WordPress 3.0 beta 1 documentation kindly provided to you by Hay Kranen
WordPress 3.0 beta 1 documentation kindly provided to you by Hay Kranen
| [ Index ] |
PHP Cross Reference of WordPress 3.0 beta 1 |
|
| [ Index ] [ Variables ] [ Functions ] [ Classes ] [ Constants ] [ Statistics ] | ||
[Summary view] [Print] [Text view]
1 <?php 2 /** 3 * HTML/XHTML filter that only allows some elements and attributes 4 * 5 * Added wp_ prefix to avoid conflicts with existing kses users 6 * 7 * @version 0.2.2 8 * @copyright (C) 2002, 2003, 2005 9 * @author Ulf Harnhammar <metaur@users.sourceforge.net> 10 * 11 * @package External 12 * @subpackage KSES 13 * 14 * @internal 15 * *** CONTACT INFORMATION *** 16 * E-mail: metaur at users dot sourceforge dot net 17 * Web page: http://sourceforge.net/projects/kses 18 * Paper mail: Ulf Harnhammar 19 * Ymergatan 17 C 20 * 753 25 Uppsala 21 * SWEDEN 22 * 23 * [kses strips evil scripts!] 24 */ 25 26 /** 27 * You can override this in a plugin. 28 * 29 * @since 1.2.0 30 */ 31 if ( ! defined( 'CUSTOM_TAGS' ) ) 32 define( 'CUSTOM_TAGS', false ); 33 34 if ( ! CUSTOM_TAGS ) { 35 /** 36 * Kses global for default allowable HTML tags. 37 * 38 * Can be override by using CUSTOM_TAGS constant. 39 * 40 * @global array $allowedposttags 41 * @since 2.0.0 42 */ 43 $allowedposttags = array( 44 'address' => array(), 45 'a' => array( 46 'class' => array (), 47 'href' => array (), 48 'id' => array (), 49 'title' => array (), 50 'rel' => array (), 51 'rev' => array (), 52 'name' => array (), 53 'target' => array()), 54 'abbr' => array( 55 'class' => array (), 56 'title' => array ()), 57 'acronym' => array( 58 'title' => array ()), 59 'b' => array(), 60 'big' => array(), 61 'blockquote' => array( 62 'id' => array (), 63 'cite' => array (), 64 'class' => array(), 65 'lang' => array(), 66 'xml:lang' => array()), 67 'br' => array ( 68 'class' => array ()), 69 'button' => array( 70 'disabled' => array (), 71 'name' => array (), 72 'type' => array (), 73 'value' => array ()), 74 'caption' => array( 75 'align' => array (), 76 'class' => array ()), 77 'cite' => array ( 78 'class' => array(), 79 'dir' => array(), 80 'lang' => array(), 81 'title' => array ()), 82 'code' => array ( 83 'style' => array()), 84 'col' => array( 85 'align' => array (), 86 'char' => array (), 87 'charoff' => array (), 88 'span' => array (), 89 'dir' => array(), 90 'style' => array (), 91 'valign' => array (), 92 'width' => array ()), 93 'del' => array( 94 'datetime' => array ()), 95 'dd' => array(), 96 'div' => array( 97 'align' => array (), 98 'class' => array (), 99 'dir' => array (), 100 'lang' => array(), 101 'style' => array (), 102 'xml:lang' => array()), 103 'dl' => array(), 104 'dt' => array(), 105 'em' => array(), 106 'fieldset' => array(), 107 'font' => array( 108 'color' => array (), 109 'face' => array (), 110 'size' => array ()), 111 'form' => array( 112 'action' => array (), 113 'accept' => array (), 114 'accept-charset' => array (), 115 'enctype' => array (), 116 'method' => array (), 117 'name' => array (), 118 'target' => array ()), 119 'h1' => array( 120 'align' => array (), 121 'class' => array (), 122 'id' => array (), 123 'style' => array ()), 124 'h2' => array ( 125 'align' => array (), 126 'class' => array (), 127 'id' => array (), 128 'style' => array ()), 129 'h3' => array ( 130 'align' => array (), 131 'class' => array (), 132 'id' => array (), 133 'style' => array ()), 134 'h4' => array ( 135 'align' => array (), 136 'class' => array (), 137 'id' => array (), 138 'style' => array ()), 139 'h5' => array ( 140 'align' => array (), 141 'class' => array (), 142 'id' => array (), 143 'style' => array ()), 144 'h6' => array ( 145 'align' => array (), 146 'class' => array (), 147 'id' => array (), 148 'style' => array ()), 149 'hr' => array ( 150 'align' => array (), 151 'class' => array (), 152 'noshade' => array (), 153 'size' => array (), 154 'width' => array ()), 155 'i' => array(), 156 'img' => array( 157 'alt' => array (), 158 'align' => array (), 159 'border' => array (), 160 'class' => array (), 161 'height' => array (), 162 'hspace' => array (), 163 'longdesc' => array (), 164 'vspace' => array (), 165 'src' => array (), 166 'style' => array (), 167 'width' => array ()), 168 'ins' => array( 169 'datetime' => array (), 170 'cite' => array ()), 171 'kbd' => array(), 172 'label' => array( 173 'for' => array ()), 174 'legend' => array( 175 'align' => array ()), 176 'li' => array ( 177 'align' => array (), 178 'class' => array ()), 179 'p' => array( 180 'class' => array (), 181 'align' => array (), 182 'dir' => array(), 183 'lang' => array(), 184 'style' => array (), 185 'xml:lang' => array()), 186 'pre' => array( 187 'style' => array(), 188 'width' => array ()), 189 'q' => array( 190 'cite' => array ()), 191 's' => array(), 192 'span' => array ( 193 'class' => array (), 194 'dir' => array (), 195 'align' => array (), 196 'lang' => array (), 197 'style' => array (), 198 'title' => array (), 199 'xml:lang' => array()), 200 'strike' => array(), 201 'strong' => array(), 202 'sub' => array(), 203 'sup' => array(), 204 'table' => array( 205 'align' => array (), 206 'bgcolor' => array (), 207 'border' => array (), 208 'cellpadding' => array (), 209 'cellspacing' => array (), 210 'class' => array (), 211 'dir' => array(), 212 'id' => array(), 213 'rules' => array (), 214 'style' => array (), 215 'summary' => array (), 216 'width' => array ()), 217 'tbody' => array( 218 'align' => array (), 219 'char' => array (), 220 'charoff' => array (), 221 'valign' => array ()), 222 'td' => array( 223 'abbr' => array (), 224 'align' => array (), 225 'axis' => array (), 226 'bgcolor' => array (), 227 'char' => array (), 228 'charoff' => array (), 229 'class' => array (), 230 'colspan' => array (), 231 'dir' => array(), 232 'headers' => array (), 233 'height' => array (), 234 'nowrap' => array (), 235 'rowspan' => array (), 236 'scope' => array (), 237 'style' => array (), 238 'valign' => array (), 239 'width' => array ()), 240 'textarea' => array( 241 'cols' => array (), 242 'rows' => array (), 243 'disabled' => array (), 244 'name' => array (), 245 'readonly' => array ()), 246 'tfoot' => array( 247 'align' => array (), 248 'char' => array (), 249 'class' => array (), 250 'charoff' => array (), 251 'valign' => array ()), 252 'th' => array( 253 'abbr' => array (), 254 'align' => array (), 255 'axis' => array (), 256 'bgcolor' => array (), 257 'char' => array (), 258 'charoff' => array (), 259 'class' => array (), 260 'colspan' => array (), 261 'headers' => array (), 262 'height' => array (), 263 'nowrap' => array (), 264 'rowspan' => array (), 265 'scope' => array (), 266 'valign' => array (), 267 'width' => array ()), 268 'thead' => array( 269 'align' => array (), 270 'char' => array (), 271 'charoff' => array (), 272 'class' => array (), 273 'valign' => array ()), 274 'title' => array(), 275 'tr' => array( 276 'align' => array (), 277 'bgcolor' => array (), 278 'char' => array (), 279 'charoff' => array (), 280 'class' => array (), 281 'style' => array (), 282 'valign' => array ()), 283 'tt' => array(), 284 'u' => array(), 285 'ul' => array ( 286 'class' => array (), 287 'style' => array (), 288 'type' => array ()), 289 'ol' => array ( 290 'class' => array (), 291 'start' => array (), 292 'style' => array (), 293 'type' => array ()), 294 'var' => array ()); 295 296 /** 297 * Kses allowed HTML elements. 298 * 299 * @global array $allowedtags 300 * @since 1.0.0 301 */ 302 $allowedtags = array( 303 'a' => array( 304 'href' => array (), 305 'title' => array ()), 306 'abbr' => array( 307 'title' => array ()), 308 'acronym' => array( 309 'title' => array ()), 310 'b' => array(), 311 'blockquote' => array( 312 'cite' => array ()), 313 // 'br' => array(), 314 'cite' => array (), 315 'code' => array(), 316 'del' => array( 317 'datetime' => array ()), 318 // 'dd' => array(), 319 // 'dl' => array(), 320 // 'dt' => array(), 321 'em' => array (), 'i' => array (), 322 // 'ins' => array('datetime' => array(), 'cite' => array()), 323 // 'li' => array(), 324 // 'ol' => array(), 325 // 'p' => array(), 326 'q' => array( 327 'cite' => array ()), 328 'strike' => array(), 329 'strong' => array(), 330 // 'sub' => array(), 331 // 'sup' => array(), 332 // 'u' => array(), 333 // 'ul' => array(), 334 ); 335 336 $allowedentitynames = array( 337 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen', 338 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo', 339 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn', 340 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm', 341 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde', 342 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute', 343 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml', 344 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde', 345 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc', 346 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute', 347 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil', 348 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute', 349 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute', 350 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave', 351 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml', 352 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig', 353 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde', 354 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm', 355 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo', 356 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil', 357 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta', 358 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta', 359 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi', 360 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon', 361 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta', 362 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta', 363 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi', 364 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau', 365 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym', 366 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime', 367 'oline', 'frasl', 'weierp', 'image', 'real', 'trade', 368 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr', 369 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr', 370 'forall', 'part', 'exist', 'empty', 'nabla', 'isin', 371 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast', 372 'radic', 'prop', 'infin', 'ang', 'and', 'or', 373 'cap', 'cup', 'int', 'sim', 'cong', 'asymp', 374 'ne', 'equiv', 'le', 'ge', 'sub', 'sup', 375 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp', 376 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang', 377 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams', 378 ); 379 } 380 381 /** 382 * Filters content and keeps only allowable HTML elements. 383 * 384 * This function makes sure that only the allowed HTML element names, attribute 385 * names and attribute values plus only sane HTML entities will occur in 386 * $string. You have to remove any slashes from PHP's magic quotes before you 387 * call this function. 388 * 389 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news', 390 * 'irc', 'gopher', 'nntp', 'feed', and finally 'telnet. This covers all common 391 * link protocols, except for 'javascript' which should not be allowed for 392 * untrusted users. 393 * 394 * @since 1.0.0 395 * 396 * @param string $string Content to filter through kses 397 * @param array $allowed_html List of allowed HTML elements 398 * @param array $allowed_protocols Optional. Allowed protocol in links. 399 * @return string Filtered content with only allowed HTML elements 400 */ 401 function wp_kses($string, $allowed_html, $allowed_protocols = array ('http', 'https', 'ftp', 'ftps', 'mailto', 'news', 'irc', 'gopher', 'nntp', 'feed', 'telnet')) { 402 $string = wp_kses_no_null($string); 403 $string = wp_kses_js_entities($string); 404 $string = wp_kses_normalize_entities($string); 405 $allowed_html_fixed = wp_kses_array_lc($allowed_html); 406 $string = wp_kses_hook($string, $allowed_html_fixed, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook 407 return wp_kses_split($string, $allowed_html_fixed, $allowed_protocols); 408 } 409 410 /** 411 * You add any kses hooks here. 412 * 413 * There is currently only one kses WordPress hook and it is called here. All 414 * parameters are passed to the hooks and expected to recieve a string. 415 * 416 * @since 1.0.0 417 * 418 * @param string $string Content to filter through kses 419 * @param array $allowed_html List of allowed HTML elements 420 * @param array $allowed_protocols Allowed protocol in links 421 * @return string Filtered content through 'pre_kses' hook 422 */ 423 function wp_kses_hook($string, $allowed_html, $allowed_protocols) { 424 $string = apply_filters('pre_kses', $string, $allowed_html, $allowed_protocols); 425 return $string; 426 } 427 428 /** 429 * This function returns kses' version number. 430 * 431 * @since 1.0.0 432 * 433 * @return string KSES Version Number 434 */ 435 function wp_kses_version() { 436 return '0.2.2'; 437 } 438 439 /** 440 * Searches for HTML tags, no matter how malformed. 441 * 442 * It also matches stray ">" characters. 443 * 444 * @since 1.0.0 445 * 446 * @param string $string Content to filter 447 * @param array $allowed_html Allowed HTML elements 448 * @param array $allowed_protocols Allowed protocols to keep 449 * @return string Content with fixed HTML tags 450 */ 451 function wp_kses_split($string, $allowed_html, $allowed_protocols) { 452 global $pass_allowed_html, $pass_allowed_protocols; 453 $pass_allowed_html = $allowed_html; 454 $pass_allowed_protocols = $allowed_protocols; 455 return preg_replace_callback('%((<!--.*?(-->|$))|(<[^>]*(>|$)|>))%', 456 create_function('$match', 'global $pass_allowed_html, $pass_allowed_protocols; return wp_kses_split2($match[1], $pass_allowed_html, $pass_allowed_protocols);'), $string); 457 } 458 459 /** 460 * Callback for wp_kses_split for fixing malformed HTML tags. 461 * 462 * This function does a lot of work. It rejects some very malformed things like 463 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no 464 * strip_tags()!). Otherwise it splits the tag into an element and an attribute 465 * list. 466 * 467 * After the tag is split into an element and an attribute list, it is run 468 * through another filter which will remove illegal attributes and once that is 469 * completed, will be returned. 470 * 471 * @access private 472 * @since 1.0.0 473 * @uses wp_kses_attr() 474 * 475 * @param string $string Content to filter 476 * @param array $allowed_html Allowed HTML elements 477 * @param array $allowed_protocols Allowed protocols to keep 478 * @return string Fixed HTML element 479 */ 480 function wp_kses_split2($string, $allowed_html, $allowed_protocols) { 481 $string = wp_kses_stripslashes($string); 482 483 if (substr($string, 0, 1) != '<') 484 return '>'; 485 # It matched a ">" character 486 487 if (preg_match('%^<!--(.*?)(-->)?$%', $string, $matches)) { 488 $string = str_replace(array('<!--', '-->'), '', $matches[1]); 489 while ( $string != $newstring = wp_kses($string, $allowed_html, $allowed_protocols) ) 490 $string = $newstring; 491 if ( $string == '' ) 492 return ''; 493 // prevent multiple dashes in comments 494 $string = preg_replace('/--+/', '-', $string); 495 // prevent three dashes closing a comment 496 $string = preg_replace('/-$/', '', $string); 497 return "<!--{$string}-->"; 498 } 499 # Allow HTML comments 500 501 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches)) 502 return ''; 503 # It's seriously malformed 504 505 $slash = trim($matches[1]); 506 $elem = $matches[2]; 507 $attrlist = $matches[3]; 508 509 if (!@isset($allowed_html[strtolower($elem)])) 510 return ''; 511 # They are using a not allowed HTML element 512 513 if ($slash != '') 514 return "<$slash$elem>"; 515 # No attributes are allowed for closing elements 516 517 return wp_kses_attr("$slash$elem", $attrlist, $allowed_html, $allowed_protocols); 518 } 519 520 /** 521 * Removes all attributes, if none are allowed for this element. 522 * 523 * If some are allowed it calls wp_kses_hair() to split them further, and then 524 * it builds up new HTML code from the data that kses_hair() returns. It also 525 * removes "<" and ">" characters, if there are any left. One more thing it does 526 * is to check if the tag has a closing XHTML slash, and if it does, it puts one 527 * in the returned code as well. 528 * 529 * @since 1.0.0 530 * 531 * @param string $element HTML element/tag 532 * @param string $attr HTML attributes from HTML element to closing HTML element tag 533 * @param array $allowed_html Allowed HTML elements 534 * @param array $allowed_protocols Allowed protocols to keep 535 * @return string Sanitized HTML element 536 */ 537 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) { 538 # Is there a closing XHTML slash at the end of the attributes? 539 540 $xhtml_slash = ''; 541 if (preg_match('%\s*/\s*$%', $attr)) 542 $xhtml_slash = ' /'; 543 544 # Are any attributes allowed at all for this element? 545 546 if (@ count($allowed_html[strtolower($element)]) == 0) 547 return "<$element$xhtml_slash>"; 548 549 # Split it 550 551 $attrarr = wp_kses_hair($attr, $allowed_protocols); 552 553 # Go through $attrarr, and save the allowed attributes for this element 554 # in $attr2 555 556 $attr2 = ''; 557 558 foreach ($attrarr as $arreach) { 559 if (!@ isset ($allowed_html[strtolower($element)][strtolower($arreach['name'])])) 560 continue; # the attribute is not allowed 561 562 $current = $allowed_html[strtolower($element)][strtolower($arreach['name'])]; 563 if ($current == '') 564 continue; # the attribute is not allowed 565 566 if (!is_array($current)) 567 $attr2 .= ' '.$arreach['whole']; 568 # there are no checks 569 570 else { 571 # there are some checks 572 $ok = true; 573 foreach ($current as $currkey => $currval) 574 if (!wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval)) { 575 $ok = false; 576 break; 577 } 578 579 if ( $arreach['name'] == 'style' ) { 580 $orig_value = $arreach['value']; 581 582 $value = safecss_filter_attr($orig_value); 583 584 if ( empty($value) ) 585 continue; 586 587 $arreach['value'] = $value; 588 589 $arreach['whole'] = str_replace($orig_value, $value, $arreach['whole']); 590 } 591 592 if ($ok) 593 $attr2 .= ' '.$arreach['whole']; # it passed them 594 } # if !is_array($current) 595 } # foreach 596 597 # Remove any "<" or ">" characters 598 599 $attr2 = preg_replace('/[<>]/', '', $attr2); 600 601 return "<$element$attr2$xhtml_slash>"; 602 } 603 604 /** 605 * Builds an attribute list from string containing attributes. 606 * 607 * This function does a lot of work. It parses an attribute list into an array 608 * with attribute data, and tries to do the right thing even if it gets weird 609 * input. It will add quotes around attribute values that don't have any quotes 610 * or apostrophes around them, to make it easier to produce HTML code that will 611 * conform to W3C's HTML specification. It will also remove bad URL protocols 612 * from attribute values. It also reduces duplicate attributes by using the 613 * attribute defined first (foo='bar' foo='baz' will result in foo='bar'). 614 * 615 * @since 1.0.0 616 * 617 * @param string $attr Attribute list from HTML element to closing HTML element tag 618 * @param array $allowed_protocols Allowed protocols to keep 619 * @return array List of attributes after parsing 620 */ 621 function wp_kses_hair($attr, $allowed_protocols) { 622 $attrarr = array (); 623 $mode = 0; 624 $attrname = ''; 625 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action'); 626 627 # Loop through the whole attribute list 628 629 while (strlen($attr) != 0) { 630 $working = 0; # Was the last operation successful? 631 632 switch ($mode) { 633 case 0 : # attribute name, href for instance 634 635 if (preg_match('/^([-a-zA-Z]+)/', $attr, $match)) { 636 $attrname = $match[1]; 637 $working = $mode = 1; 638 $attr = preg_replace('/^[-a-zA-Z]+/', '', $attr); 639 } 640 641 break; 642 643 case 1 : # equals sign or valueless ("selected") 644 645 if (preg_match('/^\s*=\s*/', $attr)) # equals sign 646 { 647 $working = 1; 648 $mode = 2; 649 $attr = preg_replace('/^\s*=\s*/', '', $attr); 650 break; 651 } 652 653 if (preg_match('/^\s+/', $attr)) # valueless 654 { 655 $working = 1; 656 $mode = 0; 657 if(FALSE === array_key_exists($attrname, $attrarr)) { 658 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y'); 659 } 660 $attr = preg_replace('/^\s+/', '', $attr); 661 } 662 663 break; 664 665 case 2 : # attribute value, a URL after href= for instance 666 667 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match)) 668 # "value" 669 { 670 $thisval = $match[1]; 671 if ( in_array($attrname, $uris) ) 672 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols); 673 674 if(FALSE === array_key_exists($attrname, $attrarr)) { 675 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n'); 676 } 677 $working = 1; 678 $mode = 0; 679 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr); 680 break; 681 } 682 683 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match)) 684 # 'value' 685 { 686 $thisval = $match[1]; 687 if ( in_array($attrname, $uris) ) 688 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols); 689 690 if(FALSE === array_key_exists($attrname, $attrarr)) { 691 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n'); 692 } 693 $working = 1; 694 $mode = 0; 695 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr); 696 break; 697 } 698 699 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match)) 700 # value 701 { 702 $thisval = $match[1]; 703 if ( in_array($attrname, $uris) ) 704 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols); 705 706 if(FALSE === array_key_exists($attrname, $attrarr)) { 707 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n'); 708 } 709 # We add quotes to conform to W3C's HTML spec. 710 $working = 1; 711 $mode = 0; 712 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr); 713 } 714 715 break; 716 } # switch 717 718 if ($working == 0) # not well formed, remove and try again 719 { 720 $attr = wp_kses_html_error($attr); 721 $mode = 0; 722 } 723 } # while 724 725 if ($mode == 1 && FALSE === array_key_exists($attrname, $attrarr)) 726 # special case, for when the attribute list ends with a valueless 727 # attribute like "selected" 728 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y'); 729 730 return $attrarr; 731 } 732 733 /** 734 * Performs different checks for attribute values. 735 * 736 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval" 737 * and "valueless" with even more checks to come soon. 738 * 739 * @since 1.0.0 740 * 741 * @param string $value Attribute value 742 * @param string $vless Whether the value is valueless. Use 'y' or 'n' 743 * @param string $checkname What $checkvalue is checking for. 744 * @param mixed $checkvalue What constraint the value should pass 745 * @return bool Whether check passes 746 */ 747 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) { 748 $ok = true; 749 750 switch (strtolower($checkname)) { 751 case 'maxlen' : 752 # The maxlen check makes sure that the attribute value has a length not 753 # greater than the given value. This can be used to avoid Buffer Overflows 754 # in WWW clients and various Internet servers. 755 756 if (strlen($value) > $checkvalue) 757 $ok = false; 758 break; 759 760 case 'minlen' : 761 # The minlen check makes sure that the attribute value has a length not 762 # smaller than the given value. 763 764 if (strlen($value) < $checkvalue) 765 $ok = false; 766 break; 767 768 case 'maxval' : 769 # The maxval check does two things: it checks that the attribute value is 770 # an integer from 0 and up, without an excessive amount of zeroes or 771 # whitespace (to avoid Buffer Overflows). It also checks that the attribute 772 # value is not greater than the given value. 773 # This check can be used to avoid Denial of Service attacks. 774 775 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value)) 776 $ok = false; 777 if ($value > $checkvalue) 778 $ok = false; 779 break; 780 781 case 'minval' : 782 # The minval check checks that the attribute value is a positive integer, 783 # and that it is not smaller than the given value. 784 785 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value)) 786 $ok = false; 787 if ($value < $checkvalue) 788 $ok = false; 789 break; 790 791 case 'valueless' : 792 # The valueless check checks if the attribute has a value 793 # (like <a href="blah">) or not (<option selected>). If the given value 794 # is a "y" or a "Y", the attribute must not have a value. 795 # If the given value is an "n" or an "N", the attribute must have one. 796 797 if (strtolower($checkvalue) != $vless) 798 $ok = false; 799 break; 800 } # switch 801 802 return $ok; 803 } 804 805 /** 806 * Sanitize string from bad protocols. 807 * 808 * This function removes all non-allowed protocols from the beginning of 809 * $string. It ignores whitespace and the case of the letters, and it does 810 * understand HTML entities. It does its work in a while loop, so it won't be 811 * fooled by a string like "javascript:javascript:alert(57)". 812 * 813 * @since 1.0.0 814 * 815 * @param string $string Content to filter bad protocols from 816 * @param array $allowed_protocols Allowed protocols to keep 817 * @return string Filtered content 818 */ 819 function wp_kses_bad_protocol($string, $allowed_protocols) { 820 $string = wp_kses_no_null($string); 821 $string2 = $string.'a'; 822 823 while ($string != $string2) { 824 $string2 = $string; 825 $string = wp_kses_bad_protocol_once($string, $allowed_protocols); 826 } # while 827 828 return $string; 829 } 830 831 /** 832 * Removes any NULL characters in $string. 833 * 834 * @since 1.0.0 835 * 836 * @param string $string 837 * @return string 838 */ 839 function wp_kses_no_null($string) { 840 $string = preg_replace('/\0+/', '', $string); 841 $string = preg_replace('/(\\\\0)+/', '', $string); 842 843 return $string; 844 } 845 846 /** 847 * Strips slashes from in front of quotes. 848 * 849 * This function changes the character sequence \" to just ". It leaves all 850 * other slashes alone. It's really weird, but the quoting from 851 * preg_replace(//e) seems to require this. 852 * 853 * @since 1.0.0 854 * 855 * @param string $string String to strip slashes 856 * @return string Fixed strings with quoted slashes 857 */ 858 function wp_kses_stripslashes($string) { 859 return preg_replace('%\\\\"%', '"', $string); 860 } 861 862 /** 863 * Goes through an array and changes the keys to all lower case. 864 * 865 * @since 1.0.0 866 * 867 * @param array $inarray Unfiltered array 868 * @return array Fixed array with all lowercase keys 869 */ 870 function wp_kses_array_lc($inarray) { 871 $outarray = array (); 872 873 foreach ( (array) $inarray as $inkey => $inval) { 874 $outkey = strtolower($inkey); 875 $outarray[$outkey] = array (); 876 877 foreach ( (array) $inval as $inkey2 => $inval2) { 878 $outkey2 = strtolower($inkey2); 879 $outarray[$outkey][$outkey2] = $inval2; 880 } # foreach $inval 881 } # foreach $inarray 882 883 return $outarray; 884 } 885 886 /** 887 * Removes the HTML JavaScript entities found in early versions of Netscape 4. 888 * 889 * @since 1.0.0 890 * 891 * @param string $string 892 * @return string 893 */ 894 function wp_kses_js_entities($string) { 895 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string); 896 } 897 898 /** 899 * Handles parsing errors in wp_kses_hair(). 900 * 901 * The general plan is to remove everything to and including some whitespace, 902 * but it deals with quotes and apostrophes as well. 903 * 904 * @since 1.0.0 905 * 906 * @param string $string 907 * @return string 908 */ 909 function wp_kses_html_error($string) { 910 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string); 911 } 912 913 /** 914 * Sanitizes content from bad protocols and other characters. 915 * 916 * This function searches for URL protocols at the beginning of $string, while 917 * handling whitespace and HTML entities. 918 * 919 * @since 1.0.0 920 * 921 * @param string $string Content to check for bad protocols 922 * @param string $allowed_protocols Allowed protocols 923 * @return string Sanitized content 924 */ 925 function wp_kses_bad_protocol_once($string, $allowed_protocols) { 926 global $_kses_allowed_protocols; 927 $_kses_allowed_protocols = $allowed_protocols; 928 929 $string2 = preg_split('/:|:|:/i', $string, 2); 930 if ( isset($string2[1]) && !preg_match('%/\?%', $string2[0]) ) 931 $string = wp_kses_bad_protocol_once2($string2[0]) . trim($string2[1]); 932 else 933 $string = preg_replace_callback('/^((&[^;]*;|[\sA-Za-z0-9])*)'.'(:|:|&#[Xx]3[Aa];)\s*/', 'wp_kses_bad_protocol_once2', $string); 934 935 return $string; 936 } 937 938 /** 939 * Callback for wp_kses_bad_protocol_once() regular expression. 940 * 941 * This function processes URL protocols, checks to see if they're in the 942 * white-list or not, and returns different data depending on the answer. 943 * 944 * @access private 945 * @since 1.0.0 946 * 947 * @param mixed $matches string or preg_replace_callback() matches array to check for bad protocols 948 * @return string Sanitized content 949 */ 950 function wp_kses_bad_protocol_once2($matches) { 951 global $_kses_allowed_protocols; 952 953 if ( is_array($matches) ) { 954 if ( empty($matches[1]) ) 955 return ''; 956 957 $string = $matches[1]; 958 } else { 959 $string = $matches; 960 } 961 962 $string2 = wp_kses_decode_entities($string); 963 $string2 = preg_replace('/\s/', '', $string2); 964 $string2 = wp_kses_no_null($string2); 965 $string2 = strtolower($string2); 966 967 $allowed = false; 968 foreach ( (array) $_kses_allowed_protocols as $one_protocol) 969 if (strtolower($one_protocol) == $string2) { 970 $allowed = true; 971 break; 972 } 973 974 if ($allowed) 975 return "$string2:"; 976 else 977 return ''; 978 } 979 980 /** 981 * Converts and fixes HTML entities. 982 * 983 * This function normalizes HTML entities. It will convert "AT&T" to the correct 984 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on. 985 * 986 * @since 1.0.0 987 * 988 * @param string $string Content to normalize entities 989 * @return string Content with normalized entities 990 */ 991 function wp_kses_normalize_entities($string) { 992 # Disarm all entities by converting & to & 993 994 $string = str_replace('&', '&', $string); 995 996 # Change back the allowed entities in our entity whitelist 997 998 $string = preg_replace_callback('/&([A-Za-z]{2,8});/', 'wp_kses_named_entities', $string); 999 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string); 1000 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string); 1001 1002 return $string; 1003 } 1004 1005 /** 1006 * Callback for wp_kses_normalize_entities() regular expression. 1007 * 1008 * This function only accepts valid named entity references, which are finite, 1009 * case-sensitive, and highly scrutinized by HTML and XML validators. 1010 * 1011 * @since 3.0.0 1012 * 1013 * @param array $matches preg_replace_callback() matches array 1014 * @return string Correctly encoded entity 1015 */ 1016 function wp_kses_named_entities($matches) { 1017 global $allowedentitynames; 1018 1019 if ( empty($matches[1]) ) 1020 return ''; 1021 1022 $i = $matches[1]; 1023 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" ); 1024 } 1025 1026 /** 1027 * Callback for wp_kses_normalize_entities() regular expression. 1028 * 1029 * This function helps wp_kses_normalize_entities() to only accept 16 bit values 1030 * and nothing more for &#number; entities. 1031 * 1032 * @access private 1033 * @since 1.0.0 1034 * 1035 * @param array $matches preg_replace_callback() matches array 1036 * @return string Correctly encoded entity 1037 */ 1038 function wp_kses_normalize_entities2($matches) { 1039 if ( empty($matches[1]) ) 1040 return ''; 1041 1042 $i = $matches[1]; 1043 if (valid_unicode($i)) { 1044 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT); 1045 $i = "&#$i;"; 1046 } else { 1047 $i = "&#$i;"; 1048 } 1049 1050 return $i; 1051 } 1052 1053 /** 1054 * Callback for wp_kses_normalize_entities() for regular expression. 1055 * 1056 * This function helps wp_kses_normalize_entities() to only accept valid Unicode 1057 * numeric entities in hex form. 1058 * 1059 * @access private 1060 * 1061 * @param array $matches preg_replace_callback() matches array 1062 * @return string Correctly encoded entity 1063 */ 1064 function wp_kses_normalize_entities3($matches) { 1065 if ( empty($matches[1]) ) 1066 return ''; 1067 1068 $hexchars = $matches[1]; 1069 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' ); 1070 } 1071 1072 /** 1073 * Helper function to determine if a Unicode value is valid. 1074 * 1075 * @param int $i Unicode value 1076 * @return bool true if the value was a valid Unicode number 1077 */ 1078 function valid_unicode($i) { 1079 return ( $i == 0x9 || $i == 0xa || $i == 0xd || 1080 ($i >= 0x20 && $i <= 0xd7ff) || 1081 ($i >= 0xe000 && $i <= 0xfffd) || 1082 ($i >= 0x10000 && $i <= 0x10ffff) ); 1083 } 1084 1085 /** 1086 * Convert all entities to their character counterparts. 1087 * 1088 * This function decodes numeric HTML entities (A and A). It doesn't do 1089 * anything with other entities like ä, but we don't need them in the URL 1090 * protocol whitelisting system anyway. 1091 * 1092 * @since 1.0.0 1093 * 1094 * @param string $string Content to change entities 1095 * @return string Content after decoded entities 1096 */ 1097 function wp_kses_decode_entities($string) { 1098 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string); 1099 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string); 1100 1101 return $string; 1102 } 1103 1104 /** 1105 * Regex callback for wp_kses_decode_entities() 1106 * 1107 * @param array $match preg match 1108 * @return string 1109 */ 1110 function _wp_kses_decode_entities_chr( $match ) { 1111 return chr( $match[1] ); 1112 } 1113 1114 /** 1115 * Regex callback for wp_kses_decode_entities() 1116 * 1117 * @param array $match preg match 1118 * @return string 1119 */ 1120 function _wp_kses_decode_entities_chr_hexdec( $match ) { 1121 return chr( hexdec( $match[1] ) ); 1122 } 1123 1124 /** 1125 * Sanitize content with allowed HTML Kses rules. 1126 * 1127 * @since 1.0.0 1128 * @uses $allowedtags 1129 * 1130 * @param string $data Content to filter, expected to be escaped with slashes 1131 * @return string Filtered content 1132 */ 1133 function wp_filter_kses($data) { 1134 global $allowedtags; 1135 return addslashes( wp_kses(stripslashes( $data ), $allowedtags) ); 1136 } 1137 1138 /** 1139 * Sanitize content with allowed HTML Kses rules. 1140 * 1141 * @since 2.9.0 1142 * @uses $allowedtags 1143 * 1144 * @param string $data Content to filter, expected to not be escaped 1145 * @return string Filtered content 1146 */ 1147 function wp_kses_data($data) { 1148 global $allowedtags; 1149 return wp_kses( $data , $allowedtags ); 1150 } 1151 1152 /** 1153 * Sanitize content for allowed HTML tags for post content. 1154 * 1155 * Post content refers to the page contents of the 'post' type and not $_POST 1156 * data from forms. 1157 * 1158 * @since 2.0.0 1159 * @uses $allowedposttags 1160 * 1161 * @param string $data Post content to filter, expected to be escaped with slashes 1162 * @return string Filtered post content with allowed HTML tags and attributes intact. 1163 */ 1164 function wp_filter_post_kses($data) { 1165 global $allowedposttags; 1166 return addslashes ( wp_kses(stripslashes( $data ), $allowedposttags) ); 1167 } 1168 1169 /** 1170 * Sanitize content for allowed HTML tags for post content. 1171 * 1172 * Post content refers to the page contents of the 'post' type and not $_POST 1173 * data from forms. 1174 * 1175 * @since 2.9.0 1176 * @uses $allowedposttags 1177 * 1178 * @param string $data Post content to filter 1179 * @return string Filtered post content with allowed HTML tags and attributes intact. 1180 */ 1181 function wp_kses_post($data) { 1182 global $allowedposttags; 1183 return wp_kses( $data , $allowedposttags ); 1184 } 1185 1186 /** 1187 * Strips all of the HTML in the content. 1188 * 1189 * @since 2.1.0 1190 * 1191 * @param string $data Content to strip all HTML from 1192 * @return string Filtered content without any HTML 1193 */ 1194 function wp_filter_nohtml_kses($data) { 1195 return addslashes ( wp_kses(stripslashes( $data ), array()) ); 1196 } 1197 1198 /** 1199 * Adds all Kses input form content filters. 1200 * 1201 * All hooks have default priority. The wp_filter_kses() function is added to 1202 * the 'pre_comment_content' and 'title_save_pre' hooks. 1203 * 1204 * The wp_filter_post_kses() function is added to the 'content_save_pre', 1205 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks. 1206 * 1207 * @since 2.0.0 1208 * @uses add_filter() See description for what functions are added to what hooks. 1209 */ 1210 function kses_init_filters() { 1211 // Normal filtering. 1212 add_filter('pre_comment_content', 'wp_filter_kses'); 1213 add_filter('title_save_pre', 'wp_filter_kses'); 1214 1215 // Post filtering 1216 add_filter('content_save_pre', 'wp_filter_post_kses'); 1217 add_filter('excerpt_save_pre', 'wp_filter_post_kses'); 1218 add_filter('content_filtered_save_pre', 'wp_filter_post_kses'); 1219 } 1220 1221 /** 1222 * Removes all Kses input form content filters. 1223 * 1224 * A quick procedural method to removing all of the filters that kses uses for 1225 * content in WordPress Loop. 1226 * 1227 * Does not remove the kses_init() function from 'init' hook (priority is 1228 * default). Also does not remove kses_init() function from 'set_current_user' 1229 * hook (priority is also default). 1230 * 1231 * @since 2.0.6 1232 */ 1233 function kses_remove_filters() { 1234 // Normal filtering. 1235 remove_filter('pre_comment_content', 'wp_filter_kses'); 1236 remove_filter('title_save_pre', 'wp_filter_kses'); 1237 1238 // Post filtering 1239 remove_filter('content_save_pre', 'wp_filter_post_kses'); 1240 remove_filter('excerpt_save_pre', 'wp_filter_post_kses'); 1241 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses'); 1242 } 1243 1244 /** 1245 * Sets up most of the Kses filters for input form content. 1246 * 1247 * If you remove the kses_init() function from 'init' hook and 1248 * 'set_current_user' (priority is default), then none of the Kses filter hooks 1249 * will be added. 1250 * 1251 * First removes all of the Kses filters in case the current user does not need 1252 * to have Kses filter the content. If the user does not have unfiltered html 1253 * capability, then Kses filters are added. 1254 * 1255 * @uses kses_remove_filters() Removes the Kses filters 1256 * @uses kses_init_filters() Adds the Kses filters back if the user 1257 * does not have unfiltered HTML capability. 1258 * @since 2.0.0 1259 */ 1260 function kses_init() { 1261 kses_remove_filters(); 1262 1263 if (current_user_can('unfiltered_html') == false) 1264 kses_init_filters(); 1265 } 1266 1267 add_action('init', 'kses_init'); 1268 add_action('set_current_user', 'kses_init'); 1269 1270 /** 1271 * Inline CSS filter 1272 * 1273 * @since 2.8.1 1274 */ 1275 function safecss_filter_attr( $css, $deprecated = '' ) { 1276 if ( !empty( $deprecated ) ) 1277 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented 1278 1279 $css = wp_kses_no_null($css); 1280 $css = str_replace(array("\n","\r","\t"), '', $css); 1281 1282 if ( preg_match( '%[\\(&]|/\*%', $css ) ) // remove any inline css containing \ ( & or comments 1283 return ''; 1284 1285 $css_array = split( ';', trim( $css ) ); 1286 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float', 1287 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color', 1288 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left', 1289 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color', 1290 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top', 1291 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side', 1292 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style', 1293 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom', 1294 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom', 1295 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align', 1296 'width' ) ); 1297 1298 if ( empty($allowed_attr) ) 1299 return $css; 1300 1301 $css = ''; 1302 foreach ( $css_array as $css_item ) { 1303 if ( $css_item == '' ) 1304 continue; 1305 $css_item = trim( $css_item ); 1306 $found = false; 1307 if ( strpos( $css_item, ':' ) === false ) { 1308 $found = true; 1309 } else { 1310 $parts = split( ':', $css_item ); 1311 if ( in_array( trim( $parts[0] ), $allowed_attr ) ) 1312 $found = true; 1313 } 1314 if ( $found ) { 1315 if( $css != '' ) 1316 $css .= ';'; 1317 $css .= $css_item; 1318 } 1319 } 1320 1321 return $css; 1322 }
title
Description
Body
title
Description
Body
title
Description
Body
title
Body
| Generated: Mon Apr 5 14:26:09 2010 | Cross-referenced by PHPXref 0.7 |
