| [ Index ] |
PHP Cross Reference of Wordpress 2.7.1 |
[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 your my-hacks.php file You can also override this 28 * in a plugin file. The my-hacks.php is deprecated in its usage. 29 * 30 * @since 1.2.0 31 */ 32 if (!defined('CUSTOM_TAGS')) 33 define('CUSTOM_TAGS', false); 34 35 if (!CUSTOM_TAGS) { 36 /** 37 * Kses global for default allowable HTML tags. 38 * 39 * Can be override by using CUSTOM_TAGS constant. 40 * 41 * @global array $allowedposttags 42 * @since 2.0.0 43 */ 44 $allowedposttags = array( 45 'address' => array(), 46 'a' => array( 47 'class' => array (), 48 'href' => array (), 49 'id' => array (), 50 'title' => array (), 51 'rel' => array (), 52 'rev' => array (), 53 'name' => array (), 54 'target' => array()), 55 'abbr' => array( 56 'class' => array (), 57 'title' => array ()), 58 'acronym' => array( 59 'title' => array ()), 60 'b' => array(), 61 'big' => array(), 62 'blockquote' => array( 63 'id' => array (), 64 'cite' => array (), 65 'class' => array(), 66 'lang' => array(), 67 'xml:lang' => array()), 68 'br' => array ( 69 'class' => array ()), 70 'button' => array( 71 'disabled' => array (), 72 'name' => array (), 73 'type' => array (), 74 'value' => array ()), 75 'caption' => array( 76 'align' => array (), 77 'class' => array ()), 78 'cite' => array ( 79 'class' => array(), 80 'dir' => array(), 81 'lang' => array(), 82 'title' => array ()), 83 'code' => array ( 84 'style' => array()), 85 'col' => array( 86 'align' => array (), 87 'char' => array (), 88 'charoff' => array (), 89 'span' => array (), 90 'dir' => array(), 91 'style' => array (), 92 'valign' => array (), 93 'width' => array ()), 94 'del' => array( 95 'datetime' => array ()), 96 'dd' => array(), 97 'div' => array( 98 'align' => array (), 99 'class' => array (), 100 'dir' => array (), 101 'lang' => array(), 102 'style' => array (), 103 'xml:lang' => array()), 104 'dl' => array(), 105 'dt' => array(), 106 'em' => array(), 107 'fieldset' => array(), 108 'font' => array( 109 'color' => array (), 110 'face' => array (), 111 'size' => array ()), 112 'form' => array( 113 'action' => array (), 114 'accept' => array (), 115 'accept-charset' => array (), 116 'enctype' => array (), 117 'method' => array (), 118 'name' => array (), 119 'target' => array ()), 120 'h1' => array( 121 'align' => array (), 122 'class' => array ()), 123 'h2' => array( 124 'align' => array (), 125 'class' => array ()), 126 'h3' => array( 127 'align' => array (), 128 'class' => array ()), 129 'h4' => array( 130 'align' => array (), 131 'class' => array ()), 132 'h5' => array( 133 'align' => array (), 134 'class' => array ()), 135 'h6' => array( 136 'align' => array (), 137 'class' => array ()), 138 'hr' => array( 139 'align' => array (), 140 'class' => array (), 141 'noshade' => array (), 142 'size' => array (), 143 'width' => array ()), 144 'i' => array(), 145 'img' => array( 146 'alt' => array (), 147 'align' => array (), 148 'border' => array (), 149 'class' => array (), 150 'height' => array (), 151 'hspace' => array (), 152 'longdesc' => array (), 153 'vspace' => array (), 154 'src' => array (), 155 'style' => array (), 156 'width' => array ()), 157 'ins' => array( 158 'datetime' => array (), 159 'cite' => array ()), 160 'kbd' => array(), 161 'label' => array( 162 'for' => array ()), 163 'legend' => array( 164 'align' => array ()), 165 'li' => array ( 166 'align' => array (), 167 'class' => array ()), 168 'p' => array( 169 'class' => array (), 170 'align' => array (), 171 'dir' => array(), 172 'lang' => array(), 173 'style' => array (), 174 'xml:lang' => array()), 175 'pre' => array( 176 'style' => array(), 177 'width' => array ()), 178 'q' => array( 179 'cite' => array ()), 180 's' => array(), 181 'span' => array ( 182 'class' => array (), 183 'dir' => array (), 184 'align' => array (), 185 'lang' => array (), 186 'style' => array (), 187 'title' => array (), 188 'xml:lang' => array()), 189 'strike' => array(), 190 'strong' => array(), 191 'sub' => array(), 192 'sup' => array(), 193 'table' => array( 194 'align' => array (), 195 'bgcolor' => array (), 196 'border' => array (), 197 'cellpadding' => array (), 198 'cellspacing' => array (), 199 'class' => array (), 200 'dir' => array(), 201 'id' => array(), 202 'rules' => array (), 203 'style' => array (), 204 'summary' => array (), 205 'width' => array ()), 206 'tbody' => array( 207 'align' => array (), 208 'char' => array (), 209 'charoff' => array (), 210 'valign' => array ()), 211 'td' => array( 212 'abbr' => array (), 213 'align' => array (), 214 'axis' => array (), 215 'bgcolor' => array (), 216 'char' => array (), 217 'charoff' => array (), 218 'class' => array (), 219 'colspan' => array (), 220 'dir' => array(), 221 'headers' => array (), 222 'height' => array (), 223 'nowrap' => array (), 224 'rowspan' => array (), 225 'scope' => array (), 226 'style' => array (), 227 'valign' => array (), 228 'width' => array ()), 229 'textarea' => array( 230 'cols' => array (), 231 'rows' => array (), 232 'disabled' => array (), 233 'name' => array (), 234 'readonly' => array ()), 235 'tfoot' => array( 236 'align' => array (), 237 'char' => array (), 238 'class' => array (), 239 'charoff' => array (), 240 'valign' => array ()), 241 'th' => array( 242 'abbr' => array (), 243 'align' => array (), 244 'axis' => array (), 245 'bgcolor' => array (), 246 'char' => array (), 247 'charoff' => array (), 248 'class' => array (), 249 'colspan' => array (), 250 'headers' => array (), 251 'height' => array (), 252 'nowrap' => array (), 253 'rowspan' => array (), 254 'scope' => array (), 255 'valign' => array (), 256 'width' => array ()), 257 'thead' => array( 258 'align' => array (), 259 'char' => array (), 260 'charoff' => array (), 261 'class' => array (), 262 'valign' => array ()), 263 'title' => array(), 264 'tr' => array( 265 'align' => array (), 266 'bgcolor' => array (), 267 'char' => array (), 268 'charoff' => array (), 269 'class' => array (), 270 'style' => array (), 271 'valign' => array ()), 272 'tt' => array(), 273 'u' => array(), 274 'ul' => array ( 275 'class' => array (), 276 'style' => array (), 277 'type' => array ()), 278 'ol' => array ( 279 'class' => array (), 280 'start' => array (), 281 'style' => array (), 282 'type' => array ()), 283 'var' => array ()); 284 285 /** 286 * Kses allowed HTML elements. 287 * 288 * @global array $allowedtags 289 * @since 1.0.0 290 */ 291 $allowedtags = array( 292 'a' => array( 293 'href' => array (), 294 'title' => array ()), 295 'abbr' => array( 296 'title' => array ()), 297 'acronym' => array( 298 'title' => array ()), 299 'b' => array(), 300 'blockquote' => array( 301 'cite' => array ()), 302 // 'br' => array(), 303 'cite' => array (), 304 'code' => array(), 305 'del' => array( 306 'datetime' => array ()), 307 // 'dd' => array(), 308 // 'dl' => array(), 309 // 'dt' => array(), 310 'em' => array (), 'i' => array (), 311 // 'ins' => array('datetime' => array(), 'cite' => array()), 312 // 'li' => array(), 313 // 'ol' => array(), 314 // 'p' => array(), 315 'q' => array( 316 'cite' => array ()), 317 'strike' => array(), 318 'strong' => array(), 319 // 'sub' => array(), 320 // 'sup' => array(), 321 // 'u' => array(), 322 // 'ul' => array(), 323 ); 324 } 325 326 /** 327 * Filters content and keeps only allowable HTML elements. 328 * 329 * This function makes sure that only the allowed HTML element names, attribute 330 * names and attribute values plus only sane HTML entities will occur in 331 * $string. You have to remove any slashes from PHP's magic quotes before you 332 * call this function. 333 * 334 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news', 335 * 'irc', 'gopher', 'nntp', 'feed', and finally 'telnet. This covers all common 336 * link protocols, except for 'javascript' which should not be allowed for 337 * untrusted users. 338 * 339 * @since 1.0.0 340 * 341 * @param string $string Content to filter through kses 342 * @param array $allowed_html List of allowed HTML elements 343 * @param array $allowed_protocols Optional. Allowed protocol in links. 344 * @return string Filtered content with only allowed HTML elements 345 */ 346 function wp_kses($string, $allowed_html, $allowed_protocols = array ('http', 'https', 'ftp', 'ftps', 'mailto', 'news', 'irc', 'gopher', 'nntp', 'feed', 'telnet')) { 347 $string = wp_kses_no_null($string); 348 $string = wp_kses_js_entities($string); 349 $string = wp_kses_normalize_entities($string); 350 $allowed_html_fixed = wp_kses_array_lc($allowed_html); 351 $string = wp_kses_hook($string, $allowed_html_fixed, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook 352 return wp_kses_split($string, $allowed_html_fixed, $allowed_protocols); 353 } 354 355 /** 356 * You add any kses hooks here. 357 * 358 * There is currently only one kses WordPress hook and it is called here. All 359 * parameters are passed to the hooks and expected to recieve a string. 360 * 361 * @since 1.0.0 362 * 363 * @param string $string Content to filter through kses 364 * @param array $allowed_html List of allowed HTML elements 365 * @param array $allowed_protocols Allowed protocol in links 366 * @return string Filtered content through 'pre_kses' hook 367 */ 368 function wp_kses_hook($string, $allowed_html, $allowed_protocols) { 369 $string = apply_filters('pre_kses', $string, $allowed_html, $allowed_protocols); 370 return $string; 371 } 372 373 /** 374 * This function returns kses' version number. 375 * 376 * @since 1.0.0 377 * 378 * @return string KSES Version Number 379 */ 380 function wp_kses_version() { 381 return '0.2.2'; 382 } 383 384 /** 385 * Searches for HTML tags, no matter how malformed. 386 * 387 * It also matches stray ">" characters. 388 * 389 * @since 1.0.0 390 * 391 * @param string $string Content to filter 392 * @param array $allowed_html Allowed HTML elements 393 * @param array $allowed_protocols Allowed protocols to keep 394 * @return string Content with fixed HTML tags 395 */ 396 function wp_kses_split($string, $allowed_html, $allowed_protocols) { 397 global $pass_allowed_html, $pass_allowed_protocols; 398 $pass_allowed_html = $allowed_html; 399 $pass_allowed_protocols = $allowed_protocols; 400 return preg_replace_callback('%((<!--.*?(-->|$))|(<[^>]*(>|$)|>))%', 401 create_function('$match', 'global $pass_allowed_html, $pass_allowed_protocols; return wp_kses_split2($match[1], $pass_allowed_html, $pass_allowed_protocols);'), $string); 402 } 403 404 /** 405 * Callback for wp_kses_split for fixing malformed HTML tags. 406 * 407 * This function does a lot of work. It rejects some very malformed things like 408 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no 409 * strip_tags()!). Otherwise it splits the tag into an element and an attribute 410 * list. 411 * 412 * After the tag is split into an element and an attribute list, it is run 413 * through another filter which will remove illegal attributes and once that is 414 * completed, will be returned. 415 * 416 * @access private 417 * @since 1.0.0 418 * @uses wp_kses_attr() 419 * 420 * @param string $string Content to filter 421 * @param array $allowed_html Allowed HTML elements 422 * @param array $allowed_protocols Allowed protocols to keep 423 * @return string Fixed HTML element 424 */ 425 function wp_kses_split2($string, $allowed_html, $allowed_protocols) { 426 $string = wp_kses_stripslashes($string); 427 428 if (substr($string, 0, 1) != '<') 429 return '>'; 430 # It matched a ">" character 431 432 if (preg_match('%^<!--(.*?)(-->)?$%', $string, $matches)) { 433 $string = str_replace(array('<!--', '-->'), '', $matches[1]); 434 while ( $string != $newstring = wp_kses($string, $allowed_html, $allowed_protocols) ) 435 $string = $newstring; 436 if ( $string == '' ) 437 return ''; 438 // prevent multiple dashes in comments 439 $string = preg_replace('/--+/', '-', $string); 440 // prevent three dashes closing a comment 441 $string = preg_replace('/-$/', '', $string); 442 return "<!--{$string}-->"; 443 } 444 # Allow HTML comments 445 446 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches)) 447 return ''; 448 # It's seriously malformed 449 450 $slash = trim($matches[1]); 451 $elem = $matches[2]; 452 $attrlist = $matches[3]; 453 454 if (!@isset($allowed_html[strtolower($elem)])) 455 return ''; 456 # They are using a not allowed HTML element 457 458 if ($slash != '') 459 return "<$slash$elem>"; 460 # No attributes are allowed for closing elements 461 462 return wp_kses_attr("$slash$elem", $attrlist, $allowed_html, $allowed_protocols); 463 } 464 465 /** 466 * Removes all attributes, if none are allowed for this element. 467 * 468 * If some are allowed it calls wp_kses_hair() to split them further, and then 469 * it builds up new HTML code from the data that kses_hair() returns. It also 470 * removes "<" and ">" characters, if there are any left. One more thing it does 471 * is to check if the tag has a closing XHTML slash, and if it does, it puts one 472 * in the returned code as well. 473 * 474 * @since 1.0.0 475 * 476 * @param string $element HTML element/tag 477 * @param string $attr HTML attributes from HTML element to closing HTML element tag 478 * @param array $allowed_html Allowed HTML elements 479 * @param array $allowed_protocols Allowed protocols to keep 480 * @return string Sanitized HTML element 481 */ 482 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) { 483 # Is there a closing XHTML slash at the end of the attributes? 484 485 $xhtml_slash = ''; 486 if (preg_match('%\s/\s*$%', $attr)) 487 $xhtml_slash = ' /'; 488 489 # Are any attributes allowed at all for this element? 490 491 if (@ count($allowed_html[strtolower($element)]) == 0) 492 return "<$element$xhtml_slash>"; 493 494 # Split it 495 496 $attrarr = wp_kses_hair($attr, $allowed_protocols); 497 498 # Go through $attrarr, and save the allowed attributes for this element 499 # in $attr2 500 501 $attr2 = ''; 502 503 foreach ($attrarr as $arreach) { 504 if (!@ isset ($allowed_html[strtolower($element)][strtolower($arreach['name'])])) 505 continue; # the attribute is not allowed 506 507 $current = $allowed_html[strtolower($element)][strtolower($arreach['name'])]; 508 if ($current == '') 509 continue; # the attribute is not allowed 510 511 if (!is_array($current)) 512 $attr2 .= ' '.$arreach['whole']; 513 # there are no checks 514 515 else { 516 # there are some checks 517 $ok = true; 518 foreach ($current as $currkey => $currval) 519 if (!wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval)) { 520 $ok = false; 521 break; 522 } 523 524 if ($ok) 525 $attr2 .= ' '.$arreach['whole']; # it passed them 526 } # if !is_array($current) 527 } # foreach 528 529 # Remove any "<" or ">" characters 530 531 $attr2 = preg_replace('/[<>]/', '', $attr2); 532 533 return "<$element$attr2$xhtml_slash>"; 534 } 535 536 /** 537 * Builds an attribute list from string containing attributes. 538 * 539 * This function does a lot of work. It parses an attribute list into an array 540 * with attribute data, and tries to do the right thing even if it gets weird 541 * input. It will add quotes around attribute values that don't have any quotes 542 * or apostrophes around them, to make it easier to produce HTML code that will 543 * conform to W3C's HTML specification. It will also remove bad URL protocols 544 * from attribute values. It also reduces duplicate attributes by using the 545 * attribute defined first (foo='bar' foo='baz' will result in foo='bar'). 546 * 547 * @since 1.0.0 548 * 549 * @param string $attr Attribute list from HTML element to closing HTML element tag 550 * @param array $allowed_protocols Allowed protocols to keep 551 * @return array List of attributes after parsing 552 */ 553 function wp_kses_hair($attr, $allowed_protocols) { 554 $attrarr = array (); 555 $mode = 0; 556 $attrname = ''; 557 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action'); 558 559 # Loop through the whole attribute list 560 561 while (strlen($attr) != 0) { 562 $working = 0; # Was the last operation successful? 563 564 switch ($mode) { 565 case 0 : # attribute name, href for instance 566 567 if (preg_match('/^([-a-zA-Z]+)/', $attr, $match)) { 568 $attrname = $match[1]; 569 $working = $mode = 1; 570 $attr = preg_replace('/^[-a-zA-Z]+/', '', $attr); 571 } 572 573 break; 574 575 case 1 : # equals sign or valueless ("selected") 576 577 if (preg_match('/^\s*=\s*/', $attr)) # equals sign 578 { 579 $working = 1; 580 $mode = 2; 581 $attr = preg_replace('/^\s*=\s*/', '', $attr); 582 break; 583 } 584 585 if (preg_match('/^\s+/', $attr)) # valueless 586 { 587 $working = 1; 588 $mode = 0; 589 if(FALSE === array_key_exists($attrname, $attrarr)) { 590 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y'); 591 } 592 $attr = preg_replace('/^\s+/', '', $attr); 593 } 594 595 break; 596 597 case 2 : # attribute value, a URL after href= for instance 598 599 if (preg_match('/^"([^"]*)"(\s+|$)/', $attr, $match)) 600 # "value" 601 { 602 $thisval = $match[1]; 603 if ( in_array($attrname, $uris) ) 604 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols); 605 606 if(FALSE === array_key_exists($attrname, $attrarr)) { 607 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n'); 608 } 609 $working = 1; 610 $mode = 0; 611 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr); 612 break; 613 } 614 615 if (preg_match("/^'([^']*)'(\s+|$)/", $attr, $match)) 616 # 'value' 617 { 618 $thisval = $match[1]; 619 if ( in_array($attrname, $uris) ) 620 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols); 621 622 if(FALSE === array_key_exists($attrname, $attrarr)) { 623 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n'); 624 } 625 $working = 1; 626 $mode = 0; 627 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr); 628 break; 629 } 630 631 if (preg_match("%^([^\s\"']+)(\s+|$)%", $attr, $match)) 632 # value 633 { 634 $thisval = $match[1]; 635 if ( in_array($attrname, $uris) ) 636 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols); 637 638 if(FALSE === array_key_exists($attrname, $attrarr)) { 639 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n'); 640 } 641 # We add quotes to conform to W3C's HTML spec. 642 $working = 1; 643 $mode = 0; 644 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr); 645 } 646 647 break; 648 } # switch 649 650 if ($working == 0) # not well formed, remove and try again 651 { 652 $attr = wp_kses_html_error($attr); 653 $mode = 0; 654 } 655 } # while 656 657 if ($mode == 1 && FALSE === array_key_exists($attrname, $attrarr)) 658 # special case, for when the attribute list ends with a valueless 659 # attribute like "selected" 660 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y'); 661 662 return $attrarr; 663 } 664 665 /** 666 * Performs different checks for attribute values. 667 * 668 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval" 669 * and "valueless" with even more checks to come soon. 670 * 671 * @since 1.0.0 672 * 673 * @param string $value Attribute value 674 * @param string $vless Whether the value is valueless or not. Use 'y' or 'n' 675 * @param string $checkname What $checkvalue is checking for. 676 * @param mixed $checkvalue What constraint the value should pass 677 * @return bool Whether check passes (true) or not (false) 678 */ 679 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) { 680 $ok = true; 681 682 switch (strtolower($checkname)) { 683 case 'maxlen' : 684 # The maxlen check makes sure that the attribute value has a length not 685 # greater than the given value. This can be used to avoid Buffer Overflows 686 # in WWW clients and various Internet servers. 687 688 if (strlen($value) > $checkvalue) 689 $ok = false; 690 break; 691 692 case 'minlen' : 693 # The minlen check makes sure that the attribute value has a length not 694 # smaller than the given value. 695 696 if (strlen($value) < $checkvalue) 697 $ok = false; 698 break; 699 700 case 'maxval' : 701 # The maxval check does two things: it checks that the attribute value is 702 # an integer from 0 and up, without an excessive amount of zeroes or 703 # whitespace (to avoid Buffer Overflows). It also checks that the attribute 704 # value is not greater than the given value. 705 # This check can be used to avoid Denial of Service attacks. 706 707 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value)) 708 $ok = false; 709 if ($value > $checkvalue) 710 $ok = false; 711 break; 712 713 case 'minval' : 714 # The minval check checks that the attribute value is a positive integer, 715 # and that it is not smaller than the given value. 716 717 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value)) 718 $ok = false; 719 if ($value < $checkvalue) 720 $ok = false; 721 break; 722 723 case 'valueless' : 724 # The valueless check checks if the attribute has a value 725 # (like <a href="blah">) or not (<option selected>). If the given value 726 # is a "y" or a "Y", the attribute must not have a value. 727 # If the given value is an "n" or an "N", the attribute must have one. 728 729 if (strtolower($checkvalue) != $vless) 730 $ok = false; 731 break; 732 } # switch 733 734 return $ok; 735 } 736 737 /** 738 * Sanitize string from bad protocols. 739 * 740 * This function removes all non-allowed protocols from the beginning of 741 * $string. It ignores whitespace and the case of the letters, and it does 742 * understand HTML entities. It does its work in a while loop, so it won't be 743 * fooled by a string like "javascript:javascript:alert(57)". 744 * 745 * @since 1.0.0 746 * 747 * @param string $string Content to filter bad protocols from 748 * @param array $allowed_protocols Allowed protocols to keep 749 * @return string Filtered content 750 */ 751 function wp_kses_bad_protocol($string, $allowed_protocols) { 752 $string = wp_kses_no_null($string); 753 $string = preg_replace('/\xad+/', '', $string); # deals with Opera "feature" 754 $string2 = $string.'a'; 755 756 while ($string != $string2) { 757 $string2 = $string; 758 $string = wp_kses_bad_protocol_once($string, $allowed_protocols); 759 } # while 760 761 return $string; 762 } 763 764 /** 765 * Removes any NULL characters in $string. 766 * 767 * @since 1.0.0 768 * 769 * @param string $string 770 * @return string 771 */ 772 function wp_kses_no_null($string) { 773 $string = preg_replace('/\0+/', '', $string); 774 $string = preg_replace('/(\\\\0)+/', '', $string); 775 776 return $string; 777 } 778 779 /** 780 * Strips slashes from in front of quotes. 781 * 782 * This function changes the character sequence \" to just ". It leaves all 783 * other slashes alone. It's really weird, but the quoting from 784 * preg_replace(//e) seems to require this. 785 * 786 * @since 1.0.0 787 * 788 * @param string $string String to strip slashes 789 * @return string Fixed strings with quoted slashes 790 */ 791 function wp_kses_stripslashes($string) { 792 return preg_replace('%\\\\"%', '"', $string); 793 } 794 795 /** 796 * Goes through an array and changes the keys to all lower case. 797 * 798 * @since 1.0.0 799 * 800 * @param array $inarray Unfiltered array 801 * @return array Fixed array with all lowercase keys 802 */ 803 function wp_kses_array_lc($inarray) { 804 $outarray = array (); 805 806 foreach ( (array) $inarray as $inkey => $inval) { 807 $outkey = strtolower($inkey); 808 $outarray[$outkey] = array (); 809 810 foreach ( (array) $inval as $inkey2 => $inval2) { 811 $outkey2 = strtolower($inkey2); 812 $outarray[$outkey][$outkey2] = $inval2; 813 } # foreach $inval 814 } # foreach $inarray 815 816 return $outarray; 817 } 818 819 /** 820 * Removes the HTML JavaScript entities found in early versions of Netscape 4. 821 * 822 * @since 1.0.0 823 * 824 * @param string $string 825 * @return string 826 */ 827 function wp_kses_js_entities($string) { 828 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string); 829 } 830 831 /** 832 * Handles parsing errors in wp_kses_hair(). 833 * 834 * The general plan is to remove everything to and including some whitespace, 835 * but it deals with quotes and apostrophes as well. 836 * 837 * @since 1.0.0 838 * 839 * @param string $string 840 * @return string 841 */ 842 function wp_kses_html_error($string) { 843 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string); 844 } 845 846 /** 847 * Sanitizes content from bad protocols and other characters. 848 * 849 * This function searches for URL protocols at the beginning of $string, while 850 * handling whitespace and HTML entities. 851 * 852 * @since 1.0.0 853 * 854 * @param string $string Content to check for bad protocols 855 * @param string $allowed_protocols Allowed protocols 856 * @return string Sanitized content 857 */ 858 function wp_kses_bad_protocol_once($string, $allowed_protocols) { 859 global $_kses_allowed_protocols; 860 $_kses_allowed_protocols = $allowed_protocols; 861 862 $string2 = preg_split('/:|:|:/i', $string, 2); 863 if ( isset($string2[1]) && !preg_match('%/\?%', $string2[0]) ) 864 $string = wp_kses_bad_protocol_once2($string2[0]) . trim($string2[1]); 865 else 866 $string = preg_replace_callback('/^((&[^;]*;|[\sA-Za-z0-9])*)'.'(:|:|&#[Xx]3[Aa];)\s*/', 'wp_kses_bad_protocol_once2', $string); 867 868 return $string; 869 } 870 871 /** 872 * Callback for wp_kses_bad_protocol_once() regular expression. 873 * 874 * This function processes URL protocols, checks to see if they're in the 875 * white-list or not, and returns different data depending on the answer. 876 * 877 * @access private 878 * @since 1.0.0 879 * 880 * @param mixed $matches string or preg_replace_callback() matches array to check for bad protocols 881 * @return string Sanitized content 882 */ 883 function wp_kses_bad_protocol_once2($matches) { 884 global $_kses_allowed_protocols; 885 886 if ( is_array($matches) ) { 887 if ( ! isset($matches[1]) || empty($matches[1]) ) 888 return ''; 889 890 $string = $matches[1]; 891 } else { 892 $string = $matches; 893 } 894 895 $string2 = wp_kses_decode_entities($string); 896 $string2 = preg_replace('/\s/', '', $string2); 897 $string2 = wp_kses_no_null($string2); 898 $string2 = preg_replace('/\xad+/', '', $string2); 899 # deals with Opera "feature" 900 $string2 = strtolower($string2); 901 902 $allowed = false; 903 foreach ( (array) $_kses_allowed_protocols as $one_protocol) 904 if (strtolower($one_protocol) == $string2) { 905 $allowed = true; 906 break; 907 } 908 909 if ($allowed) 910 return "$string2:"; 911 else 912 return ''; 913 } 914 915 /** 916 * Converts and fixes HTML entities. 917 * 918 * This function normalizes HTML entities. It will convert "AT&T" to the correct 919 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on. 920 * 921 * @since 1.0.0 922 * 923 * @param string $string Content to normalize entities 924 * @return string Content with normalized entities 925 */ 926 function wp_kses_normalize_entities($string) { 927 # Disarm all entities by converting & to & 928 929 $string = str_replace('&', '&', $string); 930 931 # Change back the allowed entities in our entity whitelist 932 933 $string = preg_replace('/&([A-Za-z][A-Za-z0-9]{0,19});/', '&\\1;', $string); 934 $string = preg_replace_callback('/&#0*([0-9]{1,5});/', 'wp_kses_normalize_entities2', $string); 935 $string = preg_replace_callback('/&#([Xx])0*(([0-9A-Fa-f]{2}){1,2});/', 'wp_kses_normalize_entities3', $string); 936 937 return $string; 938 } 939 940 /** 941 * Callback for wp_kses_normalize_entities() regular expression. 942 * 943 * This function helps wp_kses_normalize_entities() to only accept 16 bit values 944 * and nothing more for &#number; entities. 945 * 946 * @access private 947 * @since 1.0.0 948 * 949 * @param array $matches preg_replace_callback() matches array 950 * @return string Correctly encoded entity 951 */ 952 function wp_kses_normalize_entities2($matches) { 953 if ( ! isset($matches[1]) || empty($matches[1]) ) 954 return ''; 955 956 $i = $matches[1]; 957 return ( ( ! valid_unicode($i) ) || ($i > 65535) ? "&#$i;" : "&#$i;" ); 958 } 959 960 /** 961 * Callback for wp_kses_normalize_entities() for regular expression. 962 * 963 * This function helps wp_kses_normalize_entities() to only accept valid Unicode 964 * numeric entities in hex form. 965 * 966 * @access private 967 * 968 * @param array $matches preg_replace_callback() matches array 969 * @return string Correctly encoded entity 970 */ 971 function wp_kses_normalize_entities3($matches) { 972 if ( ! isset($matches[2]) || empty($matches[2]) ) 973 return ''; 974 975 $hexchars = $matches[2]; 976 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : "&#x$hexchars;" ); 977 } 978 979 /** 980 * Helper function to determine if a Unicode value is valid. 981 * 982 * @param int $i Unicode value 983 * @return bool true if the value was a valid Unicode number 984 */ 985 function valid_unicode($i) { 986 return ( $i == 0x9 || $i == 0xa || $i == 0xd || 987 ($i >= 0x20 && $i <= 0xd7ff) || 988 ($i >= 0xe000 && $i <= 0xfffd) || 989 ($i >= 0x10000 && $i <= 0x10ffff) ); 990 } 991 992 /** 993 * Convert all entities to their character counterparts. 994 * 995 * This function decodes numeric HTML entities (A and A). It doesn't do 996 * anything with other entities like ä, but we don't need them in the URL 997 * protocol whitelisting system anyway. 998 * 999 * @since 1.0.0 1000 * 1001 * @param string $string Content to change entities 1002 * @return string Content after decoded entities 1003 */ 1004 function wp_kses_decode_entities($string) { 1005 $string = preg_replace_callback('/&#([0-9]+);/', create_function('$match', 'return chr($match[1]);'), $string); 1006 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', create_function('$match', 'return chr(hexdec($match[1]));'), $string); 1007 1008 return $string; 1009 } 1010 1011 /** 1012 * Sanitize content with allowed HTML Kses rules. 1013 * 1014 * @since 1.0.0 1015 * @uses $allowedtags 1016 * 1017 * @param string $data Content to filter 1018 * @return string Filtered content 1019 */ 1020 function wp_filter_kses($data) { 1021 global $allowedtags; 1022 return addslashes( wp_kses(stripslashes( $data ), $allowedtags) ); 1023 } 1024 1025 /** 1026 * Sanitize content for allowed HTML tags for post content. 1027 * 1028 * Post content refers to the page contents of the 'post' type and not $_POST 1029 * data from forms. 1030 * 1031 * @since 2.0.0 1032 * @uses $allowedposttags 1033 * 1034 * @param string $data Post content to filter 1035 * @return string Filtered post content with allowed HTML tags and attributes intact. 1036 */ 1037 function wp_filter_post_kses($data) { 1038 global $allowedposttags; 1039 return addslashes ( wp_kses(stripslashes( $data ), $allowedposttags) ); 1040 } 1041 1042 /** 1043 * Strips all of the HTML in the content. 1044 * 1045 * @since 2.1.0 1046 * 1047 * @param string $data Content to strip all HTML from 1048 * @return string Filtered content without any HTML 1049 */ 1050 function wp_filter_nohtml_kses($data) { 1051 return addslashes ( wp_kses(stripslashes( $data ), array()) ); 1052 } 1053 1054 /** 1055 * Adds all Kses input form content filters. 1056 * 1057 * All hooks have default priority. The wp_filter_kses() function is added to 1058 * the 'pre_comment_content' and 'title_save_pre' hooks. 1059 * 1060 * The wp_filter_post_kses() function is added to the 'content_save_pre', 1061 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks. 1062 * 1063 * @since 2.0.0 1064 * @uses add_filter() See description for what functions are added to what hooks. 1065 */ 1066 function kses_init_filters() { 1067 // Normal filtering. 1068 add_filter('pre_comment_content', 'wp_filter_kses'); 1069 add_filter('title_save_pre', 'wp_filter_kses'); 1070 1071 // Post filtering 1072 add_filter('content_save_pre', 'wp_filter_post_kses'); 1073 add_filter('excerpt_save_pre', 'wp_filter_post_kses'); 1074 add_filter('content_filtered_save_pre', 'wp_filter_post_kses'); 1075 } 1076 1077 /** 1078 * Removes all Kses input form content filters. 1079 * 1080 * A quick procedural method to removing all of the filters that kses uses for 1081 * content in WordPress Loop. 1082 * 1083 * Does not remove the kses_init() function from 'init' hook (priority is 1084 * default). Also does not remove kses_init() function from 'set_current_user' 1085 * hook (priority is also default). 1086 * 1087 * @since 2.0.6 1088 */ 1089 function kses_remove_filters() { 1090 // Normal filtering. 1091 remove_filter('pre_comment_content', 'wp_filter_kses'); 1092 remove_filter('title_save_pre', 'wp_filter_kses'); 1093 1094 // Post filtering 1095 remove_filter('content_save_pre', 'wp_filter_post_kses'); 1096 remove_filter('excerpt_save_pre', 'wp_filter_post_kses'); 1097 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses'); 1098 } 1099 1100 /** 1101 * Sets up most of the Kses filters for input form content. 1102 * 1103 * If you remove the kses_init() function from 'init' hook and 1104 * 'set_current_user' (priority is default), then none of the Kses filter hooks 1105 * will be added. 1106 * 1107 * First removes all of the Kses filters in case the current user does not need 1108 * to have Kses filter the content. If the user does not have unfiltered html 1109 * capability, then Kses filters are added. 1110 * 1111 * @uses kses_remove_filters() Removes the Kses filters 1112 * @uses kses_init_filters() Adds the Kses filters back if the user 1113 * does not have unfiltered HTML capability. 1114 * @since 2.0.0 1115 */ 1116 function kses_init() { 1117 kses_remove_filters(); 1118 1119 if (current_user_can('unfiltered_html') == false) 1120 kses_init_filters(); 1121 } 1122 1123 add_action('init', 'kses_init'); 1124 add_action('set_current_user', 'kses_init'); 1125 ?>
title
Description
Body
title
Description
Body
title
Description
Body
title
Body
| Generated: Mon Mar 23 16:23:02 2009 | Cross-referenced by PHPXref 0.7 |