| [ Index ] |
|
Code source de DokuWiki 2006-11-06 |
1 <?php 2 /* vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: */ 3 4 /** 5 * Converts to and from JSON format. 6 * 7 * JSON (JavaScript Object Notation) is a lightweight data-interchange 8 * format. It is easy for humans to read and write. It is easy for machines 9 * to parse and generate. It is based on a subset of the JavaScript 10 * Programming Language, Standard ECMA-262 3rd Edition - December 1999. 11 * This feature can also be found in Python. JSON is a text format that is 12 * completely language independent but uses conventions that are familiar 13 * to programmers of the C-family of languages, including C, C++, C#, Java, 14 * JavaScript, Perl, TCL, and many others. These properties make JSON an 15 * ideal data-interchange language. 16 * 17 * This package provides a simple encoder and decoder for JSON notation. It 18 * is intended for use with client-side Javascript applications that make 19 * use of HTTPRequest to perform server communication functions - data can 20 * be encoded into JSON notation for use in a client-side javascript, or 21 * decoded from incoming Javascript requests. JSON format is native to 22 * Javascript, and can be directly eval()'ed with no further parsing 23 * overhead 24 * 25 * All strings should be in ASCII or UTF-8 format! 26 * 27 * PHP versions 4 and 5 28 * 29 * LICENSE: Redistribution and use in source and binary forms, with or 30 * without modification, are permitted provided that the following 31 * conditions are met: Redistributions of source code must retain the 32 * above copyright notice, this list of conditions and the following 33 * disclaimer. Redistributions in binary form must reproduce the above 34 * copyright notice, this list of conditions and the following disclaimer 35 * in the documentation and/or other materials provided with the 36 * distribution. 37 * 38 * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED 39 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF 40 * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN 41 * NO EVENT SHALL CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, 42 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, 43 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS 44 * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND 45 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR 46 * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE 47 * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH 48 * DAMAGE. 49 * 50 * @category 51 * @package 52 * @author Michal Migurski <mike-json@teczno.com> 53 * @author Matt Knapp <mdknapp[at]gmail[dot]com> 54 * @author Brett Stimmerman <brettstimmerman[at]gmail[dot]com> 55 * @copyright 2005 Michal Migurski 56 * @license http://www.freebsd.org/copyright/freebsd-license.html 57 * @link http://pear.php.net/pepr/pepr-proposal-show.php?id=198 58 */ 59 60 // for DokuWiki 61 if(!defined('DOKU_INC')) define('DOKU_INC',realpath(dirname(__FILE__).'/../').'/'); 62 require_once (DOKU_INC.'inc/utf8.php'); 63 64 /** 65 * Marker constant for JSON::decode(), used to flag stack state 66 */ 67 define('JSON_SLICE', 1); 68 69 /** 70 * Marker constant for JSON::decode(), used to flag stack state 71 */ 72 define('JSON_IN_STR', 2); 73 74 /** 75 * Marker constant for JSON::decode(), used to flag stack state 76 */ 77 define('JSON_IN_ARR', 4); 78 79 /** 80 * Marker constant for JSON::decode(), used to flag stack state 81 */ 82 define('JSON_IN_OBJ', 8); 83 84 /** 85 * Marker constant for JSON::decode(), used to flag stack state 86 */ 87 define('JSON_IN_CMT', 16); 88 89 /** 90 * Behavior switch for JSON::decode() 91 */ 92 define('JSON_LOOSE_TYPE', 10); 93 94 /** 95 * Behavior switch for JSON::decode() 96 */ 97 define('JSON_STRICT_TYPE', 11); 98 99 /** 100 * Converts to and from JSON format. 101 * 102 * @category 103 * @package 104 * @author Michal Migurski <mike-json@teczno.com> 105 * @author Matt Knapp <mdknapp[at]gmail[dot]com> 106 * @author Brett Stimmerman <brettstimmerman[at]gmail[dot]com> 107 * @copyright 2005 Michal Migurski 108 * @license http://www.php.net/license/3_0.txt PHP License 3.0 109 * @version 110 * @link 111 * @see 112 * @since 113 * @deprecated 114 */ 115 class JSON 116 { 117 /** 118 * constructs a new JSON instance 119 * 120 * @param int $use object behavior: when encoding or decoding, 121 * be loose or strict about object/array usage 122 * 123 * possible values: 124 * JSON_STRICT_TYPE - strict typing, default 125 * "{...}" syntax creates objects in decode. 126 * JSON_LOOSE_TYPE - loose typing 127 * "{...}" syntax creates associative arrays in decode. 128 */ 129 function JSON($use=JSON_STRICT_TYPE) 130 { 131 $this->use = $use; 132 } 133 134 /** 135 * encodes an arbitrary variable into JSON format 136 * 137 * @param mixed $var any number, boolean, string, array, or object to be encoded. 138 * see argument 1 to JSON() above for array-parsing behavior. 139 * if var is a strng, note that encode() always expects it 140 * to be in ASCII or UTF-8 format! 141 * 142 * @return string JSON string representation of input var 143 * @access public 144 */ 145 function encode($var) 146 { 147 switch (gettype($var)) { 148 case 'boolean': 149 return $var ? 'true' : 'false'; 150 151 case 'NULL': 152 return 'null'; 153 154 case 'integer': 155 return sprintf('%d', $var); 156 157 case 'double': 158 case 'float': 159 return sprintf('%f', $var); 160 161 case 'string': 162 // STRINGS ARE EXPECTED TO BE IN ASCII OR UTF-8 FORMAT 163 $ascii = ''; 164 $strlen_var = strlen($var); 165 166 /* 167 * Iterate over every character in the string, 168 * escaping with a slash or encoding to UTF-8 where necessary 169 */ 170 for ($c = 0; $c < $strlen_var; ++$c) { 171 172 $ord_var_c = ord($var{$c}); 173 174 switch ($ord_var_c) { 175 case 0x08: $ascii .= '\b'; break; 176 case 0x09: $ascii .= '\t'; break; 177 case 0x0A: $ascii .= '\n'; break; 178 case 0x0C: $ascii .= '\f'; break; 179 case 0x0D: $ascii .= '\r'; break; 180 181 case 0x22: 182 case 0x2F: 183 case 0x5C: 184 // double quote, slash, slosh 185 $ascii .= '\\'.$var{$c}; 186 break; 187 188 case (($ord_var_c >= 0x20) && ($ord_var_c <= 0x7F)): 189 // characters U-00000000 - U-0000007F (same as ASCII) 190 $ascii .= $var{$c}; 191 break; 192 193 case (($ord_var_c & 0xE0) == 0xC0): 194 // characters U-00000080 - U-000007FF, mask 110XXXXX 195 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 196 $char = pack('C*', $ord_var_c, ord($var{$c+1})); 197 $c+=1; 198 //$utf16 = mb_convert_encoding($char, 'UTF-16', 'UTF-8'); 199 $utf16 = utf8_to_utf16be($char); 200 $ascii .= sprintf('\u%04s', bin2hex($utf16)); 201 break; 202 203 case (($ord_var_c & 0xF0) == 0xE0): 204 // characters U-00000800 - U-0000FFFF, mask 1110XXXX 205 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 206 $char = pack('C*', $ord_var_c, 207 ord($var{$c+1}), 208 ord($var{$c+2})); 209 $c+=2; 210 //$utf16 = mb_convert_encoding($char, 'UTF-16', 'UTF-8'); 211 $utf16 = utf8_to_utf16be($char); 212 $ascii .= sprintf('\u%04s', bin2hex($utf16)); 213 break; 214 215 case (($ord_var_c & 0xF8) == 0xF0): 216 // characters U-00010000 - U-001FFFFF, mask 11110XXX 217 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 218 $char = pack('C*', $ord_var_c, 219 ord($var{$c+1}), 220 ord($var{$c+2}), 221 ord($var{$c+3})); 222 $c+=3; 223 //$utf16 = mb_convert_encoding($char, 'UTF-16', 'UTF-8'); 224 $utf16 = utf8_to_utf16be($char); 225 $ascii .= sprintf('\u%04s', bin2hex($utf16)); 226 break; 227 228 case (($ord_var_c & 0xFC) == 0xF8): 229 // characters U-00200000 - U-03FFFFFF, mask 111110XX 230 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 231 $char = pack('C*', $ord_var_c, 232 ord($var{$c+1}), 233 ord($var{$c+2}), 234 ord($var{$c+3}), 235 ord($var{$c+4})); 236 $c+=4; 237 //$utf16 = mb_convert_encoding($char, 'UTF-16', 'UTF-8'); 238 $utf16 = utf8_to_utf16be($char); 239 $ascii .= sprintf('\u%04s', bin2hex($utf16)); 240 break; 241 242 case (($ord_var_c & 0xFE) == 0xFC): 243 // characters U-04000000 - U-7FFFFFFF, mask 1111110X 244 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 245 $char = pack('C*', $ord_var_c, 246 ord($var{$c+1}), 247 ord($var{$c+2}), 248 ord($var{$c+3}), 249 ord($var{$c+4}), 250 ord($var{$c+5})); 251 $c+=5; 252 //$utf16 = mb_convert_encoding($char, 'UTF-16', 'UTF-8'); 253 $utf16 = utf8_to_utf16be($char); 254 $ascii .= sprintf('\u%04s', bin2hex($utf16)); 255 break; 256 } 257 } 258 259 return '"'.$ascii.'"'; 260 261 case 'array': 262 /* 263 * As per JSON spec if any array key is not an integer 264 * we must treat the the whole array as an object. We 265 * also try to catch a sparsely populated associative 266 * array with numeric keys here because some JS engines 267 * will create an array with empty indexes up to 268 * max_index which can cause memory issues and because 269 * the keys, which may be relevant, will be remapped 270 * otherwise. 271 * 272 * As per the ECMA and JSON specification an object may 273 * have any string as a property. Unfortunately due to 274 * a hole in the ECMA specification if the key is a 275 * ECMA reserved word or starts with a digit the 276 * parameter is only accessible using ECMAScript's 277 * bracket notation. 278 */ 279 280 // treat as a JSON object 281 if (is_array($var) && count($var) && (array_keys($var) !== range(0, sizeof($var) - 1))) { 282 return sprintf('{%s}', join(',', array_map(array($this, 'name_value'), 283 array_keys($var), 284 array_values($var)))); 285 } 286 287 // treat it like a regular array 288 return sprintf('[%s]', join(',', array_map(array($this, 'encode'), $var))); 289 290 case 'object': 291 $vars = get_object_vars($var); 292 return sprintf('{%s}', join(',', array_map(array($this, 'name_value'), 293 array_keys($vars), 294 array_values($vars)))); 295 296 default: 297 return ''; 298 } 299 } 300 301 /** 302 * encodes an arbitrary variable into JSON format, alias for encode() 303 */ 304 function enc($var) 305 { 306 return $this->encode($var); 307 } 308 309 /** function name_value 310 * array-walking function for use in generating JSON-formatted name-value pairs 311 * 312 * @param string $name name of key to use 313 * @param mixed $value reference to an array element to be encoded 314 * 315 * @return string JSON-formatted name-value pair, like '"name":value' 316 * @access private 317 */ 318 function name_value($name, $value) 319 { 320 return (sprintf("%s:%s", $this->encode(strval($name)), $this->encode($value))); 321 } 322 323 /** 324 * reduce a string by removing leading and trailing comments and whitespace 325 * 326 * @param $str string string value to strip of comments and whitespace 327 * 328 * @return string string value stripped of comments and whitespace 329 * @access private 330 */ 331 function reduce_string($str) 332 { 333 $str = preg_replace(array( 334 335 // eliminate single line comments in '// ...' form 336 '#^\s*//(.+)$#m', 337 338 // eliminate multi-line comments in '/* ... */' form, at start of string 339 '#^\s*/\*(.+)\*/#Us', 340 341 // eliminate multi-line comments in '/* ... */' form, at end of string 342 '#/\*(.+)\*/\s*$#Us' 343 344 ), '', $str); 345 346 // eliminate extraneous space 347 return trim($str); 348 } 349 350 /** 351 * decodes a JSON string into appropriate variable 352 * 353 * @param string $str JSON-formatted string 354 * 355 * @return mixed number, boolean, string, array, or object 356 * corresponding to given JSON input string. 357 * See argument 1 to JSON() above for object-output behavior. 358 * Note that decode() always returns strings 359 * in ASCII or UTF-8 format! 360 * @access public 361 */ 362 function decode($str) 363 { 364 $str = $this->reduce_string($str); 365 366 switch (strtolower($str)) { 367 case 'true': 368 return true; 369 370 case 'false': 371 return false; 372 373 case 'null': 374 return null; 375 376 default: 377 if (is_numeric($str)) { 378 // Lookie-loo, it's a number 379 380 // This would work on its own, but I'm trying to be 381 // good about returning integers where appropriate: 382 // return (float)$str; 383 384 // Return float or int, as appropriate 385 return ((float)$str == (integer)$str) 386 ? (integer)$str 387 : (float)$str; 388 389 } elseif (preg_match('/^("|\').+("|\')$/s', $str, $m) && $m[1] == $m[2]) { 390 // STRINGS RETURNED IN UTF-8 FORMAT 391 $delim = substr($str, 0, 1); 392 $chrs = substr($str, 1, -1); 393 $utf8 = ''; 394 $strlen_chrs = strlen($chrs); 395 396 for ($c = 0; $c < $strlen_chrs; ++$c) { 397 398 $substr_chrs_c_2 = substr($chrs, $c, 2); 399 $ord_chrs_c = ord($chrs{$c}); 400 401 switch ($substr_chrs_c_2) { 402 case '\b': $utf8 .= chr(0x08); $c+=1; break; 403 case '\t': $utf8 .= chr(0x09); $c+=1; break; 404 case '\n': $utf8 .= chr(0x0A); $c+=1; break; 405 case '\f': $utf8 .= chr(0x0C); $c+=1; break; 406 case '\r': $utf8 .= chr(0x0D); $c+=1; break; 407 408 case '\\"': 409 case '\\\'': 410 case '\\\\': 411 case '\\/': 412 if (($delim == '"' && $substr_chrs_c_2 != '\\\'') || 413 ($delim == "'" && $substr_chrs_c_2 != '\\"')) { 414 $utf8 .= $chrs{++$c}; 415 } 416 break; 417 418 default: 419 if (preg_match('/\\\u[0-9A-F]{4}/i', substr($chrs, $c, 6))) { 420 // single, escaped unicode character 421 $utf16 = chr(hexdec(substr($chrs, ($c+2), 2))) 422 . chr(hexdec(substr($chrs, ($c+4), 2))); 423 //$utf8 .= mb_convert_encoding($utf16, 'UTF-8', 'UTF-16'); 424 $utf8 .= utf16be_to_utf8($utf16); 425 $c+=5; 426 427 } elseif(($ord_chrs_c >= 0x20) && ($ord_chrs_c <= 0x7F)) { 428 $utf8 .= $chrs{$c}; 429 430 } elseif(($ord_chrs_c & 0xE0) == 0xC0) { 431 // characters U-00000080 - U-000007FF, mask 110XXXXX 432 //see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 433 $utf8 .= substr($chrs, $c, 2); $c += 1; 434 435 } elseif(($ord_chrs_c & 0xF0) == 0xE0) { 436 // characters U-00000800 - U-0000FFFF, mask 1110XXXX 437 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 438 $utf8 .= substr($chrs, $c, 3); $c += 2; 439 440 } elseif(($ord_chrs_c & 0xF8) == 0xF0) { 441 // characters U-00010000 - U-001FFFFF, mask 11110XXX 442 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 443 $utf8 .= substr($chrs, $c, 4); $c += 3; 444 445 } elseif(($ord_chrs_c & 0xFC) == 0xF8) { 446 // characters U-00200000 - U-03FFFFFF, mask 111110XX 447 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 448 $utf8 .= substr($chrs, $c, 5); $c += 4; 449 450 } elseif(($ord_chrs_c & 0xFE) == 0xFC) { 451 // characters U-04000000 - U-7FFFFFFF, mask 1111110X 452 // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 453 $utf8 .= substr($chrs, $c, 6); $c += 5; 454 455 } 456 break; 457 458 } 459 460 } 461 462 return $utf8; 463 464 } elseif (preg_match('/^\[.*\]$/s', $str) || preg_match('/^\{.*\}$/s', $str)) { 465 // array, or object notation 466 467 if ($str{0} == '[') { 468 $stk = array(JSON_IN_ARR); 469 $arr = array(); 470 } else { 471 if ($this->use == JSON_LOOSE_TYPE) { 472 $stk = array(JSON_IN_OBJ); 473 $obj = array(); 474 } else { 475 $stk = array(JSON_IN_OBJ); 476 $obj = new stdClass(); 477 } 478 } 479 480 array_push($stk, array('what' => JSON_SLICE, 481 'where' => 0, 482 'delim' => false)); 483 484 $chrs = substr($str, 1, -1); 485 $chrs = $this->reduce_string($chrs); 486 487 if ($chrs == '') { 488 if (reset($stk) == JSON_IN_ARR) { 489 return $arr; 490 491 } else { 492 return $obj; 493 494 } 495 } 496 497 //print("\nparsing {$chrs}\n"); 498 499 $strlen_chrs = strlen($chrs); 500 501 for ($c = 0; $c <= $strlen_chrs; ++$c) { 502 503 $top = end($stk); 504 $substr_chrs_c_2 = substr($chrs, $c, 2); 505 506 if (($c == $strlen_chrs) || (($chrs{$c} == ',') && ($top['what'] == JSON_SLICE))) { 507 // found a comma that is not inside a string, array, etc., 508 // OR we've reached the end of the character list 509 $slice = substr($chrs, $top['where'], ($c - $top['where'])); 510 array_push($stk, array('what' => JSON_SLICE, 'where' => ($c + 1), 'delim' => false)); 511 //print("Found split at {$c}: ".substr($chrs, $top['where'], (1 + $c - $top['where']))."\n"); 512 513 if (reset($stk) == JSON_IN_ARR) { 514 // we are in an array, so just push an element onto the stack 515 array_push($arr, $this->decode($slice)); 516 517 } elseif (reset($stk) == JSON_IN_OBJ) { 518 // we are in an object, so figure 519 // out the property name and set an 520 // element in an associative array, 521 // for now 522 if (preg_match('/^\s*(["\'].*[^\\\]["\'])\s*:\s*(\S.*),?$/Uis', $slice, $parts)) { 523 // "name":value pair 524 $key = $this->decode($parts[1]); 525 $val = $this->decode($parts[2]); 526 527 if ($this->use == JSON_LOOSE_TYPE) { 528 $obj[$key] = $val; 529 } else { 530 $obj->$key = $val; 531 } 532 } elseif (preg_match('/^\s*(\w+)\s*:\s*(\S.*),?$/Uis', $slice, $parts)) { 533 // name:value pair, where name is unquoted 534 $key = $parts[1]; 535 $val = $this->decode($parts[2]); 536 537 if ($this->use == JSON_LOOSE_TYPE) { 538 $obj[$key] = $val; 539 } else { 540 $obj->$key = $val; 541 } 542 } 543 544 } 545 546 } elseif ((($chrs{$c} == '"') || ($chrs{$c} == "'")) && 547 in_array($top['what'], array(JSON_SLICE, JSON_IN_ARR, JSON_IN_OBJ))) { 548 // found a quote, and we are not inside a string 549 array_push($stk, array('what' => JSON_IN_STR, 'where' => $c, 'delim' => $chrs{$c})); 550 //print("Found start of string at {$c}\n"); 551 552 } elseif (($chrs{$c} == $top['delim']) && 553 ($top['what'] == JSON_IN_STR) && 554 (($chrs{$c - 1} != "\\") || 555 ($chrs{$c - 1} == "\\" && $chrs{$c - 2} == "\\"))) { 556 // found a quote, we're in a string, and it's not escaped 557 array_pop($stk); 558 //print("Found end of string at {$c}: ".substr($chrs, $top['where'], (1 + 1 + $c - $top['where']))."\n"); 559 560 } elseif (($chrs{$c} == '[') && 561 in_array($top['what'], array(JSON_SLICE, JSON_IN_ARR, JSON_IN_OBJ))) { 562 // found a left-bracket, and we are in an array, object, or slice 563 array_push($stk, array('what' => JSON_IN_ARR, 'where' => $c, 'delim' => false)); 564 //print("Found start of array at {$c}\n"); 565 566 } elseif (($chrs{$c} == ']') && ($top['what'] == JSON_IN_ARR)) { 567 // found a right-bracket, and we're in an array 568 array_pop($stk); 569 //print("Found end of array at {$c}: ".substr($chrs, $top['where'], (1 + $c - $top['where']))."\n"); 570 571 } elseif (($chrs{$c} == '{') && 572 in_array($top['what'], array(JSON_SLICE, JSON_IN_ARR, JSON_IN_OBJ))) { 573 // found a left-brace, and we are in an array, object, or slice 574 array_push($stk, array('what' => JSON_IN_OBJ, 'where' => $c, 'delim' => false)); 575 //print("Found start of object at {$c}\n"); 576 577 } elseif (($chrs{$c} == '}') && ($top['what'] == JSON_IN_OBJ)) { 578 // found a right-brace, and we're in an object 579 array_pop($stk); 580 //print("Found end of object at {$c}: ".substr($chrs, $top['where'], (1 + $c - $top['where']))."\n"); 581 582 } elseif (($substr_chrs_c_2 == '/*') && 583 in_array($top['what'], array(JSON_SLICE, JSON_IN_ARR, JSON_IN_OBJ))) { 584 // found a comment start, and we are in an array, object, or slice 585 array_push($stk, array('what' => JSON_IN_CMT, 'where' => $c, 'delim' => false)); 586 $c++; 587 //print("Found start of comment at {$c}\n"); 588 589 } elseif (($substr_chrs_c_2 == '*/') && ($top['what'] == JSON_IN_CMT)) { 590 // found a comment end, and we're in one now 591 array_pop($stk); 592 $c++; 593 594 for ($i = $top['where']; $i <= $c; ++$i) 595 $chrs = substr_replace($chrs, ' ', $i, 1); 596 597 //print("Found end of comment at {$c}: ".substr($chrs, $top['where'], (1 + $c - $top['where']))."\n"); 598 599 } 600 601 } 602 603 if (reset($stk) == JSON_IN_ARR) { 604 return $arr; 605 606 } elseif (reset($stk) == JSON_IN_OBJ) { 607 return $obj; 608 609 } 610 611 } 612 } 613 } 614 615 /** 616 * decodes a JSON string into appropriate variable; alias for decode() 617 */ 618 function dec($var) 619 { 620 return $this->decode($var); 621 } 622 623 } 624 625 ?>
titre
Description
Corps
titre
Description
Corps
titre
Description
Corps
titre
Corps
| Généré le : Tue Apr 3 20:47:31 2007 | par Balluche grâce à PHPXref 0.7 |