|
1 <?php |
|
2 |
|
3 /* |
|
4 * Enano - an open-source CMS capable of wiki functions, Drupal-like sidebar blocks, and everything in between |
|
5 * Version 1.1.6 (Caoineag beta 1) |
|
6 * Copyright (C) 2006-2008 Dan Fuhry |
|
7 * |
|
8 * This program is Free Software; you can redistribute and/or modify it under the terms of the GNU General Public License |
|
9 * as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. |
|
10 * |
|
11 * This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied |
|
12 * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for details. |
|
13 * |
|
14 * This script contains code originally found in MediaWiki (http://www.mediawiki.org). MediaWiki is also licensed under |
|
15 * the GPLv2 or later; see the file GPL included with this package for details. |
|
16 */ |
|
17 |
|
18 $attrib = '[a-zA-Z0-9]'; |
|
19 $space = '[\x09\x0a\x0d\x20]'; |
|
20 |
|
21 define( 'MW_CHAR_REFS_REGEX', |
|
22 '/&([A-Za-z0-9]+); |
|
23 |&\#([0-9]+); |
|
24 |&\#x([0-9A-Za-z]+); |
|
25 |&\#X([0-9A-Za-z]+); |
|
26 |(&)/x' ); |
|
27 |
|
28 define( 'MW_ATTRIBS_REGEX', |
|
29 "/(?:^|$space)($attrib+) |
|
30 ($space*=$space* |
|
31 (?: |
|
32 # The attribute value: quoted or alone |
|
33 ".'"'."([^<".'"'."]*)".'"'." |
|
34 | '([^<']*)' |
|
35 | ([a-zA-Z0-9!#$%&()*,\\-.\\/:;<>?@[\\]^_`{|}~]+) |
|
36 | (\#[0-9a-fA-F]+) # Technically wrong, but lots of |
|
37 # colors are specified like this. |
|
38 # We'll be normalizing it. |
|
39 ) |
|
40 )?(?=$space|\$)/sx" ); |
|
41 |
|
42 /** |
|
43 * Take a tag soup fragment listing an HTML element's attributes |
|
44 * and normalize it to well-formed XML, discarding unwanted attributes. |
|
45 * Output is safe for further wikitext processing, with escaping of |
|
46 * values that could trigger problems. |
|
47 * |
|
48 * - Normalizes attribute names to lowercase |
|
49 * - Discards attributes not on a whitelist for the given element |
|
50 * - Turns broken or invalid entities into plaintext |
|
51 * - Double-quotes all attribute values |
|
52 * - Attributes without values are given the name as attribute |
|
53 * - Double attributes are discarded |
|
54 * - Unsafe style attributes are discarded |
|
55 * - Prepends space if there are attributes. |
|
56 * |
|
57 * @param string $text |
|
58 * @param string $element |
|
59 * @return string |
|
60 */ |
|
61 function fixTagAttributes( $text, $element ) { |
|
62 if( trim( $text ) == '' ) { |
|
63 return ''; |
|
64 } |
|
65 |
|
66 $stripped = validateTagAttributes( |
|
67 decodeTagAttributes( $text ), $element ); |
|
68 |
|
69 $attribs = array(); |
|
70 foreach( $stripped as $attribute => $value ) { |
|
71 $encAttribute = htmlspecialchars( $attribute ); |
|
72 $encValue = safeEncodeAttribute( $value ); |
|
73 |
|
74 $attribs[] = "$encAttribute=".'"'."$encValue".'"'.""; // " |
|
75 } |
|
76 return count( $attribs ) ? ' ' . implode( ' ', $attribs ) : ''; |
|
77 } |
|
78 |
|
79 /** |
|
80 * Encode an attribute value for HTML tags, with extra armoring |
|
81 * against further wiki processing. |
|
82 * @param $text |
|
83 * @return HTML-encoded text fragment |
|
84 */ |
|
85 function safeEncodeAttribute( $text ) { |
|
86 $encValue= encodeAttribute( $text ); |
|
87 |
|
88 # Templates and links may be expanded in later parsing, |
|
89 # creating invalid or dangerous output. Suppress this. |
|
90 $encValue = strtr( $encValue, array( |
|
91 '<' => '<', // This should never happen, |
|
92 '>' => '>', // we've received invalid input |
|
93 '"' => '"', // which should have been escaped. |
|
94 '{' => '{', |
|
95 '[' => '[', |
|
96 "''" => '''', |
|
97 'ISBN' => 'ISBN', |
|
98 'RFC' => 'RFC', |
|
99 'PMID' => 'PMID', |
|
100 '|' => '|', |
|
101 '__' => '__', |
|
102 ) ); |
|
103 |
|
104 return $encValue; |
|
105 } |
|
106 |
|
107 /** |
|
108 * Encode an attribute value for HTML output. |
|
109 * @param $text |
|
110 * @return HTML-encoded text fragment |
|
111 */ |
|
112 function encodeAttribute( $text ) { |
|
113 |
|
114 // In Enano 1.0.3, added this cheapo hack to keep ampersands |
|
115 // from being double-sanitized. Thanks to markybob from #deluge. |
|
116 |
|
117 // htmlspecialchars() the "manual" way |
|
118 $encValue = strtr( $text, array( |
|
119 '&' => '&', |
|
120 '"' => '"', |
|
121 '<' => '<', |
|
122 '>' => '>', |
|
123 ''' => "'" |
|
124 ) ); |
|
125 |
|
126 $encValue = strtr( $text, array( |
|
127 '&' => '&', |
|
128 '"' => '"', |
|
129 '<' => '<', |
|
130 '>' => '>', |
|
131 "'" => ''' |
|
132 ) ); |
|
133 |
|
134 |
|
135 // Whitespace is normalized during attribute decoding, |
|
136 // so if we've been passed non-spaces we must encode them |
|
137 // ahead of time or they won't be preserved. |
|
138 $encValue = strtr( $encValue, array( |
|
139 "\n" => ' ', |
|
140 "\r" => ' ', |
|
141 "\t" => '	', |
|
142 ) ); |
|
143 |
|
144 return $encValue; |
|
145 } |
|
146 |
|
147 function unstripForHTML( $text ) { |
|
148 global $mStripState; |
|
149 $text = unstrip( $text, $mStripState ); |
|
150 $text = unstripNoWiki( $text, $mStripState ); |
|
151 return $text; |
|
152 } |
|
153 |
|
154 /** |
|
155 * Always call this after unstrip() to preserve the order |
|
156 * |
|
157 * @private |
|
158 */ |
|
159 function unstripNoWiki( $text, &$state ) { |
|
160 if ( !isset( $state['nowiki'] ) ) { |
|
161 return $text; |
|
162 } |
|
163 |
|
164 # TODO: good candidate for FSS |
|
165 $text = strtr( $text, $state['nowiki'] ); |
|
166 |
|
167 return $text; |
|
168 } |
|
169 |
|
170 /** |
|
171 * Take an array of attribute names and values and normalize or discard |
|
172 * illegal values for the given element type. |
|
173 * |
|
174 * - Discards attributes not on a whitelist for the given element |
|
175 * - Unsafe style attributes are discarded |
|
176 * |
|
177 * @param array $attribs |
|
178 * @param string $element |
|
179 * @return array |
|
180 * |
|
181 * @todo Check for legal values where the DTD limits things. |
|
182 * @todo Check for unique id attribute :P |
|
183 */ |
|
184 function validateTagAttributes( $attribs, $element ) { |
|
185 $whitelist = array_flip( attributeWhitelist( $element ) ); |
|
186 $out = array(); |
|
187 foreach( $attribs as $attribute => $value ) { |
|
188 if( !isset( $whitelist[$attribute] ) ) { |
|
189 continue; |
|
190 } |
|
191 # Strip javascript "expression" from stylesheets. |
|
192 # http://msdn.microsoft.com/workshop/author/dhtml/overview/recalc.asp |
|
193 if( $attribute == 'style' ) { |
|
194 $value = checkCss( $value ); |
|
195 if( $value === false ) { |
|
196 # haxx0r |
|
197 continue; |
|
198 } |
|
199 } |
|
200 |
|
201 if ( $attribute === 'id' ) |
|
202 $value = escapeId( $value ); |
|
203 |
|
204 // If this attribute was previously set, override it. |
|
205 // Output should only have one attribute of each name. |
|
206 $out[$attribute] = $value; |
|
207 } |
|
208 return $out; |
|
209 } |
|
210 |
|
211 /** |
|
212 * Pick apart some CSS and check it for forbidden or unsafe structures. |
|
213 * Returns a sanitized string, or false if it was just too evil. |
|
214 * |
|
215 * Currently URL references, 'expression', 'tps' are forbidden. |
|
216 * |
|
217 * @param string $value |
|
218 * @return mixed |
|
219 */ |
|
220 function checkCss( $value ) { |
|
221 $stripped = decodeCharReferences( $value ); |
|
222 |
|
223 // Remove any comments; IE gets token splitting wrong |
|
224 $stripped = preg_replace( '!/\\*.*?\\*/!S', '', $stripped ); |
|
225 $value = $stripped; |
|
226 |
|
227 // ... and continue checks |
|
228 $stripped = preg_replace( '!\\\\([0-9A-Fa-f]{1,6})[ \\n\\r\\t\\f]?!e', |
|
229 'codepointToUtf8(hexdec("$1"))', $stripped ); |
|
230 $stripped = str_replace( '\\', '', $stripped ); |
|
231 if( preg_match( '/(expression|tps*:\/\/|url\\s*\().*/is', |
|
232 $stripped ) ) { |
|
233 # haxx0r |
|
234 return false; |
|
235 } |
|
236 |
|
237 return $value; |
|
238 } |
|
239 |
|
240 /** |
|
241 * Decode any character references, numeric or named entities, |
|
242 * in the text and return a UTF-8 string. |
|
243 * |
|
244 * @param string $text |
|
245 * @return string |
|
246 * @access public |
|
247 * @static |
|
248 */ |
|
249 function decodeCharReferences( $text ) { |
|
250 return preg_replace_callback( |
|
251 MW_CHAR_REFS_REGEX, |
|
252 'decodeCharReferencesCallback', |
|
253 $text ); |
|
254 } |
|
255 |
|
256 /** |
|
257 * Fetch the whitelist of acceptable attributes for a given |
|
258 * element name. |
|
259 * |
|
260 * @param string $element |
|
261 * @return array |
|
262 */ |
|
263 function attributeWhitelist( $element ) { |
|
264 static $list; |
|
265 if( !isset( $list ) ) { |
|
266 $list = setupAttributeWhitelist(); |
|
267 } |
|
268 return isset( $list[$element] ) |
|
269 ? $list[$element] |
|
270 : array(); |
|
271 } |
|
272 |
|
273 /** |
|
274 * @todo Document it a bit |
|
275 * @return array |
|
276 */ |
|
277 function setupAttributeWhitelist() { |
|
278 global $db, $session, $paths, $template, $plugins; |
|
279 $common = array( 'id', 'class', 'lang', 'dir', 'title', 'style' ); |
|
280 $block = array_merge( $common, array( 'align' ) ); |
|
281 $tablealign = array( 'align', 'char', 'charoff', 'valign' ); |
|
282 $tablecell = array( 'abbr', |
|
283 'axis', |
|
284 'headers', |
|
285 'scope', |
|
286 'rowspan', |
|
287 'colspan', |
|
288 'nowrap', # deprecated |
|
289 'width', # deprecated |
|
290 'height', # deprecated |
|
291 'bgcolor' # deprecated |
|
292 ); |
|
293 |
|
294 # Numbers refer to sections in HTML 4.01 standard describing the element. |
|
295 # See: http://www.w3.org/TR/html4/ |
|
296 $whitelist = array ( |
|
297 # 7.5.4 |
|
298 'div' => $block, |
|
299 'center' => $common, # deprecated |
|
300 'span' => $block, # ?? |
|
301 |
|
302 # 7.5.5 |
|
303 'h1' => $block, |
|
304 'h2' => $block, |
|
305 'h3' => $block, |
|
306 'h4' => $block, |
|
307 'h5' => $block, |
|
308 'h6' => $block, |
|
309 |
|
310 # 7.5.6 |
|
311 # address |
|
312 |
|
313 # 8.2.4 |
|
314 # bdo |
|
315 |
|
316 # 9.2.1 |
|
317 'em' => $common, |
|
318 'strong' => $common, |
|
319 'cite' => $common, |
|
320 # dfn |
|
321 'code' => $common, |
|
322 # samp |
|
323 # kbd |
|
324 'var' => $common, |
|
325 # abbr |
|
326 # acronym |
|
327 |
|
328 # 9.2.2 |
|
329 'blockquote' => array_merge( $common, array( 'cite' ) ), |
|
330 # q |
|
331 |
|
332 # 9.2.3 |
|
333 'sub' => $common, |
|
334 'sup' => $common, |
|
335 |
|
336 # 9.3.1 |
|
337 'p' => $block, |
|
338 |
|
339 # 9.3.2 |
|
340 'br' => array( 'id', 'class', 'title', 'style', 'clear' ), |
|
341 |
|
342 # 9.3.4 |
|
343 'pre' => array_merge( $common, array( 'width' ) ), |
|
344 |
|
345 # 9.4 |
|
346 'ins' => array_merge( $common, array( 'cite', 'datetime' ) ), |
|
347 'del' => array_merge( $common, array( 'cite', 'datetime' ) ), |
|
348 |
|
349 # 10.2 |
|
350 'ul' => array_merge( $common, array( 'type' ) ), |
|
351 'ol' => array_merge( $common, array( 'type', 'start' ) ), |
|
352 'li' => array_merge( $common, array( 'type', 'value' ) ), |
|
353 |
|
354 # 10.3 |
|
355 'dl' => $common, |
|
356 'dd' => $common, |
|
357 'dt' => $common, |
|
358 |
|
359 # 11.2.1 |
|
360 'table' => array_merge( $common, |
|
361 array( 'summary', 'width', 'border', 'frame', |
|
362 'rules', 'cellspacing', 'cellpadding', |
|
363 'align', 'bgcolor', |
|
364 ) ), |
|
365 |
|
366 # 11.2.2 |
|
367 'caption' => array_merge( $common, array( 'align' ) ), |
|
368 |
|
369 # 11.2.3 |
|
370 'thead' => array_merge( $common, $tablealign ), |
|
371 'tfoot' => array_merge( $common, $tablealign ), |
|
372 'tbody' => array_merge( $common, $tablealign ), |
|
373 |
|
374 # 11.2.4 |
|
375 'colgroup' => array_merge( $common, array( 'span', 'width' ), $tablealign ), |
|
376 'col' => array_merge( $common, array( 'span', 'width' ), $tablealign ), |
|
377 |
|
378 # 11.2.5 |
|
379 'tr' => array_merge( $common, array( 'bgcolor' ), $tablealign ), |
|
380 |
|
381 # 11.2.6 |
|
382 'td' => array_merge( $common, $tablecell, $tablealign ), |
|
383 'th' => array_merge( $common, $tablecell, $tablealign ), |
|
384 |
|
385 # 12.2 |
|
386 # added by dan |
|
387 'a' => array_merge( $common, array( 'href', 'name' ) ), |
|
388 |
|
389 # 13.2 |
|
390 # added by dan |
|
391 'img' => array_merge( $common, array( 'src', 'width', 'height', 'alt' ) ), |
|
392 |
|
393 # 15.2.1 |
|
394 'tt' => $common, |
|
395 'b' => $common, |
|
396 'i' => $common, |
|
397 'big' => $common, |
|
398 'small' => $common, |
|
399 'strike' => $common, |
|
400 's' => $common, |
|
401 'u' => $common, |
|
402 |
|
403 # 15.2.2 |
|
404 'font' => array_merge( $common, array( 'size', 'color', 'face' ) ), |
|
405 # basefont |
|
406 |
|
407 # 15.3 |
|
408 'hr' => array_merge( $common, array( 'noshade', 'size', 'width' ) ), |
|
409 |
|
410 # XHTML Ruby annotation text module, simple ruby only. |
|
411 # http://www.w3c.org/TR/ruby/ |
|
412 'ruby' => $common, |
|
413 # rbc |
|
414 # rtc |
|
415 'rb' => $common, |
|
416 'rt' => $common, #array_merge( $common, array( 'rbspan' ) ), |
|
417 'rp' => $common, |
|
418 |
|
419 # For compatibility with the XHTML parser. |
|
420 'nowiki' => array(), |
|
421 'noinclude' => array(), |
|
422 'nodisplay' => array(), |
|
423 'lang' => array('code'), |
|
424 |
|
425 # XHTML stuff |
|
426 'acronym' => $common |
|
427 ); |
|
428 |
|
429 // custom tags can be added by plugins |
|
430 $code = $plugins->setHook('html_attribute_whitelist'); |
|
431 foreach ( $code as $cmd ) |
|
432 { |
|
433 eval($cmd); |
|
434 } |
|
435 |
|
436 return $whitelist; |
|
437 } |
|
438 |
|
439 /** |
|
440 * Given a value escape it so that it can be used in an id attribute and |
|
441 * return it, this does not validate the value however (see first link) |
|
442 * |
|
443 * @link http://www.w3.org/TR/html401/types.html#type-name Valid characters |
|
444 * in the id and |
|
445 * name attributes |
|
446 * @link http://www.w3.org/TR/html401/struct/links.html#h-12.2.3 Anchors with the id attribute |
|
447 * |
|
448 * @bug 4461 |
|
449 * |
|
450 * @static |
|
451 * |
|
452 * @param string $id |
|
453 * @return string |
|
454 */ |
|
455 function escapeId( $id ) { |
|
456 static $replace = array( |
|
457 '%3A' => ':', |
|
458 '%' => '.' |
|
459 ); |
|
460 |
|
461 $id = urlencode( decodeCharReferences( strtr( $id, ' ', '_' ) ) ); |
|
462 |
|
463 return str_replace( array_keys( $replace ), array_values( $replace ), $id ); |
|
464 } |
|
465 |
|
466 /** |
|
467 * More or less "markup-safe" explode() |
|
468 * Ignores any instances of the separator inside <...> |
|
469 * @param string $separator |
|
470 * @param string $text |
|
471 * @return array |
|
472 */ |
|
473 function wfExplodeMarkup( $separator, $text ) { |
|
474 $placeholder = "\x00"; |
|
475 |
|
476 // Just in case... |
|
477 $text = str_replace( $placeholder, '', $text ); |
|
478 |
|
479 // Trim stuff |
|
480 $replacer = new ReplacerCallback( $separator, $placeholder ); |
|
481 $cleaned = preg_replace_callback( '/(<.*?>)/', array( $replacer, 'go' ), $text ); |
|
482 |
|
483 $items = explode( $separator, $cleaned ); |
|
484 foreach( $items as $i => $str ) { |
|
485 $items[$i] = str_replace( $placeholder, $separator, $str ); |
|
486 } |
|
487 |
|
488 return $items; |
|
489 } |
|
490 |
|
491 class ReplacerCallback { |
|
492 function ReplacerCallback( $from, $to ) { |
|
493 $this->from = $from; |
|
494 $this->to = $to; |
|
495 } |
|
496 |
|
497 function go( $matches ) { |
|
498 return str_replace( $this->from, $this->to, $matches[1] ); |
|
499 } |
|
500 } |
|
501 |
|
502 /** |
|
503 * Return an associative array of attribute names and values from |
|
504 * a partial tag string. Attribute names are forces to lowercase, |
|
505 * character references are decoded to UTF-8 text. |
|
506 * |
|
507 * @param string |
|
508 * @return array |
|
509 */ |
|
510 function decodeTagAttributes( $text ) { |
|
511 $attribs = array(); |
|
512 |
|
513 if( trim( $text ) == '' ) { |
|
514 return $attribs; |
|
515 } |
|
516 |
|
517 $pairs = array(); |
|
518 if( !preg_match_all( |
|
519 MW_ATTRIBS_REGEX, |
|
520 $text, |
|
521 $pairs, |
|
522 PREG_SET_ORDER ) ) { |
|
523 return $attribs; |
|
524 } |
|
525 |
|
526 foreach( $pairs as $set ) { |
|
527 $attribute = strtolower( $set[1] ); |
|
528 $value = getTagAttributeCallback( $set ); |
|
529 |
|
530 // Normalize whitespace |
|
531 $value = preg_replace( '/[\t\r\n ]+/', ' ', $value ); |
|
532 $value = trim( $value ); |
|
533 |
|
534 // Decode character references |
|
535 $attribs[$attribute] = decodeCharReferences( $value ); |
|
536 } |
|
537 return $attribs; |
|
538 } |
|
539 |
|
540 /** |
|
541 * Pick the appropriate attribute value from a match set from the |
|
542 * MW_ATTRIBS_REGEX matches. |
|
543 * |
|
544 * @param array $set |
|
545 * @return string |
|
546 * @access private |
|
547 */ |
|
548 function getTagAttributeCallback( $set ) { |
|
549 if( isset( $set[6] ) ) { |
|
550 # Illegal #XXXXXX color with no quotes. |
|
551 return $set[6]; |
|
552 } elseif( isset( $set[5] ) ) { |
|
553 # No quotes. |
|
554 return $set[5]; |
|
555 } elseif( isset( $set[4] ) ) { |
|
556 # Single-quoted |
|
557 return $set[4]; |
|
558 } elseif( isset( $set[3] ) ) { |
|
559 # Double-quoted |
|
560 return $set[3]; |
|
561 } elseif( !isset( $set[2] ) ) { |
|
562 # In XHTML, attributes must have a value. |
|
563 # For 'reduced' form, return explicitly the attribute name here. |
|
564 return $set[1]; |
|
565 } else { |
|
566 die_friendly('Parser error', "<p>Tag conditions not met. This should never happen and is a bug.</p>" ); |
|
567 } |
|
568 } |
|
569 |
|
570 /** |
|
571 * Strips and renders nowiki, pre, math, hiero |
|
572 * If $render is set, performs necessary rendering operations on plugins |
|
573 * Returns the text, and fills an array with data needed in unstrip() |
|
574 * If the $state is already a valid strip state, it adds to the state |
|
575 * |
|
576 * @param bool $stripcomments when set, HTML comments <!-- like this --> |
|
577 * will be stripped in addition to other tags. This is important |
|
578 * for section editing, where these comments cause confusion when |
|
579 * counting the sections in the wikisource |
|
580 * |
|
581 * @param array dontstrip contains tags which should not be stripped; |
|
582 * used to prevent stipping of <gallery> when saving (fixes bug 2700) |
|
583 * |
|
584 * @access private |
|
585 */ |
|
586 function mwStrip( $text, &$state, $stripcomments = false , $dontstrip = array () ) { |
|
587 global $wgRandomKey; |
|
588 $render = true; |
|
589 |
|
590 $wgRandomKey = "\x07UNIQ" . dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff)); |
|
591 $uniq_prefix =& $wgRandomKey; |
|
592 $commentState = array(); |
|
593 |
|
594 $elements = array( 'nowiki', 'gallery' ); |
|
595 |
|
596 # Removing $dontstrip tags from $elements list (currently only 'gallery', fixing bug 2700) |
|
597 foreach ( $elements AS $k => $v ) { |
|
598 if ( !in_array ( $v , $dontstrip ) ) continue; |
|
599 unset ( $elements[$k] ); |
|
600 } |
|
601 |
|
602 $matches = array(); |
|
603 $text = extractTagsAndParams( $elements, $text, $matches, $uniq_prefix ); |
|
604 |
|
605 foreach( $matches as $marker => $data ) { |
|
606 list( $element, $content, $params, $tag ) = $data; |
|
607 if( $render ) { |
|
608 $tagName = strtolower( $element ); |
|
609 switch( $tagName ) { |
|
610 case '!--': |
|
611 // Comment |
|
612 if( substr( $tag, -3 ) == '-->' ) { |
|
613 $output = $tag; |
|
614 } else { |
|
615 // Unclosed comment in input. |
|
616 // Close it so later stripping can remove it |
|
617 $output = "$tag-->"; |
|
618 } |
|
619 break; |
|
620 case 'html': |
|
621 if( $wgRawHtml ) { |
|
622 $output = $content; |
|
623 break; |
|
624 } |
|
625 // Shouldn't happen otherwise. :) |
|
626 case 'nowiki': |
|
627 $output = wfEscapeHTMLTagsOnly( $content ); |
|
628 break; |
|
629 default: |
|
630 } |
|
631 } else { |
|
632 // Just stripping tags; keep the source |
|
633 $output = $tag; |
|
634 } |
|
635 |
|
636 // Unstrip the output, because unstrip() is no longer recursive so |
|
637 // it won't do it itself |
|
638 $output = unstrip( $output, $state ); |
|
639 |
|
640 if( !$stripcomments && $element == '!--' ) { |
|
641 $commentState[$marker] = $output; |
|
642 } elseif ( $element == 'html' || $element == 'nowiki' ) { |
|
643 $state['nowiki'][$marker] = $output; |
|
644 } else { |
|
645 $state['general'][$marker] = $output; |
|
646 } |
|
647 } |
|
648 |
|
649 # Unstrip comments unless explicitly told otherwise. |
|
650 # (The comments are always stripped prior to this point, so as to |
|
651 # not invoke any extension tags / parser hooks contained within |
|
652 # a comment.) |
|
653 if ( !$stripcomments ) { |
|
654 // Put them all back and forget them |
|
655 $text = strtr( $text, $commentState ); |
|
656 } |
|
657 |
|
658 return $text; |
|
659 } |
|
660 |
|
661 /** |
|
662 * Replaces all occurrences of HTML-style comments and the given tags |
|
663 * in the text with a random marker and returns teh next text. The output |
|
664 * parameter $matches will be an associative array filled with data in |
|
665 * the form: |
|
666 * 'UNIQ-xxxxx' => array( |
|
667 * 'element', |
|
668 * 'tag content', |
|
669 * array( 'param' => 'x' ), |
|
670 * '<element param="x">tag content</element>' ) ) |
|
671 * |
|
672 * @param $elements list of element names. Comments are always extracted. |
|
673 * @param $text Source text string. |
|
674 * @param $uniq_prefix |
|
675 * |
|
676 * @access private |
|
677 * @static |
|
678 */ |
|
679 function extractTagsAndParams($elements, $text, &$matches, $uniq_prefix = ''){ |
|
680 static $n = 1; |
|
681 $stripped = ''; |
|
682 $matches = array(); |
|
683 |
|
684 $taglist = implode( '|', $elements ); |
|
685 $start = "/<($taglist)(\\s+[^>]*?|\\s*?)(\/?>)|<(!--)/i"; |
|
686 |
|
687 while ( '' != $text ) { |
|
688 $p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); |
|
689 $stripped .= $p[0]; |
|
690 if( count( $p ) < 5 ) { |
|
691 break; |
|
692 } |
|
693 if( count( $p ) > 5 ) { |
|
694 // comment |
|
695 $element = $p[4]; |
|
696 $attributes = ''; |
|
697 $close = ''; |
|
698 $inside = $p[5]; |
|
699 } else { |
|
700 // tag |
|
701 $element = $p[1]; |
|
702 $attributes = $p[2]; |
|
703 $close = $p[3]; |
|
704 $inside = $p[4]; |
|
705 } |
|
706 |
|
707 $marker = "$uniq_prefix-$element-" . sprintf('%08X', $n++) . '-QINU'; |
|
708 $stripped .= $marker; |
|
709 |
|
710 if ( $close === '/>' ) { |
|
711 // Empty element tag, <tag /> |
|
712 $content = null; |
|
713 $text = $inside; |
|
714 $tail = null; |
|
715 } else { |
|
716 if( $element == '!--' ) { |
|
717 $end = '/(-->)/'; |
|
718 } else { |
|
719 $end = "/(<\\/$element\\s*>)/i"; |
|
720 } |
|
721 $q = preg_split( $end, $inside, 2, PREG_SPLIT_DELIM_CAPTURE ); |
|
722 $content = $q[0]; |
|
723 if( count( $q ) < 3 ) { |
|
724 # No end tag -- let it run out to the end of the text. |
|
725 $tail = ''; |
|
726 $text = ''; |
|
727 } else { |
|
728 $tail = $q[1]; |
|
729 $text = $q[2]; |
|
730 } |
|
731 } |
|
732 |
|
733 $matches[$marker] = array( $element, |
|
734 $content, |
|
735 decodeTagAttributes( $attributes ), |
|
736 "<$element$attributes$close$content$tail" ); |
|
737 } |
|
738 return $stripped; |
|
739 } |
|
740 |
|
741 /** |
|
742 * Escape html tags |
|
743 * Basically replacing " > and < with HTML entities ( ", >, <) |
|
744 * |
|
745 * @param $in String: text that might contain HTML tags. |
|
746 * @return string Escaped string |
|
747 */ |
|
748 function wfEscapeHTMLTagsOnly( $in ) { |
|
749 return str_replace( |
|
750 array( '"', '>', '<' ), |
|
751 array( '"', '>', '<' ), |
|
752 $in ); |
|
753 } |
|
754 |
|
755 /** |
|
756 * Restores pre, math, and other extensions removed by strip() |
|
757 * |
|
758 * always call unstripNoWiki() after this one |
|
759 * @private |
|
760 */ |
|
761 function unstrip( $text, &$state ) { |
|
762 if ( !isset( $state['general'] ) ) { |
|
763 return $text; |
|
764 } |
|
765 |
|
766 # TODO: good candidate for FSS |
|
767 $text = strtr( $text, $state['general'] ); |
|
768 |
|
769 return $text; |
|
770 } |
|
771 |
|
772 /** |
|
773 * Return UTF-8 string for a codepoint if that is a valid |
|
774 * character reference, otherwise U+FFFD REPLACEMENT CHARACTER. |
|
775 * @param int $codepoint |
|
776 * @return string |
|
777 * @private |
|
778 */ |
|
779 function decodeChar( $codepoint ) { |
|
780 if( validateCodepoint( $codepoint ) ) { |
|
781 return codepointToUtf8( $codepoint ); |
|
782 } else { |
|
783 return UTF8_REPLACEMENT; |
|
784 } |
|
785 } |
|
786 |
|
787 /** |
|
788 * If the named entity is defined in the HTML 4.0/XHTML 1.0 DTD, |
|
789 * return the UTF-8 encoding of that character. Otherwise, returns |
|
790 * pseudo-entity source (eg &foo;) |
|
791 * |
|
792 * @param string $name |
|
793 * @return string |
|
794 */ |
|
795 function decodeEntity( $name ) { |
|
796 global $wgHtmlEntities; |
|
797 if( isset( $wgHtmlEntities[$name] ) ) { |
|
798 return codepointToUtf8( $wgHtmlEntities[$name] ); |
|
799 } else { |
|
800 return "&$name;"; |
|
801 } |
|
802 } |
|
803 |
|
804 /** |
|
805 * Returns true if a given Unicode codepoint is a valid character in XML. |
|
806 * @param int $codepoint |
|
807 * @return bool |
|
808 */ |
|
809 function validateCodepoint( $codepoint ) { |
|
810 return ($codepoint == 0x09) |
|
811 || ($codepoint == 0x0a) |
|
812 || ($codepoint == 0x0d) |
|
813 || ($codepoint >= 0x20 && $codepoint <= 0xd7ff) |
|
814 || ($codepoint >= 0xe000 && $codepoint <= 0xfffd) |
|
815 || ($codepoint >= 0x10000 && $codepoint <= 0x10ffff); |
|
816 } |
|
817 |
|
818 /** |
|
819 * Return UTF-8 sequence for a given Unicode code point. |
|
820 * May die if fed out of range data. |
|
821 * |
|
822 * @param $codepoint Integer: |
|
823 * @return String |
|
824 * @public |
|
825 */ |
|
826 function codepointToUtf8( $codepoint ) { |
|
827 if($codepoint < 0x80) return chr($codepoint); |
|
828 if($codepoint < 0x800) return chr($codepoint >> 6 & 0x3f | 0xc0) . |
|
829 chr($codepoint & 0x3f | 0x80); |
|
830 if($codepoint < 0x10000) return chr($codepoint >> 12 & 0x0f | 0xe0) . |
|
831 chr($codepoint >> 6 & 0x3f | 0x80) . |
|
832 chr($codepoint & 0x3f | 0x80); |
|
833 if($codepoint < 0x110000) return chr($codepoint >> 18 & 0x07 | 0xf0) . |
|
834 chr($codepoint >> 12 & 0x3f | 0x80) . |
|
835 chr($codepoint >> 6 & 0x3f | 0x80) . |
|
836 chr($codepoint & 0x3f | 0x80); |
|
837 |
|
838 echo "Asked for code outside of range ($codepoint)\n"; |
|
839 die( -1 ); |
|
840 } |
|
841 |
|
842 /** |
|
843 * @param string $matches |
|
844 * @return string |
|
845 */ |
|
846 function decodeCharReferencesCallback( $matches ) { |
|
847 if( $matches[1] != '' ) { |
|
848 return decodeEntity( $matches[1] ); |
|
849 } elseif( $matches[2] != '' ) { |
|
850 return decodeChar( intval( $matches[2] ) ); |
|
851 } elseif( $matches[3] != '' ) { |
|
852 return decodeChar( hexdec( $matches[3] ) ); |
|
853 } elseif( $matches[4] != '' ) { |
|
854 return decodeChar( hexdec( $matches[4] ) ); |
|
855 } |
|
856 # Last case should be an ampersand by itself |
|
857 return $matches[0]; |
|
858 } |
|
859 |