Various changes
[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 $str
350 *
351 * @param string HTML tag (The element name MUST be separated from the attributes by a space character! Just *whitespace* will not do)
352 * @return string Tag name in upper case
353 * @see getFirstTag()
354 */
355 function getFirstTagName($str) {
356 list($tag) = split('[[:space:]]',substr(strtoupper(trim($this->getFirstTag($str))),1,-1), 2);
357 return trim($tag);
358 }
359
360 /**
361 * Checks whether block/solo tags are found in the correct amounts in HTML content
362 * Block tags are tags which are required to have an equal amount of start and end tags, eg. "<table>...</table>"
363 * Solo tags are tags which are required to have ONLY start tags (possibly with an XHTML ending like ".../>")
364 * 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.
365 * 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!
366 * 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!
367 *
368 * @param string HTML content to analyze
369 * @param string Tag names for block tags (eg. table or div or p) in lowercase, commalist (eg. "table,div,p")
370 * @param string Tag names for solo tags (eg. img, br or input) in lowercase, commalist ("img,br,input")
371 * @return array Analyse data.
372 */
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') {
374 $content = strtolower($content);
375 $analyzedOutput=array();
376 $analyzedOutput['counts']=array(); // Counts appearances of start-tags
377 $analyzedOutput['errors']=array(); // Lists ERRORS
378 $analyzedOutput['warnings']=array(); // Lists warnings.
379 $analyzedOutput['blocks']=array(); // Lists stats for block-tags
380 $analyzedOutput['solo']=array(); // Lists stats for solo-tags
381
382 // Block tags, must have endings...
383 $blockTags = explode(',',$blockTags);
384 foreach($blockTags as $tagName) {
385 $countBegin = count(split('<'.$tagName.'[^[:alnum:]]',$content))-1;
386 $countEnd = count(split('<\/'.$tagName.'[^[:alnum:]]',$content))-1;
387 $analyzedOutput['blocks'][$tagName]=array($countBegin,$countEnd,$countBegin-$countEnd);
388 if ($countBegin) $analyzedOutput['counts'][$tagName]=$countBegin;
389 if ($countBegin-$countEnd) {
390 if ($countBegin-$countEnd > 0) {
391 $analyzedOutput['errors'][$tagName]='There were more start-tags ('.$countBegin.') than end-tags ('.$countEnd.') for the element "'.$tagName.'". There should be an equal amount!';
392 } else {
393 $analyzedOutput['warnings'][$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.';
394 }
395 }
396 }
397
398 // Solo tags, must NOT have endings...
399 $soloTags = explode(',',$soloTags);
400 foreach($soloTags as $tagName) {
401 $countBegin = count(split('<'.$tagName.'[^[:alnum:]]',$content))-1;
402 $countEnd = count(split('<\/'.$tagName.'[^[:alnum:]]',$content))-1;
403 $analyzedOutput['solo'][$tagName]=array($countBegin,$countEnd);
404 if ($countBegin) $analyzedOutput['counts'][$tagName]=$countBegin;
405 if ($countEnd) {
406 $analyzedOutput['warnings'][$tagName]='There were end-tags found ('.$countEnd.') for the element "'.$tagName.'". This was not expected (although XHTML technically allows it).';
407 }
408 }
409
410 return $analyzedOutput;
411 }
412
413 /**
414 * Function that can clean up HTML content according to configuration given in the $tags array.
415 *
416 * 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'))
417 * If the value of the $tags[$tagname] entry is an array, advanced processing of the tags is initialized. These are the options:
418 *
419 * $tags[$tagname] = Array(
420 * 'overrideAttribs' => '' If set, this string is preset as the attributes of the tag
421 * 'allowedAttribs' => '0' (zero) = no attributes allowed, '[commalist of attributes]' = only allowed attributes. If blank, all attributes are allowed.
422 * 'fixAttrib' => Array(
423 * '[attribute name]' => Array (
424 * 'default' => If no attribute exists by this name, this value is set as default value (if this value is not blank)
425 * 'always' => Boolean. If set, the attribute is always processed. Normally an attribute is processed only if it exists
426 * 'trim,intval,lower,upper' => All booleans. If any of these keys are set, the value is passed through the respective PHP-functions.
427 * 'range' => Array ('[low limit]','[high limit, optional]') Setting integer range.
428 * 'list' => Array ('[value1/default]','[value2]','[value3]') Attribute must be in this list. If not, the value is set to the first element.
429 * '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)
430 * 'removeIfEquals' => [value] If the attribute value matches the value set here, then it is removed.
431 * 'casesensitiveComp' => 1 If set, then the removeIfEquals and list comparisons will be case sensitive. Otherwise not.
432 * )
433 * ),
434 * 'protect' => '', Boolean. If set, the tag <> is converted to &lt; and &gt;
435 * 'remap' => '', String. If set, the tagname is remapped to this tagname
436 * 'rmTagIfNoAttrib' => '', Boolean. If set, then the tag is removed if no attributes happend to be there.
437 * '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>'
438 * )
439 *
440 * @param string $content; is the HTML-content being processed. This is also the result being returned.
441 * @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.
442 * @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;
443 * @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;")
444 * @param array Configuration array send along as $conf to the internal functions ->processContent() and ->processTag()
445 * @return string Processed HTML content
446 */
447 function HTMLcleaner($content, $tags=array(),$keepAll=0,$hSC=0,$addConfig=array()) {
448 $newContent = array();
449 $tokArr = explode('<',$content);
450 $newContent[]=$this->processContent(current($tokArr),$hSC,$addConfig);
451 next($tokArr);
452
453 $c=1;
454 $tagRegister=array();
455 $tagStack=array();
456 while(list(,$tok)=each($tokArr)) {
457 $firstChar = substr($tok,0,1);
458 if (strcmp(trim($firstChar),'')) { // It is a tag...
459 $tagEnd = strcspn($tok,'>');
460 if (strlen($tok)!=$tagEnd) { // If there is and end-bracket...
461 $endTag = $firstChar=='/' ? 1 : 0;
462 $tagContent = substr($tok,$endTag,$tagEnd-$endTag);
463 $tagParts = split('[[:space:]]',$tagContent,2);
464 $tagName = strtolower($tagParts[0]);
465 if (isset($tags[$tagName])) {
466 if (is_array($tags[$tagName])) { // If there is processing to do for the tag:
467 // Override attributes
468 if (strcmp($tags[$tagName]['overrideAttribs'],'')) {
469 $tagParts[1]=$tags[$tagName]['overrideAttribs'];
470 }
471
472 // Allowed tags
473 if (strcmp($tags[$tagName]['allowedAttribs'],'')) {
474 if (!strcmp($tags[$tagName]['allowedAttribs'],'0')) { // No attribs allowed
475 $tagParts[1]='';
476 } elseif (trim($tagParts[1])) {
477 $tagAttrib = $this->get_tag_attributes($tagParts[1]);
478 $tagParts[1]='';
479 $newTagAttrib = array();
480 $tList = t3lib_div::trimExplode(',',strtolower($tags[$tagName]['allowedAttribs']),1);
481 while(list(,$allowTag)=each($tList)) {
482 if (isset($tagAttrib[0][$allowTag])) $newTagAttrib[$allowTag]=$tagAttrib[0][$allowTag];
483 }
484 $tagParts[1]=$this->compileTagAttribs($newTagAttrib,$tagAttrib[1]);
485 }
486 }
487
488 // Fixed attrib values
489 if (is_array($tags[$tagName]['fixAttrib'])) {
490 $tagAttrib = $this->get_tag_attributes($tagParts[1]);
491 $tagParts[1]='';
492 reset($tags[$tagName]['fixAttrib']);
493 while(list($attr,$params)=each($tags[$tagName]['fixAttrib'])) {
494 if (strcmp($params['default'],'') && !isset($tagAttrib[0][$attr])) $tagAttrib[0][$attr]=$params['default'];
495 if ($params['always'] || isset($tagAttrib[0][$attr])) {
496 if ($params['trim']) {$tagAttrib[0][$attr]=trim($tagAttrib[0][$attr]);}
497 if ($params['intval']) {$tagAttrib[0][$attr]=intval($tagAttrib[0][$attr]);}
498 if ($params['lower']) {$tagAttrib[0][$attr]=strtolower($tagAttrib[0][$attr]);}
499 if ($params['upper']) {$tagAttrib[0][$attr]=strtoupper($tagAttrib[0][$attr]);}
500 if ($params['range']) {
501 if (isset($params['range'][1])) {
502 $tagAttrib[0][$attr]=t3lib_div::intInRange($tagAttrib[0][$attr],intval($params['range'][0]),intval($params['range'][1]));
503 } else {
504 $tagAttrib[0][$attr]=t3lib_div::intInRange($tagAttrib[0][$attr],intval($params['range'][0]));
505 }
506 }
507 if (is_array($params['list'])) {
508 if (!in_array($this->caseShift($tagAttrib[0][$attr],$params['casesensitiveComp']),$this->caseShift($params['list'],$params['casesensitiveComp'],$tagName))) $tagAttrib[0][$attr]=$params['list'][0];
509 }
510 if (($params['removeIfFalse'] && $params['removeIfFalse']!='blank' && !$tagAttrib[0][$attr]) || ($params['removeIfFalse']=='blank' && !strcmp($tagAttrib[0][$attr],''))) {
511 unset($tagAttrib[0][$attr]);
512 }
513 if (strcmp($params['removeIfEquals'],'') && !strcmp($this->caseShift($tagAttrib[0][$attr],$params['casesensitiveComp']),$this->caseShift($params['removeIfEquals'],$params['casesensitiveComp']))) {
514 unset($tagAttrib[0][$attr]);
515 }
516 }
517 }
518 $tagParts[1]=$this->compileTagAttribs($tagAttrib[0],$tagAttrib[1]);
519 }
520
521 // Protecting the tag by converting < and > to &lt; and &gt; ??
522 if ($tags[$tagName]['protect']) {
523 $lt = '&lt;'; $gt = '&gt;';
524 } else {
525 $lt = '<'; $gt = '>';
526 }
527 // Remapping tag name?
528 if ($tags[$tagName]['remap']) $tagParts[0] = $tags[$tagName]['remap'];
529
530 // rmTagIfNoAttrib
531 if ($endTag || trim($tagParts[1]) || !$tags[$tagName]['rmTagIfNoAttrib']) {
532 $setTag=1;
533
534 if ($tags[$tagName]['nesting']) {
535 if (!is_array($tagRegister[$tagName])) $tagRegister[$tagName]=array();
536
537 if ($endTag) {
538 /* if ($tags[$tagName]['nesting']=='global') {
539 $lastEl = end($tagStack);
540 $correctTag = !strcmp($tagName,$lastEl);
541 } else $correctTag=1;
542 */
543 $correctTag=1;
544 if ($tags[$tagName]['nesting']=='global') {
545 $lastEl = end($tagStack);
546 if (strcmp($tagName,$lastEl)) {
547 if (in_array($tagName,$tagStack)) {
548 while(count($tagStack) && strcmp($tagName,$lastEl)) {
549 $elPos = end($tagRegister[$lastEl]);
550 unset($newContent[$elPos]);
551
552 array_pop($tagRegister[$lastEl]);
553 array_pop($tagStack);
554 $lastEl = end($tagStack);
555 }
556 } else {
557 $correctTag=0; // In this case the
558 }
559 }
560 }
561 if (!count($tagRegister[$tagName]) || !$correctTag) {
562 $setTag=0;
563 } else {
564 array_pop($tagRegister[$tagName]);
565 if ($tags[$tagName]['nesting']=='global') {array_pop($tagStack);}
566 }
567 } else {
568 array_push($tagRegister[$tagName],$c);
569 if ($tags[$tagName]['nesting']=='global') {array_push($tagStack,$tagName);}
570 }
571 }
572
573 if ($setTag) {
574 // Setting the tag
575 $newContent[$c++]=$this->processTag($lt.($endTag?'/':'').trim($tagParts[0].' '.$tagParts[1]).$gt,$addConfig,$endTag,$lt=='&lt;');
576 }
577 }
578 } else {
579 $newContent[$c++]=$this->processTag('<'.($endTag?'/':'').$tagContent.'>',$addConfig,$endTag);
580 }
581 } elseif ($keepAll) { // This is if the tag was not defined in the array for processing:
582 if (!strcmp($keepAll,'protect')) {
583 $lt = '&lt;'; $gt = '&gt;';
584 } else {
585 $lt = '<'; $gt = '>';
586 }
587 $newContent[$c++]=$this->processTag($lt.($endTag?'/':'').$tagContent.$gt,$addConfig,$endTag,$lt=='&lt;');
588 }
589 $newContent[$c++]=$this->processContent(substr($tok,$tagEnd+1),$hSC,$addConfig);
590 } else {
591 $newContent[$c++]=$this->processContent('<'.$tok,$hSC,$addConfig); // There were not end-bracket, so no tag...
592 }
593 } else {
594 $newContent[$c++]=$this->processContent('<'.$tok,$hSC,$addConfig); // It was not a tag anyways
595 }
596 }
597
598 // Unsetting tags:
599 reset($tagRegister);
600 while(list($tag,$positions)=each($tagRegister)) {
601 reset($positions);
602 while(list(,$pKey)=each($positions)) {
603 unset($newContent[$pKey]);
604 }
605 }
606
607 return implode('',$newContent);
608 }
609
610 /**
611 * Returns an array with all attributes as keys. Attributes are only lowercase a-z
612 * If a attribute is empty (shorthand), then the value for the key is empty. You can check if it existed with isset()
613 *
614 * @param string Tag: $tag is either a whole tag (eg '<TAG OPTION ATTRIB=VALUE>') or the parameterlist (ex ' OPTION ATTRIB=VALUE>')
615 * @param boolean If set, the attribute values are de-htmlspecialchar'ed. Should actually always be set!
616 * @return array array(Tag attributes,Attribute meta-data)
617 */
618 function get_tag_attributes($tag,$deHSC=0) {
619 list($components,$metaC) = $this->split_tag_attributes($tag);
620 $name = ''; // attribute name is stored here
621 $valuemode = '';
622 $attributes=array();
623 $attributesMeta=array();
624 if (is_array($components)) {
625 while (list($key,$val) = each ($components)) {
626 if ($val != '=') { // Only if $name is set (if there is an attribute, that waits for a value), that valuemode is enabled. This ensures that the attribute is assigned it's value
627 if ($valuemode) {
628 if ($name) {
629 $attributes[$name] = $deHSC?t3lib_div::htmlspecialchars_decode($val):$val;
630 $attributesMeta[$name]['dashType']=$metaC[$key];
631 $name = '';
632 }
633 } else {
634 if ($namekey = ereg_replace('[^a-zA-Z0-9_-]','',$val)) {
635 $name = strtolower($namekey);
636 $attributesMeta[$name]=array();
637 $attributesMeta[$name]['origTag']=$namekey;
638 $attributes[$name] = '';
639 }
640 }
641 $valuemode = '';
642 } else {
643 $valuemode = 'on';
644 }
645 }
646 if (is_array($attributes)) reset($attributes);
647 return array($attributes,$attributesMeta);
648 }
649 }
650
651 /**
652 * Returns an array with the 'components' from an attribute list. The result is normally analyzed by get_tag_attributes
653 * Removes tag-name if found
654 *
655 * @param string The tag or attributes
656 * @return array
657 * @access private
658 * @see t3lib_div::split_tag_attributes()
659 */
660 function split_tag_attributes($tag) {
661 $tag_tmp = trim(eregi_replace ('^<[^[:space:]]*','',trim($tag)));
662 // Removes any > in the end of the string
663 $tag_tmp = trim(eregi_replace ('>$','',$tag_tmp));
664
665 $metaValue = array();
666 $value = array();
667 while (strcmp($tag_tmp,'')) { // Compared with empty string instead , 030102
668 $firstChar=substr($tag_tmp,0,1);
669 if (!strcmp($firstChar,'"') || !strcmp($firstChar,"'")) {
670 $reg=explode($firstChar,$tag_tmp,3);
671 $value[]=$reg[1];
672 $metaValue[]=$firstChar;
673 $tag_tmp=trim($reg[2]);
674 } elseif (!strcmp($firstChar,'=')) {
675 $value[] = '=';
676 $metaValue[]='';
677 $tag_tmp = trim(substr($tag_tmp,1)); // Removes = chars.
678 } else {
679 // There are '' around the value. We look for the next ' ' or '>'
680 $reg = split('[[:space:]=]',$tag_tmp,2);
681 $value[] = trim($reg[0]);
682 $metaValue[]='';
683 $tag_tmp = trim(substr($tag_tmp,strlen($reg[0]),1).$reg[1]);
684 }
685 }
686 if (is_array($value)) reset($value);
687 return array($value,$metaValue);
688 }
689
690 /**
691 * Converts htmlspecialchars forth ($dir=1) AND back ($dir=-1)
692 *
693 * @param string Input value
694 * @param integer Direction: forth ($dir=1, dir=2 for preserving entities) AND back ($dir=-1)
695 * @return string Output value
696 */
697 function bidir_htmlspecialchars($value,$dir) {
698 if ($dir==1) {
699 $value = htmlspecialchars($value);
700 } elseif ($dir==2) {
701 $value = t3lib_div::deHSCentities(htmlspecialchars($value));
702 } elseif ($dir==-1) {
703 $value = str_replace('&gt;','>',$value);
704 $value = str_replace('&lt;','<',$value);
705 $value = str_replace('&quot;','"',$value);
706 $value = str_replace('&amp;','&',$value);
707 }
708 return $value;
709 }
710
711 /**
712 * 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 $alternatives
713 *
714 * @param string Prefix string
715 * @param string HTML content
716 * @param array Array with alternative prefixes for certain of the tags. key=>value pairs where the keys are the tag element names in uppercase
717 * @return string Processed HTML content
718 */
719 function prefixResourcePath($main_prefix,$content,$alternatives=array()) {
720 $parts = $this->splitTags('td,table,body,img,input,form,link,script,a',$content);
721 reset($parts);
722 while(list($k,$v)=each($parts)) {
723 if ($k%2) {
724 $params = $this->get_tag_attributes($v,1);
725 $tagEnd = substr($v,-2)=='/>' ? ' />' : '>'; // Detect tag-ending so that it is re-applied correctly.
726 $firstTagName = $this->getFirstTagName($v); // The 'name' of the first tag
727 $somethingDone=0;
728 $prefix = isset($alternatives[strtoupper($firstTagName)]) ? $alternatives[strtoupper($firstTagName)] : $main_prefix;
729 switch(strtolower($firstTagName)) {
730 // background - attribute:
731 case 'td':
732 case 'body':
733 case 'table':
734 $src = $params[0]['background'];
735 if ($src) {
736 $params[0]['background'] = $this->prefixRelPath($prefix,$params[0]['background']);
737 $somethingDone=1;
738 }
739 break;
740 // src attribute
741 case 'img':
742 case 'input':
743 case 'script':
744 $src = $params[0]['src'];
745 if ($src) {
746 $params[0]['src'] = $this->prefixRelPath($prefix,$params[0]['src']);
747 $somethingDone=1;
748 }
749 break;
750 case 'link':
751 case 'a':
752 $src = $params[0]['href'];
753 if ($src) {
754 $params[0]['href'] = $this->prefixRelPath($prefix,$params[0]['href']);
755 $somethingDone=1;
756 }
757 break;
758 // action attribute
759 case 'form':
760 $src = $params[0]['action'];
761 if ($src) {
762 $params[0]['action'] = $this->prefixRelPath($prefix,$params[0]['action']);
763 $somethingDone=1;
764 }
765 break;
766 }
767 if ($somethingDone) {
768 $tagParts = split('[[:space:]]',$v,2);
769 $tagParts[1]=$this->compileTagAttribs($params[0],$params[1]);
770 $parts[$k] = '<'.trim(strtolower($firstTagName).' '.$tagParts[1]).
771 $tagEnd;
772 }
773 }
774 }
775
776 return implode('',$parts);
777 }
778
779 /**
780 * Internal sub-function for ->prefixResourcePath()
781 *
782 * @param string Prefix string
783 * @param string Relative path/URL
784 * @return string Output path, prefixed if no scheme in input string
785 * @access private
786 */
787 function prefixRelPath($prefix,$srcVal) {
788 $pU = parse_url($srcVal);
789 if (!$pU['scheme']) { // If not an absolute URL.
790 $srcVal = $prefix.$srcVal;
791 }
792 return $srcVal;
793 }
794
795 /**
796 * Cleans up the input $value for fonttags.
797 * If keepFace,-Size and -Color is set then font-tags with an allowed property is kept. Else deleted.
798 *
799 * @param string HTML content with font-tags inside to clean up.
800 * @param boolean If set, keep "face" attribute
801 * @param boolean If set, keep "size" attribute
802 * @param boolean If set, keep "color" attribute
803 * @return string Processed HTML content
804 */
805 function cleanFontTags($value,$keepFace=0,$keepSize=0,$keepColor=0) {
806 $fontSplit = $this->splitIntoBlock('font',$value); // ,1 ?? - could probably be more stable if splitTags() was used since this depends on end-tags being properly set!
807 reset($fontSplit);
808 while(list($k,$v)=each($fontSplit)) {
809 if ($k%2) { // font:
810 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v));
811 $newAttribs=array();
812 if ($keepFace && $attribArray['face']) $newAttribs[]='face="'.$attribArray['face'].'"';
813 if ($keepSize && $attribArray['size']) $newAttribs[]='size="'.$attribArray['size'].'"';
814 if ($keepColor && $attribArray['color']) $newAttribs[]='color="'.$attribArray['color'].'"';
815
816 $innerContent = $this->cleanFontTags($this->removeFirstAndLastTag($v),$keepFace,$keepSize,$keepColor);
817 if (count($newAttribs)) {
818 $fontSplit[$k]='<font '.implode(' ',$newAttribs).'>'.$innerContent.'</font>';
819 } else {
820 $fontSplit[$k]=$innerContent;
821 }
822 }
823 }
824 return implode('',$fontSplit);
825 }
826
827 /**
828 * This is used to map certain tag-names into other names.
829 *
830 * @param string HTML content
831 * @param array Array with tag key=>value pairs where key is from-tag and value is to-tag
832 * @param string Alternative less-than char to search for (search regex string)
833 * @param string Alternative less-than char to replace with (replace regex string)
834 * @return string Processed HTML content
835 */
836 function mapTags($value,$tags=array(),$ltChar='<',$ltChar2='<') {
837
838 foreach($tags as $from => $to) {
839 $value = eregi_replace($ltChar.$from.'>',$ltChar2.$to.'>',$value);
840 $value = eregi_replace($ltChar.$from.'[[:space:]]([^>]*)>',$ltChar2.$to.' \\1>',$value);
841 $value = eregi_replace($ltChar.'\/'.$from.'[^>]*>',$ltChar2.'/'.$to.'>',$value);
842 }
843 return $value;
844 }
845
846 /**
847 * This converts htmlspecialchar()'ed tags (from $tagList) back to real tags. Eg. '&lt;strong&gt' would be converted back to '<strong>' if found in $tagList
848 *
849 * @param string HTML content
850 * @param string Tag list, separated by comma. Lowercase!
851 * @return string Processed HTML content
852 */
853 function unprotectTags($content,$tagList='') {
854 $tagsArray = t3lib_div::trimExplode(',',$tagList,1);
855 $contentParts = explode('&lt;',$content);
856 next($contentParts); // bypass the first
857 while(list($k,$tok)=each($contentParts)) {
858 $firstChar = substr($tok,0,1);
859 if (strcmp(trim($firstChar),'')) {
860 $subparts = explode('&gt;',$tok,2);
861 $tagEnd = strlen($subparts[0]);
862 if (strlen($tok)!=$tagEnd) {
863 $endTag = $firstChar=='/' ? 1 : 0;
864 $tagContent = substr($tok,$endTag,$tagEnd-$endTag);
865 $tagParts = split('[[:space:]]',$tagContent,2);
866 $tagName = strtolower($tagParts[0]);
867 if (!strcmp($tagList,'') || in_array($tagName,$tagsArray)) {
868 $contentParts[$k] = '<'.$subparts[0].'>'.$subparts[1];
869 } else $contentParts[$k] = '&lt;'.$tok;
870 } else $contentParts[$k] = '&lt;'.$tok;
871 } else $contentParts[$k] = '&lt;'.$tok;
872 }
873
874 return implode('',$contentParts);
875 }
876
877 /**
878 * Strips tags except the tags in the list, $tagList
879 * OBSOLETE - use PHP function strip_tags()
880 *
881 * @param string Value to process
882 * @param string List of tags
883 * @return string Output value
884 * @ignore
885 */
886 function stripTagsExcept($value,$tagList) {
887 $tags=t3lib_div::trimExplode(',',$tagList,1);
888 $forthArr=array();
889 $backArr=array();
890 while(list(,$theTag)=each($tags)) {
891 $forthArr[$theTag]=md5($theTag);
892 $backArr[md5($theTag)]=$theTag;
893 }
894 $value = $this->mapTags($value,$forthArr,'<','_');
895 $value=strip_tags($value);
896 $value = $this->mapTags($value,$backArr,'_','<');
897 return $value;
898 }
899
900 /**
901 * Internal function for case shifting of a string or whole array
902 *
903 * @param mixed Input string/array
904 * @param boolean If $str is a string AND this boolean is true, the string is returned in uppercase
905 * @param string Key string used for internal caching of the results. Could be an MD5 hash of the serialized version of the input $str if that is an array.
906 * @return string Output string, processed
907 * @access private
908 */
909 function caseShift($str,$flag,$cacheKey='') {
910 if (is_array($str)) {
911 if (!$cacheKey || !isset($this->caseShift_cache[$cacheKey])) {
912 reset($str);
913 while(list($k)=each($str)) {
914 $str[$k] = strtoupper($str[$k]);
915 }
916 if ($cacheKey) $this->caseShift_cache[$cacheKey]=$str;
917 } else {
918 $str = $this->caseShift_cache[$cacheKey];
919 }
920 } elseif (!$flag) $str = strtoupper($str);
921 return $str;
922 }
923
924 /**
925 * Compiling an array with tag attributes into a string
926 *
927 * @param array Tag attributes
928 * @param array Meta information about these attributes (like if they were quoted)
929 * @param boolean If set, then the attribute names will be set in lower case, value quotes in double-quotes and the value will be htmlspecialchar()'ed
930 * @return string Imploded attributes, eg: 'attribute="value" attrib2="value2"'
931 * @access private
932 */
933 function compileTagAttribs($tagAttrib,$meta=array(), $xhtmlClean=0) {
934 $accu=array();
935 reset($tagAttrib);
936 while(list($k,$v)=each($tagAttrib)) {
937 if ($xhtmlClean) {
938 $attr=strtolower($k);
939 if (strcmp($v,'') || isset($meta[$k]['dashType'])) {
940 $attr.='="'.htmlspecialchars($v).'"';
941 }
942 } else {
943 $attr=$meta[$k]['origTag']?$meta[$k]['origTag']:$k;
944 if (strcmp($v,'') || isset($meta[$k]['dashType'])) {
945 $dash=$meta[$k]['dashType']?$meta[$k]['dashType']:(t3lib_div::testInt($v)?'':'"');
946 $attr.='='.$dash.$v.$dash;
947 }
948 }
949 $accu[]=$attr;
950 }
951 return implode(' ',$accu);
952 }
953
954 /**
955 * Get tag attributes, the classic version (which had some limitations?)
956 *
957 * @param string The tag
958 * @param boolean De-htmlspecialchar flag.
959 * @return array
960 * @access private
961 */
962 function get_tag_attributes_classic($tag,$deHSC=0) {
963 $attr=$this->get_tag_attributes($tag,$deHSC);
964 return is_array($attr[0])?$attr[0]:array();
965 }
966
967 /**
968 * Converts TSconfig into an array for the HTMLcleaner function.
969 *
970 * @param array TSconfig for HTMLcleaner
971 * @param array Array of tags to keep (?)
972 * @return array
973 * @access private
974 */
975 function HTMLparserConfig($TSconfig,$keepTags=array()) {
976 // Allow tags (base list, merged with incoming array)
977 $alTags = array_flip(t3lib_div::trimExplode(',',strtolower($TSconfig['allowTags']),1));
978 $keepTags = array_merge($alTags,$keepTags);
979
980 // Set config properties.
981 if (is_array($TSconfig['tags.'])) {
982 reset($TSconfig['tags.']);
983 while(list($key,$tagC)=each($TSconfig['tags.'])) {
984 if (!is_array($tagC) && $key==strtolower($key)) {
985 if (!strcmp($tagC,'0')) unset($keepTags[$key]);
986 if (!strcmp($tagC,'1') && !isset($keepTags[$key])) $keepTags[$key]=1;
987 }
988 }
989
990 reset($TSconfig['tags.']);
991 while(list($key,$tagC)=each($TSconfig['tags.'])) {
992 if (is_array($tagC) && $key==strtolower($key)) {
993 $key=substr($key,0,-1);
994 if (!is_array($keepTags[$key])) $keepTags[$key]=array();
995 if (is_array($tagC['fixAttrib.'])) {
996 reset($tagC['fixAttrib.']);
997 while(list($atName,$atConfig)=each($tagC['fixAttrib.'])) {
998 if (is_array($atConfig)) {
999 $atName=substr($atName,0,-1);
1000 if (!is_array($keepTags[$key]['fixAttrib'][$atName])) {
1001 $keepTags[$key]['fixAttrib'][$atName]=array();
1002 }
1003 $keepTags[$key]['fixAttrib'][$atName] = array_merge($keepTags[$key]['fixAttrib'][$atName],$atConfig); // Candidate for t3lib_div::array_merge() if integer-keys will some day make trouble...
1004 if (strcmp($keepTags[$key]['fixAttrib'][$atName]['range'],'')) $keepTags[$key]['fixAttrib'][$atName]['range'] = t3lib_div::trimExplode(',',$keepTags[$key]['fixAttrib'][$atName]['range']);
1005 if (strcmp($keepTags[$key]['fixAttrib'][$atName]['list'],'')) $keepTags[$key]['fixAttrib'][$atName]['list'] = t3lib_div::trimExplode(',',$keepTags[$key]['fixAttrib'][$atName]['list']);
1006 }
1007 }
1008 }
1009 unset($tagC['fixAttrib.']);
1010 unset($tagC['fixAttrib']);
1011 $keepTags[$key] = array_merge($keepTags[$key],$tagC); // Candidate for t3lib_div::array_merge() if integer-keys will some day make trouble...
1012 }
1013 }
1014 }
1015 // localNesting
1016 if ($TSconfig['localNesting']) {
1017 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['localNesting']),1);
1018 while(list(,$tn)=each($lN)) {
1019 if (isset($keepTags[$tn])) {
1020 $keepTags[$tn]['nesting']=1;
1021 }
1022 }
1023 }
1024 if ($TSconfig['globalNesting']) {
1025 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['globalNesting']),1);
1026 while(list(,$tn)=each($lN)) {
1027 if (isset($keepTags[$tn])) {
1028 if (!is_array($keepTags[$tn])) $keepTags[$tn]=array();
1029 $keepTags[$tn]['nesting']='global';
1030 }
1031 }
1032 }
1033 if ($TSconfig['rmTagIfNoAttrib']) {
1034 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['rmTagIfNoAttrib']),1);
1035 while(list(,$tn)=each($lN)) {
1036 if (isset($keepTags[$tn])) {
1037 if (!is_array($keepTags[$tn])) $keepTags[$tn]=array();
1038 $keepTags[$tn]['rmTagIfNoAttrib']=1;
1039 }
1040 }
1041 }
1042 if ($TSconfig['noAttrib']) {
1043 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['noAttrib']),1);
1044 while(list(,$tn)=each($lN)) {
1045 if (isset($keepTags[$tn])) {
1046 if (!is_array($keepTags[$tn])) $keepTags[$tn]=array();
1047 $keepTags[$tn]['allowedAttribs']=0;
1048 }
1049 }
1050 }
1051 if ($TSconfig['removeTags']) {
1052 $lN = t3lib_div::trimExplode(',',strtolower($TSconfig['removeTags']),1);
1053 while(list(,$tn)=each($lN)) {
1054 $keepTags[$tn]=array();
1055 $keepTags[$tn]['allowedAttribs']=0;
1056 $keepTags[$tn]['rmTagIfNoAttrib']=1;
1057 }
1058 }
1059
1060 // Create additional configuration:
1061 $addConfig=array();
1062 if ($TSconfig['xhtml_cleaning']) {
1063 $addConfig['xhtml']=1;
1064 }
1065
1066 return array(
1067 $keepTags,
1068 ''.$TSconfig['keepNonMatchedTags'],
1069 intval($TSconfig['htmlSpecialChars']),
1070 $addConfig
1071 );
1072 }
1073
1074 /**
1075 * Tries to convert the content to be XHTML compliant and other stuff like that.
1076 * STILL EXPERIMENTAL. See comments below.
1077 *
1078 * What it does NOT do (yet) according to XHTML specs.:
1079 * - Wellformedness: Nesting is NOT checked
1080 * - name/id attribute issue is not observed at this point.
1081 * - Certain nesting of elements not allowed. Most interesting, <PRE> cannot contain img, big,small,sub,sup ...
1082 * - Wrapping scripts and style element contents in CDATA - or alternatively they should have entitites converted.
1083 * - Setting charsets may put some special requirements on both XML declaration/ meta-http-equiv. (C.9)
1084 * - UTF-8 encoding is in fact expected by XML!!
1085 * - stylesheet element and attribute names are NOT converted to lowercase
1086 * - ampersands (and entities in general I think) MUST be converted to an entity reference! (&amps;). This may mean further conversion of non-tag content before output to page. May be related to the charset issue as a whole.
1087 * - Minimized values not allowed: Must do this: selected="selected"
1088 *
1089 * What it does at this point:
1090 * - All tags (frame,base,meta,link + img,br,hr,area,input) is ended with "/>" - others?
1091 * - Lowercase for elements and attributes
1092 * - All attributes in quotes
1093 * - Add "alt" attribute to img-tags if it's not there already.
1094 *
1095 * @param string Content to clean up
1096 * @return string Cleaned up content returned.
1097 * @access private
1098 */
1099 function XHTML_clean($content) {
1100 $content = $this->HTMLcleaner(
1101 $content,
1102 array(), // No tags treated specially
1103 1, // Keep ALL tags.
1104 0, // All content is htmlspecialchar()'ed (or ??) - if we do, <script> content will break...
1105 array('xhtml' => 1)
1106 );
1107
1108 return $content;
1109 }
1110
1111 /**
1112 * Processing all tags themselves
1113 * (Some additions by Sacha Vorbeck)
1114 *
1115 * @param string Tag to process
1116 * @param array Configuration array passing instructions for processing. If count()==0, function will return value unprocessed. See source code for details
1117 * @param boolean Is endtag, then set this.
1118 * @param boolean If set, just return value straight away
1119 * @return string Processed value.
1120 * @access private
1121 */
1122 function processTag($value,$conf,$endTag,$protected=0) {
1123 // Return immediately if protected or no parameters
1124 if ($protected || !count($conf)) return $value;
1125
1126 // OK then, begin processing for XHTML output:
1127 // STILL VERY EXPERIMENTAL!!
1128 if ($conf['xhtml']) {
1129 if ($endTag) { // Endtags are just set lowercase right away
1130 $value = strtolower($value);
1131 } elseif (substr($value,0,2)!='<!') { // ... and comments are ignored.
1132 $inValue = substr($value,1,(substr($value,-2)=='/>'?-2:-1)); // Finding inner value with out < >
1133 list($tagName,$tagP)=split('[[:space:]]',$inValue,2); // Separate attributes and tagname
1134 $tagName = strtolower($tagName);
1135
1136 // Process attributes
1137 $tagAttrib = $this->get_tag_attributes($tagP);
1138 if (!strcmp($tagName,'img') && !isset($tagAttrib[0]['alt'])) $tagAttrib[0]['alt']=''; // Set alt attribute for all images (not XHTML though...)
1139 if (!strcmp($tagName,'script') && !isset($tagAttrib[0]['type'])) $tagAttrib[0]['type']='text/javascript'; // Set type attribute for all script-tags
1140 $outA=array();
1141 reset($tagAttrib[0]);
1142 while(list($attrib_name,$attrib_value)=each($tagAttrib[0])) {
1143 // Set attributes: lowercase, always in quotes, with htmlspecialchars converted.
1144 $outA[]=$attrib_name.'="'.htmlspecialchars($this->bidir_htmlspecialchars($attrib_value,-1)).'"';
1145 }
1146 $newTag='<'.trim($tagName.' '.implode(' ',$outA));
1147 // All tags that are standalone (not wrapping, not having endtags) should be ended with '/>'
1148 if (t3lib_div::inList('img,br,hr,meta,link,base,area,input',$tagName) || substr($value,-2)=='/>') {
1149 $newTag.=' />';
1150 } else {
1151 $newTag.='>';
1152 }
1153 $value = $newTag;
1154 }
1155 }
1156 return $value;
1157
1158 }
1159
1160 /**
1161 * Processing content between tags for HTML_cleaner
1162 *
1163 * @param string The value
1164 * @param integer Direction, either -1 or +1. 0 (zero) means no change to input value.
1165 * @param mixed Not used, ignore.
1166 * @return string The processed value.
1167 * @access private
1168 */
1169 function processContent($value,$dir,$conf) {
1170 if ($dir!=0) $value = $this->bidir_htmlspecialchars($value,$dir);
1171 return $value;
1172 }
1173 }
1174
1175
1176
1177 if (defined('TYPO3_MODE') && $TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml.php']) {
1178 include_once($TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml.php']);
1179 }
1180 ?>