Added syntax highlighting class
[Packages/TYPO3.CMS.git] / t3lib / class.t3lib_parsehtml.php
1 <?php
2 /***************************************************************
3 * Copyright notice
4 *
5 * (c) 1999-2003 Kasper Skaarhoj (kasper@typo3.com)
6 * All rights reserved
7 *
8 * This script is part of the TYPO3 project. The TYPO3 project is
9 * free software; you can redistribute it and/or modify
10 * it under the terms of the GNU General Public License as published by
11 * the Free Software Foundation; either version 2 of the License, or
12 * (at your option) any later version.
13 *
14 * The GNU General Public License can be found at
15 * http://www.gnu.org/copyleft/gpl.html.
16 * A copy is found in the textfile GPL.txt and important notices to the license
17 * from the author is found in LICENSE.txt distributed with these scripts.
18 *
19 *
20 * This script is distributed in the hope that it will be useful,
21 * but WITHOUT ANY WARRANTY; without even the implied warranty of
22 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23 * GNU General Public License for more details.
24 *
25 * This copyright notice MUST APPEAR in all copies of the script!
26 ***************************************************************/
27 /**
28 * Contains class with functions for parsing HTML code.
29 *
30 * $Id$
31 * Revised for TYPO3 3.6 July/2003 by Kasper Skaarhoj
32 *
33 * @author Kasper Skaarhoj <kasper@typo3.com>
34 */
35 /**
36 * [CLASS/FUNCTION INDEX of SCRIPT]
37 *
38 *
39 *
40 * 100: class t3lib_parsehtml
41 * 117: function getSubpart($content, $marker)
42 * 145: function substituteSubpart($content,$marker,$subpartContent,$recursive=1,$keepMarker=0)
43 * 210: function splitIntoBlock($tag,$content,$eliminateExtraEndTags=0)
44 * 269: function splitTags($tag,$content)
45 * 303: function getAllParts($parts,$tag_parts=1,$include_tag=1)
46 * 322: function removeFirstAndLastTag($str)
47 * 341: function getFirstTag($str)
48 * 355: function getFirstTagName($str)
49 * 373: function checkTagTypeCounts($content,$blockTags='a,b,blockquote,body,div,em,font,form,h1,h2,h3,h4,h5,h6,i,li,map,ol,option,p,pre,select,span,strong,table,td,textarea,tr,u,ul', $soloTags='br,hr,img,input,area')
50 * 447: function HTMLcleaner($content, $tags=array(),$keepAll=0,$hSC=0,$addConfig=array())
51 * 618: function get_tag_attributes($tag,$deHSC=0)
52 * 659: function split_tag_attributes($tag)
53 * 696: function bidir_htmlspecialchars($value,$dir)
54 * 716: function prefixResourcePath($main_prefix,$content,$alternatives=array())
55 * 784: function prefixRelPath($prefix,$srcVal)
56 * 802: function cleanFontTags($value,$keepFace=0,$keepSize=0,$keepColor=0)
57 * 833: function mapTags($value,$tags=array(),$ltChar='<',$ltChar2='<')
58 * 850: function unprotectTags($content,$tagList='')
59 * 883: function stripTagsExcept($value,$tagList)
60 * 906: function caseShift($str,$flag,$cacheKey='')
61 * 930: function compileTagAttribs($tagAttrib,$meta=array(), $xhtmlClean=0)
62 * 959: function get_tag_attributes_classic($tag,$deHSC=0)
63 * 972: function HTMLparserConfig($TSconfig,$keepTags=array())
64 * 1085: function XHTML_clean($content)
65 * 1107: function processTag($value,$conf,$endTag,$protected=0)
66 * 1153: function processContent($value,$dir,$conf)
67 *
68 * TOTAL FUNCTIONS: 26
69 * (This index is automatically created/updated by the extension "extdeveval")
70 *
71 */
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92 /**
93 * Functions for parsing HTML.
94 * You are encouraged to use this class in your own applications
95 *
96 * @author Kasper Skaarhoj <kasper@typo3.com>
97 * @package TYPO3
98 * @subpackage t3lib
99 */
100 class t3lib_parsehtml {
101 var $caseShift_cache=array();
102
103
104 // *******************************************'
105 // COPY FROM class.tslib_content.php: / BEGIN
106 // substituteSubpart
107 // Cleaned locally 2/2003 !!!! (so different from tslib_content version)
108 // *******************************************'
109
110 /**
111 * Returns the first subpart encapsulated in the marker, $marker (possibly present in $content as a HTML comment)
112 *
113 * @param string Content with subpart wrapped in fx. "###CONTENT_PART###" inside.
114 * @param string Marker string, eg. "###CONTENT_PART###"
115 * @return string
116 */
117 function getSubpart($content, $marker) {
118 if ($marker && strstr($content,$marker)) {
119 $start = strpos($content, $marker)+strlen($marker);
120 $stop = @strpos($content, $marker, $start+1);
121 $sub = substr($content, $start, $stop-$start);
122
123 $reg=Array();
124 ereg('^[^<]*-->',$sub,$reg);
125 $start+=strlen($reg[0]);
126
127 $reg=Array();
128 ereg('<!--[^>]*$',$sub,$reg);
129 $stop-=strlen($reg[0]);
130
131 return substr($content, $start, $stop-$start);
132 }
133 }
134
135 /**
136 * Substitutes a subpart in $content with the content of $subpartContent.
137 *
138 * @param string Content with subpart wrapped in fx. "###CONTENT_PART###" inside.
139 * @param string Marker string, eg. "###CONTENT_PART###"
140 * @param array If $subpartContent happens to be an array, it's [0] and [1] elements are wrapped around the content of the subpart (fetched by getSubpart())
141 * @param boolean If $recursive is set, the function calls itself with the content set to the remaining part of the content after the second marker. This means that proceding subparts are ALSO substituted!
142 * @param boolean If set, the marker around the subpart is not removed, but kept in the output
143 * @return string Processed input content
144 */
145 function substituteSubpart($content,$marker,$subpartContent,$recursive=1,$keepMarker=0) {
146 $start = strpos($content, $marker);
147 $stop = @strpos($content, $marker, $start+1)+strlen($marker);
148 if ($start && $stop>$start) {
149 // code before
150 $before = substr($content, 0, $start);
151 $reg=Array();
152 ereg('<!--[^>]*$',$before,$reg);
153 $start-=strlen($reg[0]);
154 if ($keepMarker) {
155 $reg_k=Array();
156 if ($reg[0]) ereg('^[^>]*-->',substr($content,$start),$reg_k);
157 $before_marker = substr($content, $start, strlen($reg_k[0]?$reg_k[0]:$marker));
158 }
159 $before = substr($content, 0, $start);
160 // code after
161 $after = substr($content, $stop);
162 $reg=Array();
163 ereg('^[^<]*-->',$after,$reg);
164 $stop+=strlen($reg[0]);
165 if ($keepMarker) {
166 $reg_k=Array();
167 if ($reg[0]) ereg('<!--[^<]*$',substr($content,0,$stop),$reg_k);
168 $sLen = strlen($reg_k[0]?$reg_k[0]:$marker);
169 $after_marker = substr($content, $stop-$sLen,$sLen);
170 }
171 $after = substr($content, $stop);
172
173
174 // replace?
175 if (is_array($subpartContent)) {
176 $substContent=$subpartContent[0].$this->getSubpart($content,$marker).$subpartContent[1];
177 } else {
178 $substContent=$subpartContent;
179 }
180
181 if ($recursive && strpos($after, $marker)) {
182 return $before.($keepMarker?$before_marker:'').$substContent.($keepMarker?$after_marker:'').$this->substituteSubpart($after,$marker,$subpartContent);
183 } else {
184 return $before.($keepMarker?$before_marker:'').$substContent.($keepMarker?$after_marker:'').$after;
185 }
186 } else {
187 return $content;
188 }
189 }
190 // *******************************************'
191 // COPY FROM class.tslib_content.php: / END
192 // *******************************************'
193
194
195
196
197
198
199 /**
200 * Returns an array with the $content divided by tag-blocks specified with the list of tags, $tag
201 * Even numbers in the array are outside the blocks, Odd numbers are block-content.
202 * Use ->getAllParts() and ->removeFirstAndLastTag() to process the content if needed.
203 *
204 * @param string List of tags, comma separated.
205 * @param string HTML-content
206 * @param boolean If set, excessive end tags are ignored - you should probably set this in most cases.
207 * @return array Even numbers in the array are outside the blocks, Odd numbers are block-content.
208 * @see splitTags(), getAllParts(), removeFirstAndLastTag()
209 */
210 function splitIntoBlock($tag,$content,$eliminateExtraEndTags=0) {
211 $tags=array_unique(t3lib_div::trimExplode(',',$tag,1));
212 $regexStr = '</?('.implode('|',$tags).')(>|[[:space:]][^>]*>)';
213
214 $parts = spliti($regexStr,$content);
215
216 $newParts=array();
217 $pointer=strlen($parts[0]);
218 $buffer=$parts[0];
219 $nested=0;
220 reset($parts);
221 next($parts);
222 while(list($k,$v)=each($parts)) {
223 $isEndTag= substr($content,$pointer,2)=='</' ? 1 : 0;
224 $tagLen = strcspn(substr($content,$pointer),'>')+1;
225
226 if (!$isEndTag) { // We meet a start-tag:
227 if (!$nested) { // Ground level:
228 $newParts[]=$buffer; // previous buffer stored
229 $buffer='';
230 }
231 $nested++; // We are inside now!
232 $mbuffer=substr($content,$pointer,strlen($v)+$tagLen); // New buffer set and pointer increased
233 $pointer+=strlen($mbuffer);
234 $buffer.=$mbuffer;
235 } else { // If we meet an endtag:
236 $nested--; // decrease nested-level
237 $eliminated=0;
238 if ($eliminateExtraEndTags && $nested<0) {
239 $nested=0;
240 $eliminated=1;
241 } else {
242 $buffer.=substr($content,$pointer,$tagLen); // In any case, add the endtag to current buffer and increase pointer
243 }
244 $pointer+=$tagLen;
245 if (!$nested && !$eliminated) { // if we're back on ground level, (and not by eliminating tags...
246 $newParts[]=$buffer;
247 $buffer='';
248 }
249 $mbuffer=substr($content,$pointer,strlen($v)); // New buffer set and pointer increased
250 $pointer+=strlen($mbuffer);
251 $buffer.=$mbuffer;
252 }
253
254 }
255 $newParts[]=$buffer;
256 return $newParts;
257 }
258
259 /**
260 * Returns an array with the $content divided by tag-blocks specified with the list of tags, $tag
261 * Even numbers in the array are outside the blocks, Odd numbers are block-content.
262 * Use ->getAllParts() and ->removeFirstAndLastTag() to process the content if needed.
263 *
264 * @param string List of tags
265 * @param string HTML-content
266 * @return array Even numbers in the array are outside the blocks, Odd numbers are block-content.
267 * @see splitIntoBlock(), getAllParts(), removeFirstAndLastTag()
268 */
269 function splitTags($tag,$content) {
270 $tags=t3lib_div::trimExplode(',',$tag,1);
271 $regexStr = '<('.implode('|',$tags).')(>|[[:space:]][^>]*>)';
272 $parts = spliti($regexStr,$content);
273
274 $pointer=strlen($parts[0]);
275 $newParts=array();
276 $newParts[]=$parts[0];
277 reset($parts);
278 next($parts);
279 while(list($k,$v)=each($parts)) {
280 $tagLen = strcspn(substr($content,$pointer),'>')+1;
281
282 // Set tag:
283 $tag = substr($content,$pointer,$tagLen); // New buffer set and pointer increased
284 $newParts[]=$tag;
285 $pointer+=strlen($tag);
286
287 // Set content:
288 $newParts[]=$v;
289 $pointer+=strlen($v);
290 }
291 return $newParts;
292 }
293
294 /**
295 * Returns an array with either tag or non-tag content of the result from ->splitIntoBlock()/->splitTags()
296 *
297 * @param array Parts generated by ->splitIntoBlock() or >splitTags()
298 * @param boolean Whether to return the tag-parts (default,true) or what was outside the tags.
299 * @param boolean Whether to include the tags in the tag-parts (most useful for input made by ->splitIntoBlock())
300 * @return array Tag-parts/Non-tag-parts depending on input argument settings
301 * @see splitIntoBlock(), splitTags()
302 */
303 function getAllParts($parts,$tag_parts=1,$include_tag=1) {
304 reset($parts);
305 $newParts=array();
306 while(list($k,$v)=each($parts)) {
307 if (($k+($tag_parts?0:1))%2) {
308 if (!$include_tag) $v=$this->removeFirstAndLastTag($v);
309 $newParts[]=$v;
310 }
311 }
312 return $newParts;
313 }
314
315 /**
316 * Removes the first and last tag in the string
317 * Anything before and after the first and last tags respectively is also removed
318 *
319 * @param string String to process
320 * @return string
321 */
322 function removeFirstAndLastTag($str) {
323 // First:
324 $endLen = strcspn($str,'>')+1;
325 $str = substr($str,$endLen);
326 // Last:
327 $str = strrev($str);
328 $endLen = strcspn($str,'<')+1;
329 $str = substr($str,$endLen);
330 // return
331 return strrev($str);
332 }
333
334 /**
335 * Returns the first tag in $str
336 * Actually everything from the begining of the $str is returned, so you better make sure the tag is the first thing...
337 *
338 * @param string HTML string with tags
339 * @return string
340 */
341 function getFirstTag($str) {
342 // First:
343 $endLen = strcspn($str,'>')+1;
344 $str = substr($str,0,$endLen);
345 return $str;
346 }
347
348 /**
349 * Returns the NAME of the first tag in $str
350 *
351 * @param string HTML tag (The element name MUST be separated from the attributes by a space character! Just *whitespace* will not do)
352 * @param boolean If set, then the tag is NOT converted to uppercase by case is preserved.
353 * @return string Tag name in upper case
354 * @see getFirstTag()
355 */
356 function getFirstTagName($str,$preserveCase=FALSE) {
357 list($tag) = split('[[:space:]]',substr(trim($this->getFirstTag($str)),1,-1), 2);
358 if (!$preserveCase) $tag = strtoupper($tag);
359
360 return trim($tag);
361 }
362
363 /**
364 * Checks whether block/solo tags are found in the correct amounts in HTML content
365 * Block tags are tags which are required to have an equal amount of start and end tags, eg. "<table>...</table>"
366 * Solo tags are tags which are required to have ONLY start tags (possibly with an XHTML ending like ".../>")
367 * NOTICE: Correct XHTML might actually fail since "<br></br>" is allowed as well as "<br/>". However only the LATTER is accepted by this function (with "br" in the "solo-tag" list), the first example will result in a warning.
368 * NOTICE: Correct XHTML might actually fail since "<p/>" is allowed as well as "<p></p>". However only the LATTER is accepted by this function (with "p" in the "block-tag" list), the first example will result in an ERROR!
369 * NOTICE: Correct HTML version "something" allows eg. <p> and <li> to be NON-ended (implicitly ended by other tags). However this is NOT accepted by this function (with "p" and "li" in the block-tag list) and it will result in an ERROR!
370 *
371 * @param string HTML content to analyze
372 * @param string Tag names for block tags (eg. table or div or p) in lowercase, commalist (eg. "table,div,p")
373 * @param string Tag names for solo tags (eg. img, br or input) in lowercase, commalist ("img,br,input")
374 * @return array Analyse data.
375 */
376 function checkTagTypeCounts($content,$blockTags='a,b,blockquote,body,div,em,font,form,h1,h2,h3,h4,h5,h6,i,li,map,ol,option,p,pre,select,span,strong,table,td,textarea,tr,u,ul', $soloTags='br,hr,img,input,area') {
377 $content = strtolower($content);
378 $analyzedOutput=array();
379 $analyzedOutput['counts']=array(); // Counts appearances of start-tags
380 $analyzedOutput['errors']=array(); // Lists ERRORS
381 $analyzedOutput['warnings']=array(); // Lists warnings.
382 $analyzedOutput['blocks']=array(); // Lists stats for block-tags
383 $analyzedOutput['solo']=array(); // Lists stats for solo-tags
384
385 // Block tags, must have endings...
386 $blockTags = explode(',',$blockTags);
387 foreach($blockTags as $tagName) {
388 $countBegin = count(split('<'.$tagName.'[^[:alnum:]]',$content))-1;
389 $countEnd = count(split('<\/'.$tagName.'[^[:alnum:]]',$content))-1;
390 $analyzedOutput['blocks'][$tagName]=array($countBegin,$countEnd,$countBegin-$countEnd);
391 if ($countBegin) $analyzedOutput['counts'][$tagName]=$countBegin;
392 if ($countBegin-$countEnd) {
393 if ($countBegin-$countEnd > 0) {
394 $analyzedOutput['errors'][$tagName]='There were more start-tags ('.$countBegin.') than end-tags ('.$countEnd.') for the element "'.$tagName.'". There should be an equal amount!';
395 } else {
396 $analyzedOutput['warnings'][$tagName]='There were more end-tags ('.$countEnd.') than start-tags ('.$countBegin.') for the element "'.$tagName.'". There should be an equal amount! However the problem is not fatal.';
397 }
398 }
399 }
400
401 // Solo tags, must NOT have endings...
402 $soloTags = explode(',',$soloTags);
403 foreach($soloTags as $tagName) {
404 $countBegin = count(split('<'.$tagName.'[^[:alnum:]]',$content))-1;
405 $countEnd = count(split('<\/'.$tagName.'[^[:alnum:]]',$content))-1;
406 $analyzedOutput['solo'][$tagName]=array($countBegin,$countEnd);
407 if ($countBegin) $analyzedOutput['counts'][$tagName]=$countBegin;
408 if ($countEnd) {
409 $analyzedOutput['warnings'][$tagName]='There were end-tags found ('.$countEnd.') for the element "'.$tagName.'". This was not expected (although XHTML technically allows it).';
410 }
411 }
412
413 return $analyzedOutput;
414 }
415
416 /**
417 * Function that can clean up HTML content according to configuration given in the $tags array.
418 *
419 * Initializing the $tags array to allow a list of tags (in this case <B>,<I>,<U> and <A>), set it like this: $tags = array_flip(explode(',','b,a,i,u'))
420 * If the value of the $tags[$tagname] entry is an array, advanced processing of the tags is initialized. These are the options:
421 *
422 * $tags[$tagname] = Array(
423 * 'overrideAttribs' => '' If set, this string is preset as the attributes of the tag
424 * 'allowedAttribs' => '0' (zero) = no attributes allowed, '[commalist of attributes]' = only allowed attributes. If blank, all attributes are allowed.
425 * 'fixAttrib' => Array(
426 * '[attribute name]' => Array (
427 * 'default' => If no attribute exists by this name, this value is set as default value (if this value is not blank)
428 * 'always' => Boolean. If set, the attribute is always processed. Normally an attribute is processed only if it exists
429 * 'trim,intval,lower,upper' => All booleans. If any of these keys are set, the value is passed through the respective PHP-functions.
430 * 'range' => Array ('[low limit]','[high limit, optional]') Setting integer range.
431 * 'list' => Array ('[value1/default]','[value2]','[value3]') Attribute must be in this list. If not, the value is set to the first element.
432 * 'removeIfFalse' => Boolean/'blank'. If set, then the attribute is removed if it is 'false'. If this value is set to 'blank' then the value must be a blank string (that means a 'zero' value will not be removed)
433 * 'removeIfEquals' => [value] If the attribute value matches the value set here, then it is removed.
434 * 'casesensitiveComp' => 1 If set, then the removeIfEquals and list comparisons will be case sensitive. Otherwise not.
435 * )
436 * ),
437 * 'protect' => '', Boolean. If set, the tag <> is converted to &lt; and &gt;
438 * 'remap' => '', String. If set, the tagname is remapped to this tagname
439 * 'rmTagIfNoAttrib' => '', Boolean. If set, then the tag is removed if no attributes happend to be there.
440 * 'nesting' => '', Boolean/'global'. If set true, then this tag must have starting and ending tags in the correct order. Any tags not in this order will be discarded. Thus '</B><B><I></B></I></B>' will be converted to '<B><I></B></I>'. Is the value 'global' then true nesting in relation to other tags marked for 'global' nesting control is preserved. This means that if <B> and <I> are set for global nesting then this string '</B><B><I></B></I></B>' is converted to '<B></B>'
441 * )
442 *
443 * @param string $content; is the HTML-content being processed. This is also the result being returned.
444 * @param array $tags; is an array where each key is a tagname in lowercase. Only tags present as keys in this array are preserved. The value of the key can be an array with a vast number of options to configure.
445 * @param string $keepAll; boolean/'protect', if set, then all tags are kept regardless of tags present as keys in $tags-array. If 'protect' then the preserved tags have their <> converted to &lt; and &gt;
446 * @param integer $hSC; Values -1,0,1,2: Set to zero= disabled, set to 1 then the content BETWEEN tags is htmlspecialchar()'ed, set to -1 its the opposite and set to 2 the content will be HSC'ed BUT with preservation for real entities (eg. "&amp;" or "&#234;")
447 * @param array Configuration array send along as $conf to the internal functions ->processContent() and ->processTag()
448 * @return string Processed HTML content
449 */
450 function HTMLcleaner($content, $tags=array(),$keepAll=0,$hSC=0,$addConfig=array()) {
451 $newContent = array();
452 $tokArr = explode('<',$content);
453 $newContent[]=$this->processContent(current($tokArr),$hSC,$addConfig);
454 next($tokArr);
455
456 $c=1;
457 $tagRegister=array();
458 $tagStack=array();
459 while(list(,$tok)=each($tokArr)) {
460 $firstChar = substr($tok,0,1);
461 if (strcmp(trim($firstChar),'')) { // It is a tag...
462 $tagEnd = strcspn($tok,'>');
463 if (strlen($tok)!=$tagEnd) { // If there is and end-bracket...
464 $endTag = $firstChar=='/' ? 1 : 0;
465 $tagContent = substr($tok,$endTag,$tagEnd-$endTag);
466 $tagParts = split('[[:space:]]',$tagContent,2);
467 $tagName = strtolower($tagParts[0]);
468 if (isset($tags[$tagName])) {
469 if (is_array($tags[$tagName])) { // If there is processing to do for the tag:
470
471 if (!$endTag) { // If NOT an endtag, do attribute processing (added dec. 2003)
472 // Override attributes
473 if (strcmp($tags[$tagName]['overrideAttribs'],'')) {
474 $tagParts[1]=$tags[$tagName]['overrideAttribs'];
475 }
476
477 // Allowed tags
478 if (strcmp($tags[$tagName]['allowedAttribs'],'')) {
479 if (!strcmp($tags[$tagName]['allowedAttribs'],'0')) { // No attribs allowed
480 $tagParts[1]='';
481 } elseif (trim($tagParts[1])) {
482 $tagAttrib = $this->get_tag_attributes($tagParts[1]);
483 $tagParts[1]='';
484 $newTagAttrib = array();
485 $tList = t3lib_div::trimExplode(',',strtolower($tags[$tagName]['allowedAttribs']),1);
486 while(list(,$allowTag)=each($tList)) {
487 if (isset($tagAttrib[0][$allowTag])) $newTagAttrib[$allowTag]=$tagAttrib[0][$allowTag];
488 }
489 $tagParts[1]=$this->compileTagAttribs($newTagAttrib,$tagAttrib[1]);
490 }
491 }
492
493 // Fixed attrib values
494 if (is_array($tags[$tagName]['fixAttrib'])) {
495 $tagAttrib = $this->get_tag_attributes($tagParts[1]);
496 $tagParts[1]='';
497 reset($tags[$tagName]['fixAttrib']);
498 while(list($attr,$params)=each($tags[$tagName]['fixAttrib'])) {
499 if (strcmp($params['default'],'') && !isset($tagAttrib[0][$attr])) $tagAttrib[0][$attr]=$params['default'];
500 if ($params['always'] || isset($tagAttrib[0][$attr])) {
501 if ($params['trim']) {$tagAttrib[0][$attr]=trim($tagAttrib[0][$attr]);}
502 if ($params['intval']) {$tagAttrib[0][$attr]=intval($tagAttrib[0][$attr]);}
503 if ($params['lower']) {$tagAttrib[0][$attr]=strtolower($tagAttrib[0][$attr]);}
504 if ($params['upper']) {$tagAttrib[0][$attr]=strtoupper($tagAttrib[0][$attr]);}
505 if ($params['range']) {
506 if (isset($params['range'][1])) {
507 $tagAttrib[0][$attr]=t3lib_div::intInRange($tagAttrib[0][$attr],intval($params['range'][0]),intval($params['range'][1]));
508 } else {
509 $tagAttrib[0][$attr]=t3lib_div::intInRange($tagAttrib[0][$attr],intval($params['range'][0]));
510 }
511 }
512 if (is_array($params['list'])) {
513 if (!in_array($this->caseShift($tagAttrib[0][$attr],$params['casesensitiveComp']),$this->caseShift($params['list'],$params['casesensitiveComp'],$tagName))) $tagAttrib[0][$attr]=$params['list'][0];
514 }
515 if (($params['removeIfFalse'] && $params['removeIfFalse']!='blank' && !$tagAttrib[0][$attr]) || ($params['removeIfFalse']=='blank' && !strcmp($tagAttrib[0][$attr],''))) {
516 unset($tagAttrib[0][$attr]);
517 }
518 if (strcmp($params['removeIfEquals'],'') && !strcmp($this->caseShift($tagAttrib[0][$attr],$params['casesensitiveComp']),$this->caseShift($params['removeIfEquals'],$params['casesensitiveComp']))) {
519 unset($tagAttrib[0][$attr]);
520 }
521 }
522 }
523 $tagParts[1]=$this->compileTagAttribs($tagAttrib[0],$tagAttrib[1]);
524 }
525 } else { // If endTag, remove any possible attributes:
526 $tagParts[1]='';
527 }
528
529 // Protecting the tag by converting < and > to &lt; and &gt; ??
530 if ($tags[$tagName]['protect']) {
531 $lt = '&lt;'; $gt = '&gt;';
532 } else {
533 $lt = '<'; $gt = '>';
534 }
535 // Remapping tag name?
536 if ($tags[$tagName]['remap']) $tagParts[0] = $tags[$tagName]['remap'];
537
538 // rmTagIfNoAttrib
539 if ($endTag || trim($tagParts[1]) || !$tags[$tagName]['rmTagIfNoAttrib']) {
540 $setTag=1;
541
542 if ($tags[$tagName]['nesting']) {
543 if (!is_array($tagRegister[$tagName])) $tagRegister[$tagName]=array();
544
545 if ($endTag) {
546 /* if ($tags[$tagName]['nesting']=='global') {
547 $lastEl = end($tagStack);
548 $correctTag = !strcmp($tagName,$lastEl);
549 } else $correctTag=1;
550 */
551 $correctTag=1;
552 if ($tags[$tagName]['nesting']=='global') {
553 $lastEl = end($tagStack);
554 if (strcmp($tagName,$lastEl)) {
555 if (in_array($tagName,$tagStack)) {
556 while(count($tagStack) && strcmp($tagName,$lastEl)) {
557 $elPos = end($tagRegister[$lastEl]);
558 unset($newContent[$elPos]);
559
560 array_pop($tagRegister[$lastEl]);
561 array_pop($tagStack);
562 $lastEl = end($tagStack);
563 }
564 } else {
565 $correctTag=0; // In this case the
566 }
567 }
568 }
569 if (!count($tagRegister[$tagName]) || !$correctTag) {
570 $setTag=0;
571 } else {
572 array_pop($tagRegister[$tagName]);
573 if ($tags[$tagName]['nesting']=='global') {array_pop($tagStack);}
574 }
575 } else {
576 array_push($tagRegister[$tagName],$c);
577 if ($tags[$tagName]['nesting']=='global') {array_push($tagStack,$tagName);}
578 }
579 }
580
581 if ($setTag) {
582 // Setting the tag
583 $newContent[$c++]=$this->processTag($lt.($endTag?'/':'').trim($tagParts[0].' '.$tagParts[1]).$gt,$addConfig,$endTag,$lt=='&lt;');
584 }
585 }
586 } else {
587 $newContent[$c++]=$this->processTag('<'.($endTag?'/':'').$tagContent.'>',$addConfig,$endTag);
588 }
589 } elseif ($keepAll) { // This is if the tag was not defined in the array for processing:
590 if (!strcmp($keepAll,'protect')) {
591 $lt = '&lt;'; $gt = '&gt;';
592 } else {
593 $lt = '<'; $gt = '>';
594 }
595 $newContent[$c++]=$this->processTag($lt.($endTag?'/':'').$tagContent.$gt,$addConfig,$endTag,$lt=='&lt;');
596 }
597 $newContent[$c++]=$this->processContent(substr($tok,$tagEnd+1),$hSC,$addConfig);
598 } else {
599 $newContent[$c++]=$this->processContent('<'.$tok,$hSC,$addConfig); // There were not end-bracket, so no tag...
600 }
601 } else {
602 $newContent[$c++]=$this->processContent('<'.$tok,$hSC,$addConfig); // It was not a tag anyways
603 }
604 }
605
606 // Unsetting tags:
607 reset($tagRegister);
608 while(list($tag,$positions)=each($tagRegister)) {
609 reset($positions);
610 while(list(,$pKey)=each($positions)) {
611 unset($newContent[$pKey]);
612 }
613 }
614
615 return implode('',$newContent);
616 }
617
618 /**
619 * Returns an array with all attributes as keys. Attributes are only lowercase a-z
620 * If a attribute is empty (shorthand), then the value for the key is empty. You can check if it existed with isset()
621 *
622 * @param string Tag: $tag is either a whole tag (eg '<TAG OPTION ATTRIB=VALUE>') or the parameterlist (ex ' OPTION ATTRIB=VALUE>')
623 * @param boolean If set, the attribute values are de-htmlspecialchar'ed. Should actually always be set!
624 * @return array array(Tag attributes,Attribute meta-data)
625 */
626 function get_tag_attributes($tag,$deHSC=0) {
627 list($components,$metaC) = $this->split_tag_attributes($tag);
628 $name = ''; // attribute name is stored here
629 $valuemode = '';
630 $attributes=array();
631 $attributesMeta=array();
632 if (is_array($components)) {
633 while (list($key,$val) = each ($components)) {
634 if ($val != '=') { // Only if $name is set (if there is an attribute, that waits for a value), that valuemode is enabled. This ensures that the attribute is assigned it's value
635 if ($valuemode) {
636 if ($name) {
637 $attributes[$name] = $deHSC?t3lib_div::htmlspecialchars_decode($val):$val;
638 $attributesMeta[$name]['dashType']=$metaC[$key];
639 $name = '';
640 }
641 } else {
642 if ($namekey = ereg_replace('[^a-zA-Z0-9_-]','',$val)) {
643 $name = strtolower($namekey);
644 $attributesMeta[$name]=array();
645 $attributesMeta[$name]['origTag']=$namekey;
646 $attributes[$name] = '';
647 }
648 }
649 $valuemode = '';
650 } else {
651 $valuemode = 'on';
652 }
653 }
654 if (is_array($attributes)) reset($attributes);
655 return array($attributes,$attributesMeta);
656 }
657 }
658
659 /**
660 * Returns an array with the 'components' from an attribute list. The result is normally analyzed by get_tag_attributes
661 * Removes tag-name if found
662 *
663 * @param string The tag or attributes
664 * @return array
665 * @access private
666 * @see t3lib_div::split_tag_attributes()
667 */
668 function split_tag_attributes($tag) {
669 $tag_tmp = trim(eregi_replace ('^<[^[:space:]]*','',trim($tag)));
670 // Removes any > in the end of the string
671 $tag_tmp = trim(eregi_replace ('>$','',$tag_tmp));
672
673 $metaValue = array();
674 $value = array();
675 while (strcmp($tag_tmp,'')) { // Compared with empty string instead , 030102
676 $firstChar=substr($tag_tmp,0,1);
677 if (!strcmp($firstChar,'"') || !strcmp($firstChar,"'")) {
678 $reg=explode($firstChar,$tag_tmp,3);
679 $value[]=$reg[1];
680 $metaValue[]=$firstChar;
681 $tag_tmp=trim($reg[2]);
682 } elseif (!strcmp($firstChar,'=')) {
683 $value[] = '=';
684 $metaValue[]='';
685 $tag_tmp = trim(substr($tag_tmp,1)); // Removes = chars.
686 } else {
687 // There are '' around the value. We look for the next ' ' or '>'
688 $reg = split('[[:space:]=]',$tag_tmp,2);
689 $value[] = trim($reg[0]);
690 $metaValue[]='';
691 $tag_tmp = trim(substr($tag_tmp,strlen($reg[0]),1).$reg[1]);
692 }
693 }
694 if (is_array($value)) reset($value);
695 return array($value,$metaValue);
696 }
697
698 /**
699 * Converts htmlspecialchars forth ($dir=1) AND back ($dir=-1)
700 *
701 * @param string Input value
702 * @param integer Direction: forth ($dir=1, dir=2 for preserving entities) AND back ($dir=-1)
703 * @return string Output value
704 */
705 function bidir_htmlspecialchars($value,$dir) {
706 if ($dir==1) {
707 $value = htmlspecialchars($value);
708 } elseif ($dir==2) {
709 $value = t3lib_div::deHSCentities(htmlspecialchars($value));
710 } elseif ($dir==-1) {
711 $value = str_replace('&gt;','>',$value);
712 $value = str_replace('&lt;','<',$value);
713 $value = str_replace('&quot;','"',$value);
714 $value = str_replace('&amp;','&',$value);
715 }
716 return $value;
717 }
718
719 /**
720 * Prefixes the relative paths of hrefs/src/action in the tags [td,table,body,img,input,form,link,script,a] in the $content with the $main_prefix or and alternative given by $alternatives
721 *
722 * @param string Prefix string
723 * @param string HTML content
724 * @param array Array with alternative prefixes for certain of the tags. key=>value pairs where the keys are the tag element names in uppercase
725 * @return string Processed HTML content
726 */
727 function prefixResourcePath($main_prefix,$content,$alternatives=array()) {
728 $parts = $this->splitTags('td,table,body,img,input,form,link,script,a',$content);
729 reset($parts);
730 while(list($k,$v)=each($parts)) {
731 if ($k%2) {
732 $params = $this->get_tag_attributes($v,1);
733 $tagEnd = substr($v,-2)=='/>' ? ' />' : '>'; // Detect tag-ending so that it is re-applied correctly.
734 $firstTagName = $this->getFirstTagName($v); // The 'name' of the first tag
735 $somethingDone=0;
736 $prefix = isset($alternatives[strtoupper($firstTagName)]) ? $alternatives[strtoupper($firstTagName)] : $main_prefix;
737 switch(strtolower($firstTagName)) {
738 // background - attribute:
739 case 'td':
740 case 'body':
741 case 'table':
742 $src = $params[0]['background'];
743 if ($src) {
744 $params[0]['background'] = $this->prefixRelPath($prefix,$params[0]['background']);
745 $somethingDone=1;
746 }
747 break;
748 // src attribute
749 case 'img':
750 case 'input':
751 case 'script':
752 $src = $params[0]['src'];
753 if ($src) {
754 $params[0]['src'] = $this->prefixRelPath($prefix,$params[0]['src']);
755 $somethingDone=1;
756 }
757 break;
758 case 'link':
759 case 'a':
760 $src = $params[0]['href'];
761 if ($src) {
762 $params[0]['href'] = $this->prefixRelPath($prefix,$params[0]['href']);
763 $somethingDone=1;
764 }
765 break;
766 // action attribute
767 case 'form':
768 $src = $params[0]['action'];
769 if ($src) {
770 $params[0]['action'] = $this->prefixRelPath($prefix,$params[0]['action']);
771 $somethingDone=1;
772 }
773 break;
774 }
775 if ($somethingDone) {
776 $tagParts = split('[[:space:]]',$v,2);
777 $tagParts[1]=$this->compileTagAttribs($params[0],$params[1]);
778 $parts[$k] = '<'.trim(strtolower($firstTagName).' '.$tagParts[1]).
779 $tagEnd;
780 }
781 }
782 }
783
784 return implode('',$parts);
785 }
786
787 /**
788 * Internal sub-function for ->prefixResourcePath()
789 *
790 * @param string Prefix string
791 * @param string Relative path/URL
792 * @return string Output path, prefixed if no scheme in input string
793 * @access private
794 */
795 function prefixRelPath($prefix,$srcVal) {
796 $pU = parse_url($srcVal);
797 if (!$pU['scheme']) { // If not an absolute URL.
798 $srcVal = $prefix.$srcVal;
799 }
800 return $srcVal;
801 }
802
803 /**
804 * Cleans up the input $value for fonttags.
805 * If keepFace,-Size and -Color is set then font-tags with an allowed property is kept. Else deleted.
806 *
807 * @param string HTML content with font-tags inside to clean up.
808 * @param boolean If set, keep "face" attribute
809 * @param boolean If set, keep "size" attribute
810 * @param boolean If set, keep "color" attribute
811 * @return string Processed HTML content
812 */
813 function cleanFontTags($value,$keepFace=0,$keepSize=0,$keepColor=0) {
814 $fontSplit = $this->splitIntoBlock('font',$value); // ,1 ?? - could probably be more stable if splitTags() was used since this depends on end-tags being properly set!
815 reset($fontSplit);
816 while(list($k,$v)=each($fontSplit)) {
817 if ($k%2) { // font:
818 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v));
819 $newAttribs=array();
820 if ($keepFace && $attribArray['face']) $newAttribs[]='face="'.$attribArray['face'].'"';
821 if ($keepSize && $attribArray['size']) $newAttribs[]='size="'.$attribArray['size'].'"';
822 if ($keepColor && $attribArray['color']) $newAttribs[]='color="'.$attribArray['color'].'"';
823
824 $innerContent = $this->cleanFontTags($this->removeFirstAndLastTag($v),$keepFace,$keepSize,$keepColor);
825 if (count($newAttribs)) {
826 $fontSplit[$k]='<font '.implode(' ',$newAttribs).'>'.$innerContent.'</font>';
827 } else {
828 $fontSplit[$k]=$innerContent;
829 }
830 }
831 }
832 return implode('',$fontSplit);
833 }
834
835 /**
836 * This is used to map certain tag-names into other names.
837 *
838 * @param string HTML content
839 * @param array Array with tag key=>value pairs where key is from-tag and value is to-tag
840 * @param string Alternative less-than char to search for (search regex string)
841 * @param string Alternative less-than char to replace with (replace regex string)
842 * @return string Processed HTML content
843 */
844 function mapTags($value,$tags=array(),$ltChar='<',$ltChar2='<') {
845
846 foreach($tags as $from => $to) {
847 $value = eregi_replace($ltChar.$from.'>',$ltChar2.$to.'>',$value);
848 $value = eregi_replace($ltChar.$from.'[[:space:]]([^>]*)>',$ltChar2.$to.' \\1>',$value);
849 $value = eregi_replace($ltChar.'\/'.$from.'[^>]*>',$ltChar2.'/'.$to.'>',$value);
850 }
851 return $value;
852 }
853
854 /**
855 * This converts htmlspecialchar()'ed tags (from $tagList) back to real tags. Eg. '&lt;strong&gt' would be converted back to '<strong>' if found in $tagList
856 *
857 * @param string HTML content
858 * @param string Tag list, separated by comma. Lowercase!
859 * @return string Processed HTML content
860 */
861 function unprotectTags($content,$tagList='') {
862 $tagsArray = t3lib_div::trimExplode(',',$tagList,1);
863 $contentParts = explode('&lt;',$content);
864 next($contentParts); // bypass the first
865 while(list($k,$tok)=each($contentParts)) {
866 $firstChar = substr($tok,0,1);
867 if (strcmp(trim($firstChar),'')) {
868 $subparts = explode('&gt;',$tok,2);
869 $tagEnd = strlen($subparts[0]);
870 if (strlen($tok)!=$tagEnd) {
871 $endTag = $firstChar=='/' ? 1 : 0;
872 $tagContent = substr($tok,$endTag,$tagEnd-$endTag);
873 $tagParts = split('[[:space:]]',$tagContent,2);
874 $tagName = strtolower($tagParts[0]);
875 if (!strcmp($tagList,'') || in_array($tagName,$tagsArray)) {
876 $contentParts[$k] = '<'.$subparts[0].'>'.$subparts[1];
877 } else $contentParts[$k] = '&lt;'.$tok;
878 } else $contentParts[$k] = '&lt;'.$tok;
879 } else $contentParts[$k] = '&lt;'.$tok;
880 }
881
882 return implode('',$contentParts);
883 }
884
885 /**
886 * Strips tags except the tags in the list, $tagList
887 * OBSOLETE - use PHP function strip_tags()
888 *
889 * @param string Value to process
890 * @param string List of tags
891 * @return string Output value
892 * @ignore
893 */
894 function stripTagsExcept($value,$tagList) {
895 $tags=t3lib_div::trimExplode(',',$tagList,1);
896 $forthArr=array();
897 $backArr=array();
898 while(list(,$theTag)=each($tags)) {
899 $forthArr[$theTag]=md5($theTag);
900 $backArr[md5($theTag)]=$theTag;
901 }
902 $value = $this->mapTags($value,$forthArr,'<','_');
903 $value=strip_tags($value);
904 $value = $this->mapTags($value,$backArr,'_','<');
905 return $value;
906 }
907
908 /**
909 * Internal function for case shifting of a string or whole array
910 *
911 * @param mixed Input string/array
912 * @param boolean If $str is a string AND this boolean is true, the string is returned in uppercase
913 * @param string Key string used for internal caching of the results. Could be an MD5 hash of the serialized version of the input $str if that is an array.
914 * @return string Output string, processed
915 * @access private
916 */
917 function caseShift($str,$flag,$cacheKey='') {
918 if (is_array($str)) {
919 if (!$cacheKey || !isset($this->caseShift_cache[$cacheKey])) {
920 reset($str);
921 while(list($k)=each($str)) {
922 $str[$k] = strtoupper($str[$k]);
923 }
924 if ($cacheKey) $this->caseShift_cache[$cacheKey]=$str;
925 } else {
926 $str = $this->caseShift_cache[$cacheKey];
927 }
928 } elseif (!$flag) $str = strtoupper($str);
929 return $str;
930 }
931
932 /**
933 * Compiling an array with tag attributes into a string
934 *
935 * @param array Tag attributes
936 * @param array Meta information about these attributes (like if they were quoted)
937 * @param boolean If set, then the attribute names will be set in lower case, value quotes in double-quotes and the value will be htmlspecialchar()'ed
938 * @return string Imploded attributes, eg: 'attribute="value" attrib2="value2"'
939 * @access private
940 */
941 function compileTagAttribs($tagAttrib,$meta=array(), $xhtmlClean=0) {
942 $accu=array();
943 reset($tagAttrib);
944 while(list($k,$v)=each($tagAttrib)) {
945 if ($xhtmlClean) {
946 $attr=strtolower($k);
947 if (strcmp($v,'') || isset($meta[$k]['dashType'])) {
948 $attr.='="'.htmlspecialchars($v).'"';
949 }
950 } else {
951 $attr=$meta[$k]['origTag']?$meta[$k]['origTag']:$k;
952 if (strcmp($v,'') || isset($meta[$k]['dashType'])) {
953 $dash=$meta[$k]['dashType']?$meta[$k]['dashType']:(t3lib_div::testInt($v)?'':'"');
954 $attr.='='.$dash.$v.$dash;
955 }
956 }
957 $accu[]=$attr;
958 }
959 return implode(' ',$accu);
960 }
961
962 /**
963 * Get tag attributes, the classic version (which had some limitations?)
964 *
965 * @param string The tag
966 * @param boolean De-htmlspecialchar flag.
967 * @return array
968 * @access private
969 */
970 function get_tag_attributes_classic($tag,$deHSC=0) {
971 $attr=$this->get_tag_attributes($tag,$deHSC);
972 return is_array($attr[0])?$attr[0]:array();
973 }
974
975 /**
976 * Converts TSconfig into an array for the HTMLcleaner function.
977 *
978 * @param array TSconfig for HTMLcleaner
979 * @param array Array of tags to keep (?)
980 * @return array
981 * @access private
982 */
983 function HTMLparserConfig($TSconfig,$keepTags=array()) {
984 // Allow tags (base list, merged with incoming array)
985 $alTags = array_flip(t3lib_div::trimExplode(',',strtolower($TSconfig['allowTags']),1));
986 $keepTags = array_merge($alTags,$keepTags);
987
988 // Set config properties.
989 if (is_array($TSconfig['tags.'])) {
990 reset($TSconfig['tags.']);
991 while(list($key,$tagC)=each($TSconfig['tags.'])) {
992 if (!is_array($tagC) && $key==strtolower($key)) {
993 if (!strcmp($tagC,'0')) unset($keepTags[$key]);
994 if (!strcmp($tagC,'1') && !isset($keepTags[$key])) $keepTags[$key]=1;
995 }
996 }
997
998 reset($TSconfig['tags.']);
999 while(list($key,$tagC)=each($TSconfig['tags.'])) {
1000 if (is_array($tagC) && $key==strtolower($key)) {
1001 $key=substr($key,0,-1);
1002 if (!is_array($keepTags[$key])) $keepTags[$key]=array();
1003 if (is_array($tagC['fixAttrib.'])) {
1004 reset($tagC['fixAttrib.']);
1005 while(list($atName,$atConfig)=each($tagC['fixAttrib.'])) {
1006 if (is_array($atConfig)) {
1007 $atName=substr($atName,0,-1);
1008 if (!is_array($keepTags[$key]['fixAttrib'][$atName])) {
1009 $keepTags[$key]['fixAttrib'][$atName]=array();
1010 }
1011 $keepTags[$key]['fixAttrib'][$atName] = array_merge($keepTags[$key]['fixAttrib'][$atName],$atConfig); // Candidate for t3lib_div::array_merge() if integer-keys will some day make trouble...
1012 if (strcmp($keepTags[$key]['fixAttrib'][$atName]['range'],'')) $keepTags[$key]['fixAttrib'][$atName]['range'] = t3lib_div::trimExplode(',',$keepTags[$key]['fixAttrib'][$atName]['range']);
1013 if (strcmp($keepTags[$key]['fixAttrib'][$atName]['list'],'')) $keepTags[$key]['fixAttrib'][$atName]['list'] = t3lib_div::trimExplode(',',$keepTags[$key]['fixAttrib'][$atName]['list']);
1014 }
1015 }
1016 }
1017 unset($tagC['fixAttrib.']);
1018 unset($tagC['fixAttrib']);
1019 $keepTags[$key] = array_merge($keepTags[$key],$tagC); // Candidate for t3lib_div::array_merge() if integer-keys will some day make trouble...
1020 }
1021 }
1022 }
1023 // localNesting
1024 if ($TSconfig['localNesting']) {
1025 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['localNesting']),1);
1026 while(list(,$tn)=each($lN)) {
1027 if (isset($keepTags[$tn])) {
1028 $keepTags[$tn]['nesting']=1;
1029 }
1030 }
1031 }
1032 if ($TSconfig['globalNesting']) {
1033 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['globalNesting']),1);
1034 while(list(,$tn)=each($lN)) {
1035 if (isset($keepTags[$tn])) {
1036 if (!is_array($keepTags[$tn])) $keepTags[$tn]=array();
1037 $keepTags[$tn]['nesting']='global';
1038 }
1039 }
1040 }
1041 if ($TSconfig['rmTagIfNoAttrib']) {
1042 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['rmTagIfNoAttrib']),1);
1043 while(list(,$tn)=each($lN)) {
1044 if (isset($keepTags[$tn])) {
1045 if (!is_array($keepTags[$tn])) $keepTags[$tn]=array();
1046 $keepTags[$tn]['rmTagIfNoAttrib']=1;
1047 }
1048 }
1049 }
1050 if ($TSconfig['noAttrib']) {
1051 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['noAttrib']),1);
1052 while(list(,$tn)=each($lN)) {
1053 if (isset($keepTags[$tn])) {
1054 if (!is_array($keepTags[$tn])) $keepTags[$tn]=array();
1055 $keepTags[$tn]['allowedAttribs']=0;
1056 }
1057 }
1058 }
1059 if ($TSconfig['removeTags']) {
1060 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['removeTags']),1);
1061 while(list(,$tn)=each($lN)) {
1062 $keepTags[$tn]=array();
1063 $keepTags[$tn]['allowedAttribs']=0;
1064 $keepTags[$tn]['rmTagIfNoAttrib']=1;
1065 }
1066 }
1067
1068 // Create additional configuration:
1069 $addConfig=array();
1070 if ($TSconfig['xhtml_cleaning']) {
1071 $addConfig['xhtml']=1;
1072 }
1073
1074 return array(
1075 $keepTags,
1076 ''.$TSconfig['keepNonMatchedTags'],
1077 intval($TSconfig['htmlSpecialChars']),
1078 $addConfig
1079 );
1080 }
1081
1082 /**
1083 * Tries to convert the content to be XHTML compliant and other stuff like that.
1084 * STILL EXPERIMENTAL. See comments below.
1085 *
1086 * What it does NOT do (yet) according to XHTML specs.:
1087 * - Wellformedness: Nesting is NOT checked
1088 * - name/id attribute issue is not observed at this point.
1089 * - Certain nesting of elements not allowed. Most interesting, <PRE> cannot contain img, big,small,sub,sup ...
1090 * - Wrapping scripts and style element contents in CDATA - or alternatively they should have entitites converted.
1091 * - Setting charsets may put some special requirements on both XML declaration/ meta-http-equiv. (C.9)
1092 * - UTF-8 encoding is in fact expected by XML!!
1093 * - stylesheet element and attribute names are NOT converted to lowercase
1094 * - ampersands (and entities in general I think) MUST be converted to an entity reference! (&amps;). This may mean further conversion of non-tag content before output to page. May be related to the charset issue as a whole.
1095 * - Minimized values not allowed: Must do this: selected="selected"
1096 *
1097 * What it does at this point:
1098 * - All tags (frame,base,meta,link + img,br,hr,area,input) is ended with "/>" - others?
1099 * - Lowercase for elements and attributes
1100 * - All attributes in quotes
1101 * - Add "alt" attribute to img-tags if it's not there already.
1102 *
1103 * @param string Content to clean up
1104 * @return string Cleaned up content returned.
1105 * @access private
1106 */
1107 function XHTML_clean($content) {
1108 $content = $this->HTMLcleaner(
1109 $content,
1110 array(), // No tags treated specially
1111 1, // Keep ALL tags.
1112 0, // All content is htmlspecialchar()'ed (or ??) - if we do, <script> content will break...
1113 array('xhtml' => 1)
1114 );
1115
1116 return $content;
1117 }
1118
1119 /**
1120 * Processing all tags themselves
1121 * (Some additions by Sacha Vorbeck)
1122 *
1123 * @param string Tag to process
1124 * @param array Configuration array passing instructions for processing. If count()==0, function will return value unprocessed. See source code for details
1125 * @param boolean Is endtag, then set this.
1126 * @param boolean If set, just return value straight away
1127 * @return string Processed value.
1128 * @access private
1129 */
1130 function processTag($value,$conf,$endTag,$protected=0) {
1131 // Return immediately if protected or no parameters
1132 if ($protected || !count($conf)) return $value;
1133
1134 // OK then, begin processing for XHTML output:
1135 // STILL VERY EXPERIMENTAL!!
1136 if ($conf['xhtml']) {
1137 if ($endTag) { // Endtags are just set lowercase right away
1138 $value = strtolower($value);
1139 } elseif (substr($value,0,2)!='<!') { // ... and comments are ignored.
1140 $inValue = substr($value,1,(substr($value,-2)=='/>'?-2:-1)); // Finding inner value with out < >
1141 list($tagName,$tagP)=split('[[:space:]]',$inValue,2); // Separate attributes and tagname
1142 $tagName = strtolower($tagName);
1143
1144 // Process attributes
1145 $tagAttrib = $this->get_tag_attributes($tagP);
1146 if (!strcmp($tagName,'img') && !isset($tagAttrib[0]['alt'])) $tagAttrib[0]['alt']=''; // Set alt attribute for all images (not XHTML though...)
1147 if (!strcmp($tagName,'script') && !isset($tagAttrib[0]['type'])) $tagAttrib[0]['type']='text/javascript'; // Set type attribute for all script-tags
1148 $outA=array();
1149 reset($tagAttrib[0]);
1150 while(list($attrib_name,$attrib_value)=each($tagAttrib[0])) {
1151 // Set attributes: lowercase, always in quotes, with htmlspecialchars converted.
1152 $outA[]=$attrib_name.'="'.htmlspecialchars($this->bidir_htmlspecialchars($attrib_value,-1)).'"';
1153 }
1154 $newTag='<'.trim($tagName.' '.implode(' ',$outA));
1155 // All tags that are standalone (not wrapping, not having endtags) should be ended with '/>'
1156 if (t3lib_div::inList('img,br,hr,meta,link,base,area,input',$tagName) || substr($value,-2)=='/>') {
1157 $newTag.=' />';
1158 } else {
1159 $newTag.='>';
1160 }
1161 $value = $newTag;
1162 }
1163 }
1164
1165 return $value;
1166 }
1167
1168 /**
1169 * Processing content between tags for HTML_cleaner
1170 *
1171 * @param string The value
1172 * @param integer Direction, either -1 or +1. 0 (zero) means no change to input value.
1173 * @param mixed Not used, ignore.
1174 * @return string The processed value.
1175 * @access private
1176 */
1177 function processContent($value,$dir,$conf) {
1178 if ($dir!=0) $value = $this->bidir_htmlspecialchars($value,$dir);
1179 return $value;
1180 }
1181 }
1182
1183
1184
1185 if (defined('TYPO3_MODE') && $TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml.php']) {
1186 include_once($TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml.php']);
1187 }
1188 ?>