// so the final packed script is different of the Dean's javascript version |
| 256 |
// but equivalent. |
| 257 |
// the ECMAscript standard does not guarantee this behaviour, |
| 258 |
// and thus not all browsers (e.g. Mozilla versions dating back to at |
| 259 |
// least 2003) respect this. |
| 260 |
usort( $unsorted, array( &$this, '_sortWords' ) ); |
| 261 |
$j = 0; |
| 262 |
// because there are "protected" words in the list |
| 263 |
// we must add the sorted words around them |
| 264 |
do { |
| 265 |
if ( !isset( $_sorted[ $i ] ) ) |
| 266 |
$_sorted[ $i ] = substr( $unsorted[ $j++ ], 1 ); |
| 267 |
$_encoded[ $_sorted[ $i ] ] = $values[ $i ]; |
| 268 |
} while ( ++$i < count( $unsorted ) ); |
| 269 |
} |
| 270 |
return array( |
| 271 |
'sorted' => $_sorted, |
| 272 |
'encoded' => $_encoded, |
| 273 |
'protected' => $_protected ); |
| 274 |
} |
| 275 |
|
| 276 |
private function _encodeKeywords( $script ) |
| 277 |
{ |
| 278 |
// escape high-ascii values already in the script (i.e. in strings) |
| 279 |
if ( $this->_encoding > 62 ) |
| 280 |
$script = $this->_escape95( $script ); |
| 281 |
// create the parser |
| 282 |
$parser = new ParseMaster(); |
| 283 |
$encode = $this->_getEncoder( $this->_encoding ); |
| 284 |
// for high-ascii, don't encode single character low-ascii |
| 285 |
$regexp = ($this->_encoding > 62) ? '/\\w\\w+/' : '/\\w+/'; |
| 286 |
// build the word list |
| 287 |
$keywords = $this->_analyze( $script, $regexp, $encode ); |
| 288 |
$encoded = $keywords[ 'encoded' ]; |
| 289 |
// encode |
| 290 |
$parser->add( $regexp, |
| 291 |
array( |
| 292 |
'fn' => '_replace_encoded', |
| 293 |
'data' => $encoded, |
| 294 |
) |
| 295 |
); |
| 296 |
if ( empty( $script ) ) return $script; |
| 297 |
else { |
| 298 |
//$res = $parser->exec($script); |
| 299 |
//$res = $this->_bootStrap($res, $keywords); |
| 300 |
//return $res; |
| 301 |
return $this->_bootStrap( $parser->exec( $script ), $keywords ); |
| 302 |
} |
| 303 |
} |
| 304 |
|
| 305 |
private function _escape95( $script ) |
| 306 |
{ |
| 307 |
return preg_replace_callback( |
| 308 |
'/[\\xa1-\\xff]/', |
| 309 |
array( &$this, '_escape95Bis' ), |
| 310 |
$script |
| 311 |
); |
| 312 |
} |
| 313 |
|
| 314 |
// mmm.. ..which one do i need ?? |
| 315 |
|
| 316 |
private function _getEncoder( $ascii ) |
| 317 |
{ |
| 318 |
return $ascii > 10 ? $ascii > 36 ? $ascii > 62 ? |
| 319 |
'_encode95' : '_encode62' : '_encode36' : '_encode10'; |
| 320 |
} |
| 321 |
// zero encoding |
| 322 |
// characters: 0123456789 |
| 323 |
|
| 324 |
private function _bootStrap( $packed, $keywords ) |
| 325 |
{ |
| 326 |
$ENCODE = $this->_safeRegExp( '$encode\\($count\\)' ); |
| 327 |
// $packed: the packed script |
| 328 |
$packed = "'" . $this->_escape( $packed ) . "'"; |
| 329 |
// $ascii: base for encoding |
| 330 |
$ascii = min( count( $keywords[ 'sorted' ] ), $this->_encoding ); |
| 331 |
if ( $ascii == 0 ) $ascii = 1; |
| 332 |
// $count: number of words contained in the script |
| 333 |
$count = count( $keywords[ 'sorted' ] ); |
| 334 |
// $keywords: list of words contained in the script |
| 335 |
foreach ( $keywords[ 'protected' ] as $i => $value ) { |
| 336 |
$keywords[ 'sorted' ][ $i ] = ''; |
| 337 |
} |
| 338 |
// convert from a string to an array |
| 339 |
ksort( $keywords[ 'sorted' ] ); |
| 340 |
$keywords = "'" . implode( '|', $keywords[ 'sorted' ] ) . "'.split('|')"; |
| 341 |
$encode = ($this->_encoding > 62) ? '_encode95' : $this->_getEncoder( $ascii ); |
| 342 |
$encode = $this->_getJSFunction( $encode ); |
| 343 |
$encode = preg_replace( '/_encoding/', '$ascii', $encode ); |
| 344 |
$encode = preg_replace( '/arguments\\.callee/', '$encode', $encode ); |
| 345 |
$inline = '\\$count' . ($ascii > 10 ? '.toString(\\$ascii)' : ''); |
| 346 |
// $decode: code snippet to speed up decoding |
| 347 |
if ( $this->_fastDecode ) { |
| 348 |
// create the decoder |
| 349 |
$decode = $this->_getJSFunction( '_decodeBody' ); |
| 350 |
if ( $this->_encoding > 62 ) |
| 351 |
$decode = preg_replace( '/\\\\w/', '[\\xa1-\\xff]', $decode ); |
| 352 |
// perform the encoding inline for lower ascii values |
| 353 |
elseif ( $ascii < 36 ) |
| 354 |
$decode = preg_replace( $ENCODE, $inline, $decode ); |
| 355 |
// special case: when $count==0 there are no keywords. I want to keep |
| 356 |
// the basic shape of the unpacking funcion so i'll frig the code... |
| 357 |
if ( $count == 0 ) |
| 358 |
$decode = preg_replace( $this->_safeRegExp( '($count)\\s*=\\s*1' ), '$1=0', $decode, 1 ); |
| 359 |
} |
| 360 |
// boot function |
| 361 |
$unpack = $this->_getJSFunction( '_unpack' ); |
| 362 |
if ( $this->_fastDecode ) { |
| 363 |
// insert the decoder |
| 364 |
$this->buffer = $decode; |
| 365 |
$unpack = preg_replace_callback( '/\\{/', array( &$this, '_insertFastDecode' ), $unpack, 1 ); |
| 366 |
} |
| 367 |
$unpack = preg_replace( '/"/', "'", $unpack ); |
| 368 |
if ( $this->_encoding > 62 ) { // high-ascii |
| 369 |
// get rid of the word-boundaries for regexp matches |
| 370 |
$unpack = preg_replace( '/\'\\\\\\\\b\'\s*\\+|\\+\s*\'\\\\\\\\b\'/', '', $unpack ); |
| 371 |
} |
| 372 |
if ( $ascii > 36 || $this->_encoding > 62 || $this->_fastDecode ) { |
| 373 |
// insert the encode function |
| 374 |
$this->buffer = $encode; |
| 375 |
$unpack = preg_replace_callback( '/\\{/', array( &$this, '_insertFastEncode' ), $unpack, 1 ); |
| 376 |
} else { |
| 377 |
// perform the encoding inline |
| 378 |
$unpack = preg_replace( $ENCODE, $inline, $unpack ); |
| 379 |
} |
| 380 |
// pack the boot function too |
| 381 |
$unpackPacker = new JavaScriptPacker( $unpack, 0, false, true ); |
| 382 |
$unpack = $unpackPacker->pack(); |
| 383 |
// arguments |
| 384 |
$params = array( $packed, $ascii, $count, $keywords ); |
| 385 |
if ( $this->_fastDecode ) { |
| 386 |
$params[] = 0; |
| 387 |
$params[] = '{}'; |
| 388 |
} |
| 389 |
$params = implode( ',', $params ); |
| 390 |
// the whole thing |
| 391 |
return 'eval(' . $unpack . '(' . $params . "))\n"; |
| 392 |
} |
| 393 |
// inherent base36 support |
| 394 |
// characters: 0123456789abcdefghijklmnopqrstuvwxyz |
| 395 |
|
| 396 |
private function _safeRegExp( $string ) |
| 397 |
{ |
| 398 |
return '/' . preg_replace( '/\$/', '\\\$', $string ) . '/'; |
| 399 |
} |
| 400 |
// hitch a ride on base36 and add the upper case alpha characters |
| 401 |
// characters: 0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ |
| 402 |
|
| 403 |
private function _escape( $script ) |
| 404 |
{ |
| 405 |
return preg_replace( '/([\\\\\'])/', '\\\$1', $script ); |
| 406 |
} |
| 407 |
// use high-ascii values |
| 408 |
// characters: ¡¢£¤¥¦§¨©ª«¬®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄ� |
| 409 |
ÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþ |
| 410 |
|
| 411 |
private function _getJSFunction( $aName ) |
| 412 |
{ |
| 413 |
if ( defined( 'self::JSFUNCTION' . $aName ) ) |
| 414 |
return constant( 'self::JSFUNCTION' . $aName ); |
| 415 |
else |
| 416 |
return ''; |
| 417 |
} |
| 418 |
|
| 419 |
public function pack() |
| 420 |
{ |
| 421 |
$this->_addParser( '_basicCompression' ); |
| 422 |
if ( $this->_specialChars ) |
| 423 |
$this->_addParser( '_encodeSpecialChars' ); |
| 424 |
if ( $this->_encoding ) |
| 425 |
$this->_addParser( '_encodeKeywords' ); |
| 426 |
// go! |
| 427 |
return $this->_pack( $this->_script ); |
| 428 |
} |
| 429 |
|
| 430 |
private function _addParser( $parser ) |
| 431 |
{ |
| 432 |
$this->_parsers[] = $parser; |
| 433 |
} |
| 434 |
|
| 435 |
// protect characters used by the parser |
| 436 |
|
| 437 |
private function _pack( $script ) |
| 438 |
{ |
| 439 |
for ( $i = 0; isset( $this->_parsers[ $i ] ); $i++ ) { |
| 440 |
$script = call_user_func( array( &$this, $this->_parsers[ $i ] ), $script ); |
| 441 |
} |
| 442 |
return $script; |
| 443 |
} |
| 444 |
|
| 445 |
// protect high-ascii characters already in the script |
| 446 |
|
| 447 |
private function _sortWords( $match1, $match2 ) |
| 448 |
{ |
| 449 |
return $this->_count[ $match2 ] - $this->_count[ $match1 ]; |
| 450 |
} |
| 451 |
|
| 452 |
private function _insertFastDecode( $match ) |
| 453 |
{ |
| 454 |
return '{' . $this->buffer . ';'; |
| 455 |
} |
| 456 |
|
| 457 |
private function _insertFastEncode( $match ) |
| 458 |
{ |
| 459 |
return '{$encode=' . $this->buffer . ';'; |
| 460 |
} |
| 461 |
// JavaScript Functions used. |
| 462 |
// Note : In Dean's version, these functions are converted |
| 463 |
// with 'String(aFunctionName);'. |
| 464 |
// This internal conversion complete the original code, ex : |
| 465 |
// 'while (aBool) anAction();' is converted to |
| 466 |
// 'while (aBool) { anAction(); }'. |
| 467 |
// The JavaScript functions below are corrected. |
| 468 |
// unpacking function - this is the boot strap function |
| 469 |
// data extracted from this packing routine is passed to |
| 470 |
// this function when decoded in the target |
| 471 |
// NOTE ! : without the ';' final. |
| 472 |
|
| 473 |
private function _encode10( $charCode ) |
| 474 |
{ |
| 475 |
return $charCode; |
| 476 |
} |
| 477 |
/* |
| 478 |
'function($packed, $ascii, $count, $keywords, $encode, $decode) { |
| 479 |
while ($count--) |
| 480 |
if ($keywords[$count]) |
| 481 |
$packed = $packed.replace(new RegExp(\'\\\\b\' + $encode($count) + \'\\\\b\', \'g\'), $keywords[$count]); |
| 482 |
return $packed; |
| 483 |
}'; |
| 484 |
*/ |
| 485 |
// code-snippet inserted into the unpacker to speed up decoding |
| 486 |
|
| 487 |
private function _encode36( $charCode ) |
| 488 |
{ |
| 489 |
return base_convert( $charCode, 10, 36 ); |
| 490 |
} |
| 491 |
//}; |
| 492 |
/* |
| 493 |
' if (!\'\'.replace(/^/, String)) { |
| 494 |
// decode all the values we need |
| 495 |
while ($count--) $decode[$encode($count)] = $keywords[$count] || $encode($count); |
| 496 |
// global replacement function |
| 497 |
$keywords = [function ($encoded) {return $decode[$encoded]}]; |
| 498 |
// generic match |
| 499 |
$encode = function () {return\'\\\\w+\'}; |
| 500 |
// reset the loop counter - we are now doing a global replace |
| 501 |
$count = 1; |
| 502 |
}'; |
| 503 |
*/ |
| 504 |
// zero encoding |
| 505 |
// characters: 0123456789 |
| 506 |
|
| 507 |
private function _encode62( $charCode ) |
| 508 |
{ |
| 509 |
$res = ''; |
| 510 |
if ( $charCode >= $this->_encoding ) { |
| 511 |
$res = $this->_encode62( (int)($charCode / $this->_encoding) ); |
| 512 |
} |
| 513 |
$charCode = $charCode % $this->_encoding; |
| 514 |
if ( $charCode > 35 ) |
| 515 |
return $res . chr( $charCode + 29 ); |
| 516 |
else |
| 517 |
return $res . base_convert( $charCode, 10, 36 ); |
| 518 |
}//;'; |
| 519 |
// inherent base36 support |
| 520 |
// characters: 0123456789abcdefghijklmnopqrstuvwxyz |
| 521 |
|
| 522 |
private function _encode95( $charCode ) |
| 523 |
{ |
| 524 |
$res = ''; |
| 525 |
if ( $charCode >= $this->_encoding ) |
| 526 |
$res = $this->_encode95( $charCode / $this->_encoding ); |
| 527 |
return $res . chr( ($charCode % $this->_encoding) + 161 ); |
| 528 |
}//;'; |
| 529 |
// hitch a ride on base36 and add the upper case alpha characters |
| 530 |
// characters: 0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ |
| 531 |
|
| 532 |
private function _encodePrivate( $charCode ) |
| 533 |
{ |
| 534 |
return "_" . $charCode; |
| 535 |
} |
| 536 |
// use high-ascii values |
| 537 |
// characters: ¡¢£¤¥¦§¨©ª«¬®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄ� |
| 538 |
ÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþ |
| 539 |
|
| 540 |
private function _escape95Bis( $match ) |
| 541 |
{ |
| 542 |
return '\x' . ((string)dechex( ord( $match ) )); |
| 543 |
} |
| 544 |
} |
| 545 |
|
| 546 |
class ParseMaster |
| 547 |
{ |
| 548 |
const EXPRESSION = 0; |
| 549 |
const REPLACEMENT = 1; |
| 550 |
// constants |
| 551 |
const LENGTH = 2; |
| 552 |
public $ignoreCase = false; |
| 553 |
public $escapeChar = ''; |
| 554 |
// used to determine nesting levels |
| 555 |
private $GROUPS = '/\\(/';//g |
| 556 |
private $SUB_REPLACE = '/\\$\\d/'; |
| 557 |
private $INDEXED = '/^\\$\\d+$/'; |
| 558 |
private $TRIM = '/([\'"])\\1\\.(.*)\\.\\1\\1$/'; |
| 559 |
private $ESCAPE = '/\\\./';//g |
| 560 |
private $QUOTE = '/\'/'; |
| 561 |
private $DELETED = '/\\x01[^\\x01]*\\x01/';//g |
| 562 |
private $_escaped = array(); |
| 563 |
private $_patterns = array(); |
| 564 |
private $buffer; |
| 565 |
|
| 566 |
// private |
| 567 |
|
| 568 |
public function add( $expression, $replacement = '' ) |
| 569 |
{ |
| 570 |
// count the number of sub-expressions |
| 571 |
// - add one because each pattern is itself a sub-expression |
| 572 |
$length = 1 + preg_match_all( $this->GROUPS, $this->_internalEscape( (string)$expression ), $out ); |
| 573 |
// treat only strings $replacement |
| 574 |
if ( is_string( $replacement ) ) { |
| 575 |
// does the pattern deal with sub-expressions? |
| 576 |
if ( preg_match( $this->SUB_REPLACE, $replacement ) ) { |
| 577 |
// a simple lookup? (e.g. "$2") |
| 578 |
if ( preg_match( $this->INDEXED, $replacement ) ) { |
| 579 |
// store the index (used for fast retrieval of matched strings) |
| 580 |
$replacement = (int)(substr( $replacement, 1 )) - 1; |
| 581 |
} else { // a complicated lookup (e.g. "Hello $2 $1") |
| 582 |
// build a function to do the lookup |
| 583 |
$quote = preg_match( $this->QUOTE, $this->_internalEscape( $replacement ) ) |
| 584 |
? '"' : "'"; |
| 585 |
$replacement = array( |
| 586 |
'fn' => '_backReferences', |
| 587 |
'data' => array( |
| 588 |
'replacement' => $replacement, |
| 589 |
'length' => $length, |
| 590 |
'quote' => $quote, |
| 591 |
), |
| 592 |
); |
| 593 |
} |
| 594 |
} |
| 595 |
} |
| 596 |
// pass the modified arguments |
| 597 |
if ( !empty( $expression ) ) $this->_add( $expression, $replacement, $length ); |
| 598 |
else $this->_add( '/^$/', $replacement, $length ); |
| 599 |
} // escaped characters |
| 600 |
|
| 601 |
private function _internalEscape( $string ) |
| 602 |
{ |
| 603 |
return preg_replace( $this->ESCAPE, '', $string ); |
| 604 |
} // patterns stored by index |
| 605 |
|
| 606 |
// create and add a new pattern to the patterns collection |
| 607 |
|
| 608 |
private function _add() |
| 609 |
{ |
| 610 |
$arguments = func_get_args(); |
| 611 |
$this->_patterns[] = $arguments; |
| 612 |
} |
| 613 |
|
| 614 |
// this is the global replace function (it's quite complicated) |
| 615 |
|
| 616 |
public function exec( $string ) |
| 617 |
{ |
| 618 |
// execute the global replacement |
| 619 |
$this->_escaped = array(); |
| 620 |
// simulate the _patterns.toSTring of Dean |
| 621 |
$regexp = '/'; |
| 622 |
foreach ( $this->_patterns as $reg ) { |
| 623 |
$regexp .= '(' . substr( $reg[ self::EXPRESSION ], 1, -1 ) . ')|'; |
| 624 |
} |
| 625 |
$regexp = substr( $regexp, 0, -1 ) . '/'; |
| 626 |
$regexp .= ($this->ignoreCase) ? 'i' : ''; |
| 627 |
$string = $this->_escape( $string, $this->escapeChar ); |
| 628 |
$string = preg_replace_callback( |
| 629 |
$regexp, |
| 630 |
array( |
| 631 |
&$this, |
| 632 |
'_replacement', |
| 633 |
), |
| 634 |
$string |
| 635 |
); |
| 636 |
$string = $this->_unescape( $string, $this->escapeChar ); |
| 637 |
return preg_replace( $this->DELETED, '', $string ); |
| 638 |
} |
| 639 |
|
| 640 |
private function _escape( $string, $escapeChar ) |
| 641 |
{ |
| 642 |
if ( $escapeChar ) { |
| 643 |
$this->buffer = $escapeChar; |
| 644 |
return preg_replace_callback( |
| 645 |
'/\\' . $escapeChar . '(.)' . '/', |
| 646 |
array( &$this, '_escapeBis' ), |
| 647 |
$string |
| 648 |
); |
| 649 |
} else { |
| 650 |
return $string; |
| 651 |
} |
| 652 |
} |
| 653 |
|
| 654 |
private function _unescape( $string, $escapeChar ) |
| 655 |
{ |
| 656 |
if ( $escapeChar ) { |
| 657 |
$regexp = '/' . '\\' . $escapeChar . '/'; |
| 658 |
$this->buffer = array( 'escapeChar' => $escapeChar, 'i' => 0 ); |
| 659 |
return preg_replace_callback |
| 660 |
( |
| 661 |
$regexp, |
| 662 |
array( &$this, '_unescapeBis' ), |
| 663 |
$string |
| 664 |
); |
| 665 |
} else { |
| 666 |
return $string; |
| 667 |
} |
| 668 |
} |
| 669 |
|
| 670 |
public function reset() |
| 671 |
{ |
| 672 |
// clear the patterns collection so that this object may be re-used |
| 673 |
$this->_patterns = array(); |
| 674 |
} |
| 675 |
// php : we cannot pass additional data to preg_replace_callback, |
| 676 |
// and we cannot use &$this in create_function, so let's go to lower level |
| 677 |
|
| 678 |
private function _replacement( $arguments ) |
| 679 |
{ |
| 680 |
if ( empty( $arguments ) ) return ''; |
| 681 |
$i = 1; |
| 682 |
$j = 0; |
| 683 |
// loop through the patterns |
| 684 |
while ( isset( $this->_patterns[ $j ] ) ) { |
| 685 |
$pattern = $this->_patterns[ $j++ ]; |
| 686 |
// do we have a result? |
| 687 |
if ( isset( $arguments[ $i ] ) && ($arguments[ $i ] != '') ) { |
| 688 |
$replacement = $pattern[ self::REPLACEMENT ]; |
| 689 |
if ( is_array( $replacement ) && isset( $replacement[ 'fn' ] ) ) { |
| 690 |
if ( isset( $replacement[ 'data' ] ) ) $this->buffer = $replacement[ 'data' ]; |
| 691 |
return call_user_func( array( &$this, $replacement[ 'fn' ] ), $arguments, $i ); |
| 692 |
} elseif ( is_int( $replacement ) ) { |
| 693 |
return $arguments[ $replacement + $i ]; |
| 694 |
} |
| 695 |
$delete = ($this->escapeChar == '' || |
| 696 |
strpos( $arguments[ $i ], $this->escapeChar ) === false) |
| 697 |
? '' : "\x01" . $arguments[ $i ] . "\x01"; |
| 698 |
return $delete . $replacement; |
| 699 |
// skip over references to sub-expressions |
| 700 |
} else { |
| 701 |
$i += $pattern[ self::LENGTH ]; |
| 702 |
} |
| 703 |
} |
| 704 |
} |
| 705 |
|
| 706 |
// encode escaped characters |
| 707 |
|
| 708 |
private function _backReferences( $match, $offset ) |
| 709 |
{ |
| 710 |
$replacement = $this->buffer[ 'replacement' ]; |
| 711 |
$quote = $this->buffer[ 'quote' ]; |
| 712 |
$i = $this->buffer[ 'length' ]; |
| 713 |
while ( $i ) { |
| 714 |
$replacement = str_replace( '$' . $i--, $match[ $offset + $i ], $replacement ); |
| 715 |
} |
| 716 |
return $replacement; |
| 717 |
} |
| 718 |
|
| 719 |
private function _replace_name( $match, $offset ) |
| 720 |
{ |
| 721 |
$length = strlen( $match[ $offset + 2 ] ); |
| 722 |
$start = $length - max( $length - strlen( $match[ $offset + 3 ] ), 0 ); |
| 723 |
return substr( $match[ $offset + 1 ], $start, $length ) . $match[ $offset + 4 ]; |
| 724 |
} |
| 725 |
|
| 726 |
// decode escaped characters |
| 727 |
|
| 728 |
private function _replace_encoded( $match, $offset ) |
| 729 |
{ |
| 730 |
return $this->buffer[ $match[ $offset ] ]; |
| 731 |
} |
| 732 |
|
| 733 |
private function _escapeBis( $match ) |
| 734 |
{ |
| 735 |
$this->_escaped[] = $match[ 1 ]; |
| 736 |
return $this->buffer; |
| 737 |
} |
| 738 |
|
| 739 |
private function _unescapeBis() |
| 740 |
{ |
| 741 |
if ( isset( $this->_escaped[ $this->buffer[ 'i' ] ] ) |
| 742 |
&& $this->_escaped[ $this->buffer[ 'i' ] ] != '' ) { |
| 743 |
$temp = $this->_escaped[ $this->buffer[ 'i' ] ]; |
| 744 |
} else { |
| 745 |
$temp = ''; |
| 746 |
} |
| 747 |
$this->buffer[ 'i' ]++; |
| 748 |
return $this->buffer[ 'escapeChar' ] . $temp; |
| 749 |
} |
| 750 |
} |
| 751 |
|
| 752 |
?> |
