Changed lots of stuff...
[Packages/TYPO3.CMS.git] / t3lib / class.t3lib_parsehtml_proc.php
1 <?php
2 /***************************************************************
3 * Copyright notice
4 *
5 * (c) 1999-2004 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 * Functions for parsing HTML, specially for TYPO3 processing in relation to TCEmain and Rich Text Editor (RTE)
29 *
30 * $Id$
31 * Revised for TYPO3 3.6 December/2003 by Kasper Skaarhoj
32 * XHTML compatible.
33 *
34 * @author Kasper Skaarhoj <kasper@typo3.com>
35 * @internal
36 */
37 /**
38 * [CLASS/FUNCTION INDEX of SCRIPT]
39 *
40 *
41 *
42 * 102: class t3lib_parsehtml_proc extends t3lib_parsehtml
43 * 137: function init($elRef='',$recPid=0)
44 * 149: function setRelPath($path)
45 * 173: function evalWriteFile($pArr,$currentRecord)
46 *
47 * SECTION: Main function
48 * 231: function RTE_transform($value,$specConf,$direction='rte',$thisConfig=array())
49 *
50 * SECTION: Specific RTE TRANSFORMATION functions
51 * 378: function TS_images_db($value)
52 * 479: function TS_images_rte($value)
53 * 513: function TS_reglinks($value,$direction)
54 * 547: function TS_links_db($value)
55 * 591: function TS_links_rte($value)
56 * 666: function TS_preserve_db($value)
57 * 690: function TS_preserve_rte($value)
58 * 711: function TS_transform_db($value,$css=FALSE)
59 * 822: function TS_transform_rte($value,$css=0)
60 * 893: function TS_strip_db($value)
61 *
62 * SECTION: Generic RTE transformation, analysis and helper functions
63 * 924: function getURL($url)
64 * 938: function HTMLcleaner_db($content,$tagList='')
65 * 959: function getKeepTags($direction='rte',$tagList='')
66 * 1068: function divideIntoLines($value,$count=5,$returnArray=FALSE)
67 * 1172: function setDivTags($value,$dT='p')
68 * 1217: function internalizeFontTags($value)
69 * 1253: function siteUrl()
70 * 1263: function rteImageStorageDir()
71 * 1275: function removeTables($value,$breakChar='<br />')
72 * 1307: function defaultTStagMapping($code,$direction='rte')
73 * 1330: function getWHFromAttribs($attribArray)
74 * 1356: function urlInfoForLinkTags($url)
75 * 1415: function TS_AtagToAbs($value,$dontSetRTEKEEP=FALSE)
76 *
77 * TOTAL FUNCTIONS: 27
78 * (This index is automatically created/updated by the extension "extdeveval")
79 *
80 */
81
82 require_once (PATH_t3lib.'class.t3lib_parsehtml.php');
83
84
85
86
87
88
89
90
91
92
93
94
95 /**
96 * Class for parsing HTML for the Rich Text Editor. (also called transformations)
97 *
98 * @author Kasper Skaarhoj <kasper@typo3.com>
99 * @package TYPO3
100 * @subpackage t3lib
101 */
102 class t3lib_parsehtml_proc extends t3lib_parsehtml {
103
104 // Static:
105 var $headListTags = 'PRE,UL,OL,H1,H2,H3,H4,H5,H6'; // List of tags for header, pre and list containers
106
107 // Internal, static:
108 var $recPid = 0; // Set this to the pid of the record manipulated by the class.
109 var $elRef = ''; // Element reference [table]:[field], eg. "tt_content:bodytext"
110 var $relPath=''; // Relative path
111 var $relBackPath=''; // Relative back-path
112 var $procOptions = ''; // Set to the TSconfig options coming from Page TSconfig
113
114 // Internal, dynamic
115 var $TS_transform_db_safecounter=100; // Run-away brake for recursive calls.
116 var $rte_p=''; // Parameters from TCA types configuration related to the RTE
117 var $getKeepTags_cache=array(); // Data caching for processing function
118 var $allowedClasses=array(); // Storage of the allowed CSS class names in the RTE
119 var $preserveTags = ''; // Set to tags to preserve from Page TSconfig configuration
120
121
122
123
124
125
126
127
128
129
130 /**
131 * Initialize, setting element reference and record PID
132 *
133 * @param string Element reference, eg "tt_content:bodytext"
134 * @param integer PID of the record (page id)
135 * @return void
136 */
137 function init($elRef='',$recPid=0) {
138 $this->recPid=$recPid;
139 $this->elRef=$elRef;
140 }
141
142 /**
143 * Setting the ->relPath and ->relBackPath to proper values so absolute references to links and images can be converted to relative dittos.
144 * This is used when editing files with the RTE
145 *
146 * @param string The relative path from PATH_site to the place where the file being edited is. Eg. "fileadmin/static".
147 * @return void There is no output, it is set in internal variables. With the above example of "fileadmin/static" as input this will yield ->relPath to be "fileadmin/static/" and ->relBackPath to be "../../"
148 */
149 function setRelPath($path) {
150 $path = trim($path);
151 $path = ereg_replace('^/','',$path);
152 $path = ereg_replace('/$','',$path);
153 if ($path) {
154 $this->relPath = $path;
155 $this->relBackPath = '';
156 $partsC=count(explode('/',$this->relPath));
157 for ($a=0;$a<$partsC;$a++) {
158 $this->relBackPath.='../';
159 }
160 $this->relPath.='/';
161 }
162 }
163
164 /**
165 * Evaluate the environment for editing a staticFileEdit file.
166 * Called for almost all fields being saved in the database. Is called without an instance of the object: t3lib_parsehtml_proc::evalWriteFile()
167 *
168 * @param array Parameters for the current field as found in types-config
169 * @param array Current record we are editing.
170 * @return mixed On success an array with various information is returned, otherwise a string with an error message
171 * @see t3lib_TCEmain, t3lib_transferData
172 */
173 function evalWriteFile($pArr,$currentRecord) {
174
175 // Write file configuration:
176 if (is_array($pArr)) {
177 if ($GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath']
178 && substr($GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath'],-1)=='/'
179 && @is_dir(PATH_site.$GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath'])) {
180
181 $SW_p = $pArr['parameters'];
182 $SW_editFileField = trim($SW_p[0]);
183 $SW_editFile = $currentRecord[$SW_editFileField];
184 if ($SW_editFileField && $SW_editFile && t3lib_div::validPathStr($SW_editFile)) {
185 $SW_relpath = $GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath'].$SW_editFile;
186 $SW_editFile = PATH_site.$SW_relpath;
187 if (@is_file($SW_editFile)) {
188 return array(
189 'editFile' => $SW_editFile,
190 'relEditFile' => $SW_relpath,
191 'contentField' => trim($SW_p[1]),
192 'markerField' => trim($SW_p[2]),
193 'loadFromFileField' => trim($SW_p[3]),
194 'statusField' => trim($SW_p[4])
195 );
196 } else return "ERROR: Editfile '".$SW_relpath."' did not exist";
197 } else return "ERROR: Edit file name could not be found or was bad.";
198 } else return "ERROR: staticFileEditPath was not set, not set correctly or did not exist!";
199 }
200 }
201
202
203
204
205
206
207
208
209
210
211
212
213
214 /**********************************************
215 *
216 * Main function
217 *
218 **********************************************/
219
220 /**
221 * Tranform value for RTE based on specConf in the direction specified by $direction (rte/db)
222 * This is the main function called from tcemain and transfer data classes
223 *
224 * @param string Input value
225 * @param array Special configuration for a field; This is coming from the types-configuration of the field in the TCA. In the types-configuration you can setup features for the field rendering and in particular the RTE takes al its major configuration options from there!
226 * @param string Direction of the transformation. Two keywords are allowed; "db" or "rte". If "db" it means the transformation will clean up content coming from the Rich Text Editor and goes into the database. The other direction, "rte", is of course when content is coming from database and must be transformed to fit the RTE.
227 * @param array Parsed TypoScript content configuring the RTE, probably coming from Page TSconfig.
228 * @return string Output value
229 * @see t3lib_TCEmain::fillInFieldArray(), t3lib_transferData::renderRecord_typesProc()
230 */
231 function RTE_transform($value,$specConf,$direction='rte',$thisConfig=array()) {
232
233 // Init:
234 $this->procOptions=$thisConfig['proc.'];
235 $this->preserveTags = strtoupper(implode(',',t3lib_div::trimExplode(',',$this->procOptions['preserveTags'])));
236
237 // Get parameters for rte_transformation:
238 $p = $this->rte_p = t3lib_BEfunc::getSpecConfParametersFromArray($specConf['rte_transform']['parameters']);
239
240 // Setting modes:
241 if (strcmp($this->procOptions['overruleMode'],'')) {
242 $modes=array_unique(t3lib_div::trimExplode(',',$this->procOptions['overruleMode']));
243 } else {
244 $modes=array_unique(t3lib_div::trimExplode('-',$p['mode']));
245 }
246 $revmodes=array_flip($modes);
247
248 // Find special modes and extract them:
249 if (isset($revmodes['ts'])) {
250 $modes[$revmodes['ts']]='ts_transform,ts_preserve,ts_images,ts_links';
251 }
252 // Find special modes and extract them:
253 if (isset($revmodes['ts_css'])) {
254 $modes[$revmodes['ts_css']]='css_transform,ts_images,ts_links';
255 }
256 $modes = array_unique(t3lib_div::trimExplode(',',implode(',',$modes),1));
257 if ($direction=='rte') {
258 $modes=array_reverse($modes);
259 }
260
261 // Getting additional HTML cleaner configuration. These are applied either before or after the main transformation is done and is thus totally independant processing options you can set up:
262 $entry_HTMLparser = $this->procOptions['entryHTMLparser_'.$direction] ? $this->HTMLparserConfig($this->procOptions['entryHTMLparser_'.$direction.'.']) : '';
263 $exit_HTMLparser = $this->procOptions['exitHTMLparser_'.$direction] ? $this->HTMLparserConfig($this->procOptions['exitHTMLparser_'.$direction.'.']) : '';
264
265 // Line breaks of content is unified into char-10 only (removing char 13)
266 if (!$this->procOptions['disableUnifyLineBreaks']) {
267 $value = str_replace(chr(13).chr(10),chr(10),$value);
268 }
269
270 // In an entry-cleaner was configured, pass value through the HTMLcleaner with that:
271 if (is_array($entry_HTMLparser)) {
272 $value = $this->HTMLcleaner($value,$entry_HTMLparser[0],$entry_HTMLparser[1],$entry_HTMLparser[2],$entry_HTMLparser[3]);
273 }
274
275 // Traverse modes:
276 foreach($modes as $cmd) {
277 // ->DB
278 if ($direction=='db') {
279 switch($cmd) {
280 case 'ts_images':
281 $value=$this->TS_images_db($value);
282 break;
283 case 'ts_reglinks':
284 $value=$this->TS_reglinks($value,'db');
285 break;
286 case 'ts_links':
287 $value=$this->TS_links_db($value);
288 break;
289 case 'ts_preserve':
290 $value=$this->TS_preserve_db($value);
291 break;
292 case 'ts_transform':
293 case 'css_transform':
294 $value = str_replace(chr(13),'',$value); // Has a very disturbing effect, so just remove all '13' - depend on '10'
295 $this->allowedClasses = t3lib_div::trimExplode(',',strtoupper($this->procOptions['allowedClasses']),1);
296 $value=$this->TS_transform_db($value,$cmd=='css_transform');
297 break;
298 case 'ts_strip':
299 $value=$this->TS_strip_db($value);
300 break;
301 case 'dummy':
302 break;
303 }
304 }
305 // ->RTE
306 if ($direction=='rte') {
307 switch($cmd) {
308 case 'ts_images':
309 $value=$this->TS_images_rte($value);
310 break;
311 case 'ts_reglinks':
312 $value=$this->TS_reglinks($value,'rte');
313 break;
314 case 'ts_links':
315 $value=$this->TS_links_rte($value);
316 break;
317 case 'ts_preserve':
318 $value=$this->TS_preserve_rte($value);
319 break;
320 case 'ts_transform':
321 case 'css_transform':
322 $value = str_replace(chr(13),'',$value); // Has a very disturbing effect, so just remove all '13' - depend on '10'
323 $value=$this->TS_transform_rte($value,$cmd=='css_transform');
324 break;
325 case 'dummy':
326 break;
327 }
328 }
329 }
330
331 // In an exit-cleaner was configured, pass value through the HTMLcleaner with that:
332 if (is_array($exit_HTMLparser)) {
333 $value = $this->HTMLcleaner($value,$exit_HTMLparser[0],$exit_HTMLparser[1],$exit_HTMLparser[2],$exit_HTMLparser[3]);
334 }
335
336 // Final clean up of linebreaks:
337 if (!$this->procOptions['disableUnifyLineBreaks']) {
338 $value = str_replace(chr(13).chr(10),chr(10),$value); // Make sure no \r\n sequences has entered in the meantime...
339 $value = str_replace(chr(10),chr(13).chr(10),$value); // ... and then change all \n into \r\n
340 }
341
342 // Return value:
343 return $value;
344 }
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361 /************************************
362 *
363 * Specific RTE TRANSFORMATION functions
364 *
365 *************************************/
366
367 /**
368 * Transformation handler: 'ts_images' / direction: "db"
369 * Processing images inserted in the RTE.
370 * This is used when content goes from the RTE to the database.
371 * Images inserted in the RTE has an absolute URL applied to the src attribute. This URL is converted to a relative URL
372 * If it turns out that the URL is from another website than the current the image is read from that external URL and moved to the local server.
373 * Also "magic" images are processed here.
374 *
375 * @param string The content from RTE going to Database
376 * @return string Processed content
377 */
378 function TS_images_db($value) {
379
380 // Split content by <img> tags and traverse the resulting array for processing:
381 $imgSplit = $this->splitTags('img',$value);
382 foreach($imgSplit as $k => $v) {
383 if ($k%2) { // image found, do processing:
384
385 // Init
386 $attribArray=$this->get_tag_attributes_classic($v,1);
387 $siteUrl = $this->siteUrl();
388 $absRef = trim($attribArray['src']); // It's always a absolute URL coming from the RTE into the Database.
389
390 // External image from another URL? In that case, fetch image (unless disabled feature).
391 if (!t3lib_div::isFirstPartOfStr($absRef,$siteUrl) && !$this->procOptions['dontFetchExtPictures']) {
392 $externalFile = $this->getUrl($absRef); // Get it
393 if ($externalFile) {
394 $pU = parse_url($absRef);
395 $pI=pathinfo($pU['path']);
396
397 if (t3lib_div::inList('gif,png,jpeg,jpg',strtolower($pI['extension']))) {
398 $filename = t3lib_div::shortMD5($absRef).'.'.$pI['extension'];
399 $origFilePath = PATH_site.$this->rteImageStorageDir().'RTEmagicP_'.$filename;
400 $C_origFilePath = PATH_site.$this->rteImageStorageDir().'RTEmagicC_'.$filename.'.'.$pI['extension'];
401 if (!@is_file($origFilePath)) {
402 t3lib_div::writeFile($origFilePath,$externalFile);
403 t3lib_div::writeFile($C_origFilePath,$externalFile);
404 }
405 $absRef = $siteUrl.$this->rteImageStorageDir().'RTEmagicC_'.$filename.'.'.$pI['extension'];
406
407 $attribArray['src']=$absRef;
408 $params = t3lib_div::implodeParams($attribArray,1);
409 $imgSplit[$k]='<img '.$params.' />';
410 }
411 }
412 }
413 // Check image as local file (siteURL equals the one of the image)
414 if (t3lib_div::isFirstPartOfStr($absRef,$siteUrl)) {
415 $path = rawurldecode(substr($absRef,strlen($siteUrl))); // Rel-path, rawurldecoded for special characters.
416 $filepath = t3lib_div::getFileAbsFileName($path); // Abs filepath, locked to relative path of this project.
417
418 // Check file existence (in relative dir to this installation!)
419 if ($filepath && @is_file($filepath)) {
420
421 // If "magic image":
422 $pathPre=$this->rteImageStorageDir().'RTEmagicC_';
423 if (t3lib_div::isFirstPartOfStr($path,$pathPre)) {
424 // Find original file:
425 $pI=pathinfo(substr($path,strlen($pathPre)));
426 $filename = substr($pI['basename'],0,-strlen('.'.$pI['extension']));
427 $origFilePath = PATH_site.$this->rteImageStorageDir().'RTEmagicP_'.$filename;
428 if (@is_file($origFilePath)) {
429 $imgObj = t3lib_div::makeInstance('t3lib_stdGraphic');
430 $imgObj->init();
431 $imgObj->mayScaleUp=0;
432 $imgObj->tempPath=PATH_site.$imgObj->tempPath;
433
434 $curInfo = $imgObj->getImageDimensions($filepath); // Image dimensions of the current image
435 $curWH = $this->getWHFromAttribs($attribArray); // Image dimensions as set in the image tag
436 // Compare dimensions:
437 if ($curWH[0]!=$curInfo[0] || $curWH[1]!=$curInfo[1]) {
438 $origImgInfo = $imgObj->getImageDimensions($origFilePath); // Image dimensions of the current image
439 $cW = $curWH[0];
440 $cH = $curWH[1];
441 $cH = 1000; // Make the image based on the width solely...
442 $imgI = $imgObj->imageMagickConvert($origFilePath,$pI['extension'],$cW.'m',$cH.'m');
443 if ($imgI[3]) {
444 $fI=pathinfo($imgI[3]);
445 @copy($imgI[3],$filepath); // Override the child file
446 unset($attribArray['style']);
447 $attribArray['width']=$imgI[0];
448 $attribArray['height']=$imgI[1];
449 if (!$attribArray['border']) $attribArray['border']=0;
450 $params = t3lib_div::implodeParams($attribArray,1);
451 $imgSplit[$k]='<img '.$params.' />';
452 }
453 }
454 }
455
456 } elseif ($this->procOptions['plainImageMode']) { // If "plain image" has been configured:
457
458 // Image dimensions as set in the image tag
459 $curWH = $this->getWHFromAttribs($attribArray);
460 $attribArray['width'] = $curWH[0];
461 $attribArray['height'] = $curWH[1];
462
463 // Forcing values for style and border:
464 unset($attribArray['style']);
465 if (!$attribArray['border']) $attribArray['border']=0;
466
467 // Finding dimensions of image file:
468 $fI = @getimagesize($filepath);
469
470 // Perform corrections to aspect ratio based on configuration:
471 switch((string)$this->procOptions['plainImageMode']) {
472 case 'lockDimensions':
473 $attribArray['width']=$fI[0];
474 $attribArray['height']=$fI[1];
475 break;
476 case 'lockRatioWhenSmaller': // If the ratio has to be smaller, then first set the width...:
477 if ($attribArray['width']>$fI[0]) $attribArray['width'] = $fI[0];
478 case 'lockRatio':
479 if ($fI[0]>0) {
480 $attribArray['height']=round($attribArray['width']*($fI[1]/$fI[0]));
481 }
482 break;
483 }
484
485 // Compile the image tag again:
486 $params = t3lib_div::implodeParams($attribArray,1);
487 $imgSplit[$k]='<img '.$params.' />';
488 }
489 } else { // Remove image if it was not found in a proper position on the server!
490 $imgSplit[$k]='';
491 }
492 }
493
494 // Convert abs to rel url
495 if ($imgSplit[$k]) {
496 $attribArray=$this->get_tag_attributes_classic($imgSplit[$k],1);
497 $absRef = trim($attribArray['src']);
498 if (t3lib_div::isFirstPartOfStr($absRef,$siteUrl)) {
499 $attribArray['src'] = $this->relBackPath.substr($absRef,strlen($siteUrl));
500 if (!isset($attribArray['alt'])) $attribArray['alt']=''; // Must have alt-attribute for XHTML compliance.
501 $imgSplit[$k]='<img '.t3lib_div::implodeParams($attribArray,1,1).' />';
502 }
503 }
504 }
505 }
506 return implode('',$imgSplit);
507 }
508
509 /**
510 * Transformation handler: 'ts_images' / direction: "rte"
511 * Processing images from database content going into the RTE.
512 * Processing includes converting the src attribute to an absolute URL.
513 *
514 * @param string Content input
515 * @return string Content output
516 */
517 function TS_images_rte($value) {
518
519 // Split content by <img> tags and traverse the resulting array for processing:
520 $imgSplit = $this->splitTags('img',$value);
521 foreach($imgSplit as $k => $v) {
522 if ($k%2) { // image found:
523
524 // Init
525 $attribArray=$this->get_tag_attributes_classic($v,1);
526 $siteUrl = $this->siteUrl();
527 $absRef = trim($attribArray['src']);
528
529 // Unless the src attribute is already pointing to an external URL:
530 if (strtolower(substr($absRef,0,4))!='http') {
531 $attribArray['src'] = $siteUrl.substr($attribArray['src'],strlen($this->relBackPath));
532 if (!isset($attribArray['alt'])) $attribArray['alt']='';
533 $params = t3lib_div::implodeParams($attribArray);
534 $imgSplit[$k]='<img '.$params.' />';
535 }
536 }
537 }
538
539 // return processed content:
540 return implode('',$imgSplit);
541 }
542
543 /**
544 * Transformation handler: 'ts_reglinks' / direction: "db"+"rte" depending on $direction variable.
545 * Converting <A>-tags to/from abs/rel
546 *
547 * @param string Content input
548 * @param string Direction of conversion; "rte" (from database to RTE) or "db" (from RTE to database)
549 * @return string Content output
550 */
551 function TS_reglinks($value,$direction) {
552 switch($direction) {
553 case 'rte':
554 return $this->TS_AtagToAbs($value,1);
555 break;
556 case 'db':
557 $siteURL = $this->siteUrl();
558 $blockSplit = $this->splitIntoBlock('A',$value);
559 reset($blockSplit);
560 while(list($k,$v)=each($blockSplit)) {
561 if ($k%2) { // block:
562 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v),1);
563 // If the url is local, remove url-prefix
564 if ($siteURL && substr($attribArray['href'],0,strlen($siteURL))==$siteURL) {
565 $attribArray['href']=$this->relBackPath.substr($attribArray['href'],strlen($siteURL));
566 }
567 $bTag='<a '.t3lib_div::implodeParams($attribArray,1).'>';
568 $eTag='</a>';
569 $blockSplit[$k] = $bTag.$this->TS_reglinks($this->removeFirstAndLastTag($blockSplit[$k]),$direction).$eTag;
570 }
571 }
572 return implode('',$blockSplit);
573 break;
574 }
575 }
576
577 /**
578 * Transformation handler: 'ts_links' / direction: "db"
579 * Converting <A>-tags to <LINK tags>
580 *
581 * @param string Content input
582 * @return string Content output
583 * @see TS_links_rte()
584 */
585 function TS_links_db($value) {
586
587 // Split content into <a> tag blocks and process:
588 $blockSplit = $this->splitIntoBlock('A',$value);
589 foreach($blockSplit as $k => $v) {
590 if ($k%2) { // If an A-tag was found:
591 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v),1);
592 $info = $this->urlInfoForLinkTags($attribArray['href']);
593
594 // Check options:
595 $attribArray_copy = $attribArray;
596 unset($attribArray_copy['href']);
597 unset($attribArray_copy['target']);
598 unset($attribArray_copy['class']);
599 if (!count($attribArray_copy)) { // Only if href, target and class are the only attributes, we can alter the link!
600 // Creating the TYPO3 pseudo-tag "<LINK>" for the link (includes href/url, target and class attributes):
601 $bTag='<LINK '.$info['url'].($attribArray['target']?' '.$attribArray['target']:($attribArray['class']?' -':'')).($attribArray['class']?' '.$attribArray['class']:'').'>';
602 $eTag='</LINK>';
603 $blockSplit[$k] = $bTag.$this->TS_links_db($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
604 } else { // ... otherwise store the link as a-tag.
605 // Unsetting 'rtekeep' attribute if that had been set.
606 unset($attribArray['rtekeep']);
607 // If the url is local, remove url-prefix
608 $siteURL = $this->siteUrl();
609 if ($siteURL && substr($attribArray['href'],0,strlen($siteURL))==$siteURL) {
610 $attribArray['href']=$this->relBackPath.substr($attribArray['href'],strlen($siteURL));
611 }
612 $bTag='<a '.t3lib_div::implodeParams($attribArray,1).'>';
613 $eTag='</a>';
614 $blockSplit[$k] = $bTag.$this->TS_links_db($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
615 }
616 }
617 }
618 return implode('',$blockSplit);
619 }
620
621 /**
622 * Transformation handler: 'ts_links' / direction: "rte"
623 * Converting <LINK tags> to <A>-tags
624 *
625 * @param string Content input
626 * @return string Content output
627 * @see TS_links_rte()
628 */
629 function TS_links_rte($value) {
630 $value = $this->TS_AtagToAbs($value);
631
632 // Split content by the TYPO3 pseudo tag "<LINK>":
633 $blockSplit = $this->splitIntoBlock('link',$value,1);
634 foreach($blockSplit as $k => $v) {
635 if ($k%2) { // block:
636 $tagCode = t3lib_div::trimExplode(' ',trim(substr($this->getFirstTag($v),0,-1)),1);
637 $link_param=$tagCode[1];
638 $href='';
639 $siteUrl = $this->siteUrl();
640 // Parsing the typolink data. This parsing is roughly done like in tslib_content->typolink()
641 if(strstr($link_param,'@')) { // mailadr
642 $href = 'mailto:'.eregi_replace('^mailto:','',$link_param);
643 } elseif (substr($link_param,0,1)=='#') { // check if anchor
644 $href = $siteUrl.$link_param;
645 } else {
646 $fileChar=intval(strpos($link_param, '/'));
647 $urlChar=intval(strpos($link_param, '.'));
648
649 // Detects if a file is found in site-root OR is a simulateStaticDocument.
650 list($rootFileDat) = explode('?',$link_param);
651 $rFD_fI = pathinfo($rootFileDat);
652 if (trim($rootFileDat) && !strstr($link_param,'/') && (@is_file(PATH_site.$rootFileDat) || t3lib_div::inList('php,html,htm',strtolower($rFD_fI['extension'])))) {
653 $href = $siteUrl.$link_param;
654 } elseif($urlChar && (strstr($link_param,'//') || !$fileChar || $urlChar<$fileChar)) { // url (external): If doubleSlash or if a '.' comes before a '/'.
655 if (!ereg('^[a-z]*://',trim(strtolower($link_param)))) {$scheme='http://';} else {$scheme='';}
656 $href = $scheme.$link_param;
657 } elseif($fileChar) { // file (internal)
658 $href = $siteUrl.$link_param;
659 } else { // integer or alias (alias is without slashes or periods or commas, that is 'nospace,alphanum_x,lower,unique' according to tables.php!!)
660 $link_params_parts=explode('#',$link_param);
661 $idPart = trim($link_params_parts[0]); // Link-data del
662 if (!strcmp($idPart,'')) {$idPart=$this->recPid;} // If no id or alias is given, set it to class record pid
663 if ($link_params_parts[1] && !$sectionMark) {
664 $sectionMark='#'.trim($link_params_parts[1]);
665 }
666 // Splitting the parameter by ',' and if the array counts more than 1 element it's a id/type/? pair
667 $pairParts = t3lib_div::trimExplode(',',$idPart);
668 if (count($pairParts)>1) {
669 $idPart = $pairParts[0];
670 // Type ? future support for?
671 }
672 // Checking if the id-parameter is an alias.
673 if (!t3lib_div::testInt($idPart)) {
674 list($idPartR) = t3lib_BEfunc::getRecordsByField('pages','alias',$idPart);
675 $idPart=intval($idPartR['uid']);
676 }
677 $page = t3lib_BEfunc::getRecord('pages',$idPart);
678 if (is_array($page)) { // Page must exist...
679 $href = $siteUrl.'?id='.$link_param;
680 } else {
681 $href='';
682 $error='no page: '.$idPart;
683 }
684 }
685 }
686
687 // Setting the A-tag:
688 $bTag='<a href="'.htmlspecialchars($href).'"'.($tagCode[2]&&$tagCode[2]!='-'?' target="'.htmlspecialchars($tagCode[2]).'"':'').($tagCode[3]?' class="'.htmlspecialchars($tagCode[3]).'"':'').'>';
689 $eTag='</a>';
690 $blockSplit[$k] = $bTag.$this->TS_links_rte($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
691 }
692 }
693
694 // Return content:
695 return implode('',$blockSplit);
696 }
697
698 /**
699 * Preserve special tags
700 *
701 * @param string Content input
702 * @return string Content output
703 */
704 function TS_preserve_db($value) {
705 if (!$this->preserveTags) return $value;
706
707 // Splitting into blocks for processing (span-tags are used for special tags)
708 $blockSplit = $this->splitIntoBlock('span',$value);
709 foreach($blockSplit as $k => $v) {
710 if ($k%2) { // block:
711 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v));
712 if ($attribArray['specialtag']) {
713 $theTag = rawurldecode($attribArray['specialtag']);
714 $theTagName = $this->getFirstTagName($theTag);
715 $blockSplit[$k] = $theTag.$this->removeFirstAndLastTag($blockSplit[$k]).'</'.$theTagName.'>';
716 }
717 }
718 }
719 return implode('',$blockSplit);
720 }
721
722 /**
723 * Preserve special tags
724 *
725 * @param string Content input
726 * @return string Content output
727 */
728 function TS_preserve_rte($value) {
729 if (!$this->preserveTags) return $value;
730
731 $blockSplit = $this->splitIntoBlock($this->preserveTags,$value);
732 foreach($blockSplit as $k => $v) {
733 if ($k%2) { // block:
734 $blockSplit[$k] = '<span specialtag="'.rawurlencode($this->getFirstTag($v)).'">'.$this->removeFirstAndLastTag($blockSplit[$k]).'</span>';
735 }
736 }
737 return implode('',$blockSplit);
738 }
739
740 /**
741 * Transformation handler: 'ts_transform' + 'css_transform' / direction: "db"
742 * Cleaning (->db) for standard content elements (ts)
743 *
744 * @param string Content input
745 * @param boolean If true, the transformation was "css_transform", otherwise "ts_transform"
746 * @return string Content output
747 * @see TS_transform_rte()
748 */
749 function TS_transform_db($value,$css=FALSE) {
750
751 // safety... so forever loops are avoided (they should not occur, but an error would potentially do this...)
752 $this->TS_transform_db_safecounter--;
753 if ($this->TS_transform_db_safecounter<0) return $value;
754
755 // Split the content from RTE by the occurence of these blocks:
756 $blockSplit = $this->splitIntoBlock('TABLE,BLOCKQUOTE,'.$this->headListTags,$value);
757
758 $cc=0;
759 $aC = count($blockSplit);
760
761 // Traverse the blocks
762 foreach($blockSplit as $k => $v) {
763 $cc++;
764 $lastBR = $cc==$aC ? '' : chr(10);
765
766 if ($k%2) { // Inside block:
767
768 // Init:
769 $tag=$this->getFirstTag($v);
770 $tagName=strtolower($this->getFirstTagName($v));
771
772 // Process based on the tag:
773 switch($tagName) {
774 case 'blockquote': // Keep blockquotes, but clean the inside recursively in the same manner as the main code
775 $blockSplit[$k]='<'.$tagName.'>'.$this->TS_transform_db($this->removeFirstAndLastTag($blockSplit[$k]),$css).'</'.$tagName.'>'.$lastBR;
776 break;
777 case 'ol':
778 case 'ul': // Transform lists into <typolist>-tags:
779 if (!$css) {
780 if (!isset($this->procOptions['typolist']) || $this->procOptions['typolist']) {
781 $parts = $this->getAllParts($this->splitIntoBlock('LI',$this->removeFirstAndLastTag($blockSplit[$k])),1,0);
782 while(list($k2)=each($parts)) {
783 $parts[$k2]=ereg_replace(chr(10).'|'.chr(13),'',$parts[$k2]); // remove all linesbreaks!
784 $parts[$k2]=$this->defaultTStagMapping($parts[$k2],'db');
785 $parts[$k2]=$this->cleanFontTags($parts[$k2],0,0,0);
786 $parts[$k2] = $this->HTMLcleaner_db($parts[$k2],strtolower($this->procOptions['allowTagsInTypolists']?$this->procOptions['allowTagsInTypolists']:'br,font,b,i,u,a,img,span,strong,em'));
787 }
788 if ($tagName=='ol') { $params=' type="1"'; } else { $params=''; }
789 $blockSplit[$k]='<typolist'.$params.'>'.chr(10).implode(chr(10),$parts).chr(10).'</typolist>'.$lastBR;
790 }
791 } else {
792 $blockSplit[$k].=$lastBR;
793 }
794 break;
795 case 'table': // Tables are NOT allowed in any form (unless preserveTables is set or CSS is the mode)
796 if (!$this->procOptions['preserveTables'] && !$css) {
797 $blockSplit[$k]=$this->TS_transform_db($this->removeTables($blockSplit[$k]));
798 } else {
799 $blockSplit[$k]=str_replace(chr(10),'',$blockSplit[$k]).$lastBR;
800 }
801 break;
802 case 'h1':
803 case 'h2':
804 case 'h3':
805 case 'h4':
806 case 'h5':
807 case 'h6':
808 if (!$css) {
809 $attribArray=$this->get_tag_attributes_classic($tag);
810 // Processing inner content here:
811 $innerContent = $this->HTMLcleaner_db($this->removeFirstAndLastTag($blockSplit[$k]));
812
813 if (!isset($this->procOptions['typohead']) || $this->procOptions['typohead']) {
814 $type = intval(substr($tagName,1));
815 $blockSplit[$k]='<typohead'.
816 ($type!=6?' type="'.$type.'"':'').
817 ($attribArray['align']?' align="'.$attribArray['align'].'"':'').
818 ($attribArray['class']?' class="'.$attribArray['class'].'"':'').
819 '>'.
820 $innerContent.
821 '</typohead>'.
822 $lastBR;
823 } else {
824 $blockSplit[$k]='<'.$tagName.
825 ($attribArray['align']?' align="'.htmlspecialchars($attribArray['align']).'"':'').
826 ($attribArray['class']?' class="'.htmlspecialchars($attribArray['class']).'"':'').
827 '>'.
828 $innerContent.
829 '</'.$tagName.'>'.
830 $lastBR;
831 }
832 } else {
833 $blockSplit[$k].=$lastBR;
834 }
835 break;
836 default:
837 $blockSplit[$k].=$lastBR;
838 break;
839 }
840 } else { // NON-block:
841 if (strcmp(trim($blockSplit[$k]),'')) {
842 $blockSplit[$k]=$this->divideIntoLines($blockSplit[$k]).$lastBR;
843 } else unset($blockSplit[$k]);
844 }
845 }
846 $this->TS_transform_db_safecounter++;
847
848 return implode('',$blockSplit);
849 }
850
851 /**
852 * Transformation handler: 'ts_transform' + 'css_transform' / direction: "rte"
853 * Set (->rte) for standard content elements (ts)
854 *
855 * @param string Content input
856 * @param boolean If true, the transformation was "css_transform", otherwise "ts_transform"
857 * @return string Content output
858 * @see TS_transform_db()
859 */
860 function TS_transform_rte($value,$css=0) {
861
862 // Split the content from Database by the occurence of these blocks:
863 $blockSplit = $this->splitIntoBlock('TABLE,BLOCKQUOTE,TYPOLIST,TYPOHEAD,'.$this->headListTags,$value);
864
865 // Traverse the blocks
866 foreach($blockSplit as $k => $v) {
867 if ($k%2) { // Inside one of the blocks:
868
869 // Init:
870 $tag=$this->getFirstTag($v);
871 $tagName=strtolower($this->getFirstTagName($v));
872 $attribArray=$this->get_tag_attributes_classic($tag);
873
874 // Based on tagname, we do transformations:
875 switch($tagName) {
876 case 'blockquote': // Keep blockquotes:
877 $blockSplit[$k] = $tag.
878 $this->TS_transform_rte($this->removeFirstAndLastTag($blockSplit[$k]),$css).
879 '</'.$tagName.'>';
880 break;
881 case 'typolist': // Transform typolist blocks into OL/UL lists. Type 1 is expected to be numerical block
882 if (!isset($this->procOptions['typolist']) || $this->procOptions['typolist']) {
883 $tListContent = $this->removeFirstAndLastTag($blockSplit[$k]);
884 $tListContent = ereg_replace('^[ ]*'.chr(10),'',$tListContent);
885 $tListContent = ereg_replace(chr(10).'[ ]*$','',$tListContent);
886 $lines=explode(chr(10),$tListContent);
887 $typ= $attribArray['type']==1?'ol':'ul';
888 $blockSplit[$k] = '<'.$typ.'>'.chr(10).
889 '<li>'.implode('</li>'.chr(10).'<li>',$lines).'</li>'.
890 '</'.$typ.'>';
891 }
892 break;
893 case 'typohead': // Transform typohead into Hx tags.
894 if (!isset($this->procOptions['typohead']) || $this->procOptions['typohead']) {
895 $tC=$this->removeFirstAndLastTag($blockSplit[$k]);
896 $typ=t3lib_div::intInRange($attribArray['type'],0,6);
897 if (!$typ) $typ=6;
898 $align = $attribArray['align']?' align="'.$attribArray['align'].'"': '';
899 $class = $attribArray['class']?' class="'.$attribArray['class'].'"': '';
900 $blockSplit[$k] = '<h'.$typ.$align.$class.'>'.
901 $tC.
902 '</h'.$typ.'>';
903 }
904 break;
905 }
906 $blockSplit[$k+1]=ereg_replace('^[ ]*'.chr(10),'',$blockSplit[$k+1]); // Removing linebreak if typohead
907 } else { // NON-block:
908 $nextFTN = $this->getFirstTagName($blockSplit[$k+1]);
909 $singleLineBreak = $blockSplit[$k]==chr(10);
910 if (t3lib_div::inList('TABLE,BLOCKQUOTE,TYPOLIST,TYPOHEAD,'.$this->headListTags,$nextFTN)) { // Removing linebreak if typolist/typohead
911 $blockSplit[$k]=ereg_replace(chr(10).'[ ]*$','',$blockSplit[$k]);
912 }
913 // If $blockSplit[$k] is blank then unset the line. UNLESS the line happend to be a single line break.
914 if (!strcmp($blockSplit[$k],'') && !$singleLineBreak) {
915 unset($blockSplit[$k]);
916 } else {
917 $blockSplit[$k]=$this->setDivTags($blockSplit[$k],($this->procOptions['useDIVasParagraphTagForRTE']?'div':'p'));
918 }
919 }
920 }
921 return implode(chr(10),$blockSplit);
922 }
923
924 /**
925 * Transformation handler: 'ts_strip' / direction: "db"
926 * Removing all non-allowed tags
927 *
928 * @param string Content input
929 * @return string Content output
930 */
931 function TS_strip_db($value) {
932 $value = strip_tags($value,'<'.implode('><',explode(',','b,i,u,a,img,br,div,center,pre,font,hr,sub,sup,p,strong,em,li,ul,ol,blockquote')).'>');
933 return $value;
934 }
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949 /***************************************************************
950 *
951 * Generic RTE transformation, analysis and helper functions
952 *
953 **************************************************************/
954
955 /**
956 * Reads the file or url $url and returns the content
957 *
958 * @param string Filepath/URL to read
959 * @return string The content from the resource given as input.
960 * @see t3lib_div::getURL()
961 */
962 function getURL($url) {
963 return t3lib_div::getURL($url);
964 }
965
966 /**
967 * Function for cleaning content going into the database.
968 * Content is cleaned eg. by removing unallowed HTML and ds-HSC content
969 * It is basically calling HTMLcleaner from the parent class with some preset configuration specifically set up for cleaning content going from the RTE into the db
970 *
971 * @param string Content to clean up
972 * @param string Comma list of tags to specifically allow. Default comes from getKeepTags and is ""
973 * @return string Clean content
974 * @see getKeepTags()
975 */
976 function HTMLcleaner_db($content,$tagList='') {
977 if (!$tagList) {
978 $keepTags = $this->getKeepTags('db');
979 } else {
980 $keepTags = $this->getKeepTags('db',$tagList);
981 }
982 $kUknown = $this->procOptions['dontRemoveUnknownTags_db'] ? 1 : 0; // Default: remove unknown tags.
983 $hSC = $this->procOptions['dontUndoHSC_db'] ? 0 : -1; // Default: re-convert literals to characters (that is &lt; to <)
984
985 return $this->HTMLcleaner($content,$keepTags,$kUknown,$hSC);
986 }
987
988 /**
989 * Creates an array of configuration for the HTMLcleaner function based on whether content go TO or FROM the Rich Text Editor ($direction)
990 * Unless "tagList" is given, the function will cache the configuration for next time processing goes on. (In this class that is the case only if we are processing a bulletlist)
991 *
992 * @param string The direction of the content being processed by the output configuration; "db" (content going into the database FROM the rte) or "rte" (content going into the form)
993 * @param string Comma list of tags to keep (overriding default which is to keep all + take notice of internal configuration)
994 * @return array Configuration array
995 * @see HTMLcleaner_db()
996 */
997 function getKeepTags($direction='rte',$tagList='') {
998 if (!is_array($this->getKeepTags_cache[$direction]) || $tagList) {
999
1000 // Setting up allowed tags:
1001 if (strcmp($tagList,'')) { // If the $tagList input var is set, this will take precedence
1002 $keepTags = array_flip(t3lib_div::trimExplode(',',$tagList,1));
1003 } else { // Default is to get allowed/denied tags from internal array of processing options:
1004 // Construct default list of tags to keep:
1005 $typoScript_list = 'b,i,u,a,img,br,div,center,pre,font,hr,sub,sup,p,strong,em,li,ul,ol,blockquote,strike,span';
1006 $keepTags = array_flip(t3lib_div::trimExplode(',',$typoScript_list.','.strtolower($this->procOptions['allowTags']),1));
1007
1008 // For tags to deny, remove them from $keepTags array:
1009 $denyTags = t3lib_div::trimExplode(',',$this->procOptions['denyTags'],1);
1010 foreach($denyTags as $dKe) {
1011 unset($keepTags[$dKe]);
1012 }
1013 }
1014
1015 // Based on the direction of content, set further options:
1016 switch ($direction) {
1017
1018 // GOING from database to Rich Text Editor:
1019 case 'rte':
1020 // Transform bold/italics tags to strong/em
1021 if (isset($keepTags['b'])) {$keepTags['b']=array('remap'=>'STRONG');}
1022 if (isset($keepTags['i'])) {$keepTags['i']=array('remap'=>'EM');}
1023
1024 // Transforming keepTags array so it can be understood by the HTMLcleaner function. This basically converts the format of the array from TypoScript (having .'s) to plain multi-dimensional array.
1025 list($keepTags) = $this->HTMLparserConfig($this->procOptions['HTMLparser_rte.'],$keepTags);
1026 break;
1027
1028 // GOING from RTE to database:
1029 case 'db':
1030 // Transform strong/em back to bold/italics:
1031 if (isset($keepTags['strong'])) { $keepTags['strong']=array('remap'=>'b'); }
1032 if (isset($keepTags['em'])) { $keepTags['em']=array('remap'=>'i'); }
1033
1034 // Setting up span tags if they are allowed:
1035 if (isset($keepTags['span'])) {
1036 $classes=array_merge(array(''),$this->allowedClasses);
1037 $keepTags['span']=array(
1038 'allowedAttribs'=>'class',
1039 'fixAttrib' => Array(
1040 'class' => Array (
1041 'list' => $classes,
1042 'removeIfFalse' => 1
1043 )
1044 ),
1045 'rmTagIfNoAttrib' => 1
1046 );
1047 if (!$this->procOptions['allowedClasses']) unset($keepTags['span']['fixAttrib']['class']['list']);
1048 }
1049
1050 // Setting up font tags if they are allowed:
1051 if (isset($keepTags['font'])) {
1052 $colors=array_merge(array(''),t3lib_div::trimExplode(',',$this->procOptions['allowedFontColors'],1));
1053 $keepTags['font']=array(
1054 'allowedAttribs'=>'face,color,size',
1055 'fixAttrib' => Array(
1056 'face' => Array (
1057 'removeIfFalse' => 1
1058 ),
1059 'color' => Array (
1060 'removeIfFalse' => 1,
1061 'list'=>$colors
1062 ),
1063 'size' => Array (
1064 'removeIfFalse' => 1,
1065 )
1066 ),
1067 'rmTagIfNoAttrib' => 1
1068 );
1069 if (!$this->procOptions['allowedFontColors']) unset($keepTags['font']['fixAttrib']['color']['list']);
1070 }
1071
1072 // Setting further options, getting them from the processiong options:
1073 $TSc = $this->procOptions['HTMLparser_db.'];
1074 if (!$TSc['globalNesting']) $TSc['globalNesting']='b,i,u,a,center,font,sub,sup,strong,em,strike,span';
1075 if (!$TSc['noAttrib']) $TSc['noAttrib']='b,i,u,br,center,hr,sub,sup,strong,em,li,ul,ol,blockquote,strike';
1076
1077 // Transforming the array from TypoScript to regular array:
1078 list($keepTags) = $this->HTMLparserConfig($TSc,$keepTags);
1079 break;
1080 }
1081
1082 // Caching (internally, in object memory) the result unless tagList is set:
1083 if (!$tagList) {
1084 $this->getKeepTags_cache[$direction] = $keepTags;
1085 } else {
1086 return $keepTags;
1087 }
1088 }
1089
1090 // Return result:
1091 return $this->getKeepTags_cache[$direction];
1092 }
1093
1094 /**
1095 * This resolves the $value into parts based on <div></div>-sections and <p>-sections and <br />-tags. These are returned as lines separated by chr(10).
1096 * This point is to resolve the HTML-code returned from RTE into ordinary lines so it's 'human-readable'
1097 * The function ->setDivTags does the opposite.
1098 * This function processes content to go into the database.
1099 *
1100 * @param string Value to process.
1101 * @param integer Recursion brake. Decremented on each recursion down to zero. Default is 5 (which equals the allowed nesting levels of p/div tags).
1102 * @param boolean If true, an array with the lines is returned, otherwise a string of the processed input value.
1103 * @return string Processed input value.
1104 * @see setDivTags()
1105 */
1106 function divideIntoLines($value,$count=5,$returnArray=FALSE) {
1107
1108 // Internalize font tags (move them from OUTSIDE p/div to inside it that is the case):
1109 if ($this->procOptions['internalizeFontTags']) {$value = $this->internalizeFontTags($value);}
1110
1111 // Setting configuration for processing:
1112 $allowTagsOutside = t3lib_div::trimExplode(',',strtolower($this->procOptions['allowTagsOutside']?$this->procOptions['allowTagsOutside']:'img'),1);
1113 $remapParagraphTag = strtoupper($this->procOptions['remapParagraphTag']);
1114 $divSplit = $this->splitIntoBlock('div,p',$value,1); // Setting the third param to 1 will eliminate false end-tags. Maybe this is a good thing to do...?
1115
1116 if ($this->procOptions['keepPDIVattribs']) {
1117 $keepAttribListArr = t3lib_div::trimExplode(',',strtolower($this->procOptions['keepPDIVattribs']),1);
1118 } else {
1119 $keepAttribListArr = array();
1120 }
1121
1122 // Returns plainly the value if there was no div/p sections in it
1123 if (count($divSplit)<=1 || $count<=0) {
1124 return $value;
1125 }
1126
1127 // Traverse the splitted sections:
1128 foreach($divSplit as $k => $v) {
1129 if ($k%2) { // Inside
1130 $v=$this->removeFirstAndLastTag($v);
1131
1132 // Fetching 'sub-lines' - which will explode any further p/div nesting...
1133 $subLines = $this->divideIntoLines($v,$count-1,1);
1134 if (is_array($subLines)) { // So, if there happend to be sub-nesting of p/div, this is written directly as the new content of THIS section. (This would be considered 'an error')
1135 // No noting.
1136 } else { //... but if NO subsection was found, we process it as a TRUE line without erronous content:
1137 $subLines=array($subLines);
1138 if (!$this->procOptions['dontConvBRtoParagraph']) { // process break-tags, if configured for. Simply, the breaktags will here be treated like if each was a line of content...
1139 $subLines = spliti('<br[[:space:]]*[\/]?>',$v);
1140 }
1141
1142 // Traverse sublines (there is typically one, except if <br/> has been converted to lines as well!)
1143 reset($subLines);
1144 while(list($sk)=each($subLines)) {
1145
1146 // Clear up the subline for DB.
1147 $subLines[$sk]=$this->HTMLcleaner_db($subLines[$sk]);
1148
1149 // Get first tag, attributes etc:
1150 $fTag = $this->getFirstTag($divSplit[$k]);
1151 $tagName=strtolower($this->getFirstTagName($divSplit[$k]));
1152 $attribs=$this->get_tag_attributes($fTag);
1153
1154 // Keep attributes (lowercase)
1155 $newAttribs=array();
1156 if (count($keepAttribListArr)) {
1157 foreach($keepAttribListArr as $keepA) {
1158 if (isset($attribs[0][$keepA])) { $newAttribs[$keepA] = $attribs[0][$keepA]; }
1159 }
1160 }
1161
1162 // ALIGN attribute:
1163 if (!$this->procOptions['skipAlign'] && strcmp(trim($attribs[0]['align']),'') && strtolower($attribs[0]['align'])!='left') { // Set to value, but not 'left'
1164 $newAttribs['align']=strtolower($attribs[0]['align']);
1165 }
1166
1167 // CLASS attribute:
1168 if (!$this->procOptions['skipClass'] && strcmp(trim($attribs[0]['class']),'')) { // Set to whatever value
1169 if (!count($this->allowedClasses) || in_array(strtoupper($attribs[0]['class']),$this->allowedClasses)) {
1170 $newAttribs['class']=$attribs[0]['class'];
1171 }
1172 }
1173
1174 // Remove any line break char (10 or 13)
1175 $subLines[$sk]=ereg_replace(chr(10).'|'.chr(13),'',$subLines[$sk]);
1176
1177 // If there are any attributes or if we are supposed to remap the tag, then do so:
1178 if (count($newAttribs) && strcmp($remapParagraphTag,'1')) {
1179 if ($remapParagraphTag=='P') $tagName='p';
1180 if ($remapParagraphTag=='DIV') $tagName='div';
1181 $subLines[$sk]='<'.trim($tagName.' '.$this->compileTagAttribs($newAttribs)).'>'.$subLines[$sk].'</'.$tagName.'>';
1182 }
1183 }
1184 }
1185 // Add the processed line(s)
1186 $divSplit[$k] = implode(chr(10),$subLines);
1187
1188 // If it turns out the line is just blank (containing a &nbsp; possibly) then just make it pure blank:
1189 if (trim(strip_tags($divSplit[$k]))=='&nbsp;') $divSplit[$k]='';
1190 } else { // outside div:
1191 // Remove positions which are outside div/p tags and without content
1192 $divSplit[$k]=trim(strip_tags($divSplit[$k],'<'.implode('><',$allowTagsOutside).'>'));
1193 if (!strcmp($divSplit[$k],'')) unset($divSplit[$k]); // Remove part if it's empty
1194 }
1195 }
1196
1197 // Return value:
1198 return $returnArray ? $divSplit : implode(chr(10),$divSplit);
1199 }
1200
1201 /**
1202 * Converts all lines into <div></div>/<p></p>-sections (unless the line is a div-section already)
1203 * For processing of content going FROM database TO RTE.
1204 *
1205 * @param string Value to convert
1206 * @param string Tag to wrap with. Either "p" or "div" should it be. Lowercase preferably.
1207 * @return string Processed value.
1208 * @see divideIntoLines()
1209 */
1210 function setDivTags($value,$dT='p') {
1211
1212 // First, setting configuration for the HTMLcleaner function. This will process each line between the <div>/<p> section on their way to the RTE
1213 $keepTags = $this->getKeepTags('rte');
1214 $kUknown = $this->procOptions['dontProtectUnknownTags_rte'] ? 0 : 'protect'; // Default: remove unknown tags.
1215 $hSC = $this->procOptions['dontHSC_rte'] ? 0 : 1; // Default: re-convert literals to characters (that is &lt; to <)
1216 $convNBSP = !$this->procOptions['dontConvAmpInNBSP_rte']?1:0;
1217
1218 // Divide the content into lines, based on chr(10):
1219 $parts = explode(chr(10),$value);
1220 foreach($parts as $k => $v) {
1221
1222 // Processing of line content:
1223 if (!strcmp(trim($parts[$k]),'')) { // If the line is blank, set it to &nbsp;
1224 $parts[$k]='&nbsp;';
1225 } else { // Clean the line content:
1226 $parts[$k]=$this->HTMLcleaner($parts[$k],$keepTags,$kUknown,$hSC);
1227 if ($convNBSP) $parts[$k]=str_replace('&amp;nbsp;','&nbsp;',$parts[$k]);
1228 }
1229
1230 // Wrapping the line in <$dT> is not already wrapped:
1231 $testStr = strtolower(trim($parts[$k]));
1232 if (substr($testStr,0,4)!='<div' || substr($testStr,-6)!='</div>') {
1233 if (substr($testStr,0,2)!='<p' || substr($testStr,-4)!='</p>') {
1234 // Only set p-tags if there is not already div or p tags:
1235 $parts[$k]='<'.$dT.'>'.$parts[$k].'</'.$dT.'>';
1236 }
1237 }
1238 }
1239
1240 // Implode result:
1241 return implode(chr(10),$parts);
1242 }
1243
1244 /**
1245 * This splits the $value in font-tag chunks.
1246 * If there are any <P>/<DIV> sections inside of them, the font-tag is wrapped AROUND the content INSIDE of the P/DIV sections and the outer font-tag is removed.
1247 * This functions seems to be a good choice for pre-processing content if it has been pasted into the RTE from eg. star-office.
1248 * In that case the font-tags are normally on the OUTSIDE of the sections.
1249 * This function is used by eg. divideIntoLines() if the procesing option 'internalizeFontTags' is set.
1250 *
1251 * @param string Input content
1252 * @return string Output content
1253 * @see divideIntoLines()
1254 */
1255 function internalizeFontTags($value) {
1256
1257 // Splitting into font tag blocks:
1258 $fontSplit = $this->splitIntoBlock('font',$value);
1259
1260 foreach($fontSplit as $k => $v) {
1261 if ($k%2) { // Inside
1262 $fTag = $this->getFirstTag($v); // Fint font-tag
1263
1264 $divSplit_sub = $this->splitIntoBlock('div,p',$this->removeFirstAndLastTag($v),1);
1265 if (count($divSplit_sub)>1) { // If there were div/p sections inside the font-tag, do something about it...
1266 // traverse those sections:
1267 foreach($divSplit_sub as $k2 => $v2) {
1268 if ($k2%2) { // Inside
1269 $div_p = $this->getFirstTag($v2); // Fint font-tag
1270 $div_p_tagname = $this->getFirstTagName($v2); // Fint font-tag
1271 $v2=$this->removeFirstAndLastTag($v2); // ... and remove it from original.
1272 $divSplit_sub[$k2]=$div_p.$fTag.$v2.'</font>'.'</'.$div_p_tagname.'>';
1273 } elseif (trim(strip_tags($v2))) {
1274 $divSplit_sub[$k2]=$fTag.$v2.'</font>';
1275 }
1276 }
1277 $fontSplit[$k]=implode('',$divSplit_sub);
1278 }
1279 }
1280 }
1281
1282 return implode('',$fontSplit);
1283 }
1284
1285 /**
1286 * Returns SiteURL based on thisScript.
1287 *
1288 * @return string Value of t3lib_div::getIndpEnv('TYPO3_SITE_URL');
1289 * @see t3lib_div::getIndpEnv()
1290 */
1291 function siteUrl() {
1292 return t3lib_div::getIndpEnv('TYPO3_SITE_URL');
1293 }
1294
1295 /**
1296 * Return the storage folder of RTE image files.
1297 * Default is $GLOBALS['TYPO3_CONF_VARS']['BE']['RTE_imageStorageDir'] unless something else is configured in the types configuration for the RTE.
1298 *
1299 * @return string
1300 */
1301 function rteImageStorageDir() {
1302 return $this->rte_p['imgpath'] ? $this->rte_p['imgpath'] : $GLOBALS['TYPO3_CONF_VARS']['BE']['RTE_imageStorageDir'];
1303 }
1304
1305 /**
1306 * Remove all tables from incoming code
1307 * The function is trying to to this is some more or less respectfull way. The approach is to resolve each table cells content and implode it all by <br /> chars. Thus at least the content is preserved in some way.
1308 *
1309 * @param string Input value
1310 * @param string Break character to use for linebreaks.
1311 * @return string Output value
1312 */
1313 function removeTables($value,$breakChar='<br />') {
1314
1315 // Splitting value into table blocks:
1316 $tableSplit = $this->splitIntoBlock('table',$value);
1317
1318 // Traverse blocks of tables:
1319 foreach($tableSplit as $k => $v) {
1320 if ($k%2) {
1321 $tableSplit[$k]='';
1322 $rowSplit = $this->splitIntoBlock('tr',$v);
1323 foreach($rowSplit as $k2 => $v2) {
1324 if ($k2%2) {
1325 $cellSplit = $this->getAllParts($this->splitIntoBlock('td',$v2),1,0);
1326 foreach($cellSplit as $k3 => $v3) {
1327 $tableSplit[$k].=$v3.$breakChar;
1328 }
1329 }
1330 }
1331 }
1332 }
1333
1334 // Implode it all again:
1335 return implode($breakChar,$tableSplit);
1336 }
1337
1338 /**
1339 * Default tag mapping for TS
1340 *
1341 * @param string Input code to process
1342 * @param string Direction To databsae (db) or from database to RTE (rte)
1343 * @return string Processed value
1344 */
1345 function defaultTStagMapping($code,$direction='rte') {
1346 if ($direction=='db') {
1347 $code=$this->mapTags($code,array( // Map tags
1348 'strong' => 'b',
1349 'em' => 'i'
1350 ));
1351 }
1352 if ($direction=='rte') {
1353 $code=$this->mapTags($code,array( // Map tags
1354 'b' => 'strong',
1355 'i' => 'em'
1356 ));
1357 }
1358 return $code;
1359 }
1360
1361 /**
1362 * Finds width and height from attrib-array
1363 * If the width and height is found in the style-attribute, use that!
1364 *
1365 * @param array Array of attributes from tag in which to search. More specifically the content of the key "style" is used to extract "width:xxx / height:xxx" information
1366 * @return array Integer w/h in key 0/1. Zero is returned if not found.
1367 */
1368 function getWHFromAttribs($attribArray) {
1369 $style =trim($attribArray['style']);
1370 if ($style) {
1371 $regex='[[:space:]]*:[[:space:]]*([0-9]*)[[:space:]]*px';
1372 // Width
1373 eregi('width'.$regex,$style,$reg);
1374 $w = intval($reg[1]);
1375 // Height
1376 eregi('height'.$regex,$style,$reg);
1377 $h = intval($reg[1]);
1378 }
1379 if (!$w) {
1380 $w = $attribArray['width'];
1381 }
1382 if (!$h) {
1383 $h = $attribArray['height'];
1384 }
1385 return array(intval($w),intval($h));
1386 }
1387
1388 /**
1389 * Parse <A>-tag href and return status of email,external,file or page
1390 *
1391 * @param string URL to analyse.
1392 * @return array Information in an array about the URL
1393 */
1394 function urlInfoForLinkTags($url) {
1395 $info = array();
1396 $url = trim($url);
1397 if (substr(strtolower($url),0,7)=='mailto:') {
1398 $info['url']=trim(substr($url,7));
1399 $info['type']='email';
1400 } else {
1401 $curURL = $this->siteUrl(); // 100502, removed this: 'http://'.t3lib_div::getThisUrl(); Reason: The url returned had typo3/ in the end - should be only the site's url as far as I see...
1402 for($a=0;$a<strlen($url);$a++) {
1403 if ($url[$a]!=$curURL[$a]) {
1404 break;
1405 }
1406 }
1407
1408 $info['relScriptPath']=substr($curURL,$a);
1409 $info['relUrl']=substr($url,$a);
1410 $info['url']=$url;
1411 $info['type']='ext';
1412
1413 $siteUrl_parts = parse_url($url);
1414 $curUrl_parts = parse_url($curURL);
1415
1416 if ($siteUrl_parts['host']==$curUrl_parts['host'] // Hosts should match
1417 && (!$info['relScriptPath'] || (defined('TYPO3_mainDir') && substr($info['relScriptPath'],0,strlen(TYPO3_mainDir))==TYPO3_mainDir))) { // If the script path seems to match or is empty (FE-EDIT)
1418
1419 // New processing order 100502
1420 $uP=parse_url($info['relUrl']);
1421
1422 if (!strcmp('#'.$siteUrl_parts['fragment'],$info['relUrl'])) {
1423 $info['url']=$info['relUrl'];
1424 $info['type']='anchor';
1425 } elseif (!trim($uP['path']) || !strcmp($uP['path'],'index.php')) {
1426 $pp = explode('id=',$uP['query']);
1427 $id = trim($pp[1]);
1428 if ($id) {
1429 $info['pageid']=$id;
1430 $info['cElement']=$uP['fragment'];
1431 $info['url']=$id.($info['cElement']?'#'.$info['cElement']:'');
1432 $info['type']='page';
1433 }
1434 } else {
1435 $info['url']=$info['relUrl'];
1436 $info['type']='file';
1437 }
1438 } else {
1439 unset($info['relScriptPath']);
1440 unset($info['relUrl']);
1441 }
1442 }
1443 return $info;
1444 }
1445
1446 /**
1447 * Converting <A>-tags to absolute URLs (+ setting rtekeep attribute)
1448 *
1449 * @param string Content input
1450 * @param boolean If true, then the "rtekeep" attribute will not be set.
1451 * @return string Content output
1452 */
1453 function TS_AtagToAbs($value,$dontSetRTEKEEP=FALSE) {
1454 $blockSplit = $this->splitIntoBlock('A',$value);
1455 reset($blockSplit);
1456 while(list($k,$v)=each($blockSplit)) {
1457 if ($k%2) { // block:
1458 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v),1);
1459 // Checking if there is a scheme, and if not, prepend the current url.
1460 $uP = parse_url(strtolower($attribArray['href']));
1461 if (!$uP['scheme']) {
1462 $attribArray['href']=$this->siteUrl().substr($attribArray['href'],strlen($this->relBackPath));
1463 }
1464 if (!$dontSetRTEKEEP) $attribArray['rtekeep']=1;
1465
1466 $bTag='<a '.t3lib_div::implodeParams($attribArray,1).'>';
1467 $eTag='</a>';
1468 $blockSplit[$k] = $bTag.$this->TS_AtagToAbs($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
1469 }
1470 }
1471 return implode('',$blockSplit);
1472 }
1473 }
1474
1475
1476 if (defined('TYPO3_MODE') && $TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml_proc.php']) {
1477 include_once($TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml_proc.php']);
1478 }
1479 ?>