Quote table names in admin_get_XXX() functions of t3lib_db
[Packages/TYPO3.CMS.git] / t3lib / class.t3lib_parsehtml_proc.php
1 <?php
2 /***************************************************************
3 * Copyright notice
4 *
5 * (c) 1999-2006 Kasper Skaarhoj (kasperYYYY@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 <kasperYYYY@typo3.com>
35 * @internal
36 */
37 /**
38 * [CLASS/FUNCTION INDEX of SCRIPT]
39 *
40 *
41 *
42 * 103: class t3lib_parsehtml_proc extends t3lib_parsehtml
43 * 138: function init($elRef='',$recPid=0)
44 * 150: function setRelPath($path)
45 * 174: function evalWriteFile($pArr,$currentRecord)
46 *
47 * SECTION: Main function
48 * 232: function RTE_transform($value,$specConf,$direction='rte',$thisConfig=array())
49 *
50 * SECTION: Specific RTE TRANSFORMATION functions
51 * 398: function TS_images_db($value)
52 * 550: function TS_images_rte($value)
53 * 589: function TS_reglinks($value,$direction)
54 * 626: function TS_links_db($value)
55 * 675: function TS_links_rte($value)
56 * 760: function TS_preserve_db($value)
57 * 784: function TS_preserve_rte($value)
58 * 805: function TS_transform_db($value,$css=FALSE)
59 * 922: function transformStyledATags($value)
60 * 948: function TS_transform_rte($value,$css=0)
61 * 1019: function TS_strip_db($value)
62 *
63 * SECTION: Generic RTE transformation, analysis and helper functions
64 * 1050: function getURL($url)
65 * 1064: function HTMLcleaner_db($content,$tagList='')
66 * 1091: function getKeepTags($direction='rte',$tagList='')
67 * 1200: function divideIntoLines($value,$count=5,$returnArray=FALSE)
68 * 1304: function setDivTags($value,$dT='p')
69 * 1349: function internalizeFontTags($value)
70 * 1385: function siteUrl()
71 * 1395: function rteImageStorageDir()
72 * 1407: function removeTables($value,$breakChar='<br />')
73 * 1439: function defaultTStagMapping($code,$direction='rte')
74 * 1462: function getWHFromAttribs($attribArray)
75 * 1489: function urlInfoForLinkTags($url)
76 * 1548: function TS_AtagToAbs($value,$dontSetRTEKEEP=FALSE)
77 *
78 * TOTAL FUNCTIONS: 28
79 * (This index is automatically created/updated by the extension "extdeveval")
80 *
81 */
82
83 require_once (PATH_t3lib.'class.t3lib_parsehtml.php');
84
85
86
87
88
89
90
91
92
93
94
95
96 /**
97 * Class for parsing HTML for the Rich Text Editor. (also called transformations)
98 *
99 * @author Kasper Skaarhoj <kasperYYYY@typo3.com>
100 * @package TYPO3
101 * @subpackage t3lib
102 */
103 class t3lib_parsehtml_proc extends t3lib_parsehtml {
104
105 // Static:
106 var $headListTags = 'PRE,UL,OL,H1,H2,H3,H4,H5,H6,HR,ADDRESS,DL'; // List of tags for these elements
107
108 // Internal, static:
109 var $recPid = 0; // Set this to the pid of the record manipulated by the class.
110 var $elRef = ''; // Element reference [table]:[field], eg. "tt_content:bodytext"
111 var $relPath=''; // Relative path
112 var $relBackPath=''; // Relative back-path
113 var $procOptions = ''; // Set to the TSconfig options coming from Page TSconfig
114
115 // Internal, dynamic
116 var $TS_transform_db_safecounter=100; // Run-away brake for recursive calls.
117 var $rte_p=''; // Parameters from TCA types configuration related to the RTE
118 var $getKeepTags_cache=array(); // Data caching for processing function
119 var $allowedClasses=array(); // Storage of the allowed CSS class names in the RTE
120 var $preserveTags = ''; // Set to tags to preserve from Page TSconfig configuration
121
122
123
124
125
126
127
128
129
130
131 /**
132 * Initialize, setting element reference and record PID
133 *
134 * @param string Element reference, eg "tt_content:bodytext"
135 * @param integer PID of the record (page id)
136 * @return void
137 */
138 function init($elRef='',$recPid=0) {
139 $this->recPid = $recPid;
140 $this->elRef = $elRef;
141 }
142
143 /**
144 * Setting the ->relPath and ->relBackPath to proper values so absolute references to links and images can be converted to relative dittos.
145 * This is used when editing files with the RTE
146 *
147 * @param string The relative path from PATH_site to the place where the file being edited is. Eg. "fileadmin/static".
148 * @return void There is no output, it is set in internal variables. With the above example of "fileadmin/static" as input this will yield ->relPath to be "fileadmin/static/" and ->relBackPath to be "../../"
149 */
150 function setRelPath($path) {
151 $path = trim($path);
152 $path = ereg_replace('^/','',$path);
153 $path = ereg_replace('/$','',$path);
154 if ($path) {
155 $this->relPath = $path;
156 $this->relBackPath = '';
157 $partsC=count(explode('/',$this->relPath));
158 for ($a=0;$a<$partsC;$a++) {
159 $this->relBackPath.='../';
160 }
161 $this->relPath.='/';
162 }
163 }
164
165 /**
166 * Evaluate the environment for editing a staticFileEdit file.
167 * Called for almost all fields being saved in the database. Is called without an instance of the object: t3lib_parsehtml_proc::evalWriteFile()
168 *
169 * @param array Parameters for the current field as found in types-config
170 * @param array Current record we are editing.
171 * @return mixed On success an array with various information is returned, otherwise a string with an error message
172 * @see t3lib_TCEmain, t3lib_transferData
173 */
174 function evalWriteFile($pArr,$currentRecord) {
175
176 // Write file configuration:
177 if (is_array($pArr)) {
178 if ($GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath']
179 && substr($GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath'],-1)=='/'
180 && @is_dir(PATH_site.$GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath'])) {
181
182 $SW_p = $pArr['parameters'];
183 $SW_editFileField = trim($SW_p[0]);
184 $SW_editFile = $currentRecord[$SW_editFileField];
185 if ($SW_editFileField && $SW_editFile && t3lib_div::validPathStr($SW_editFile)) {
186 $SW_relpath = $GLOBALS['TYPO3_CONF_VARS']['BE']['staticFileEditPath'].$SW_editFile;
187 $SW_editFile = PATH_site.$SW_relpath;
188 if (@is_file($SW_editFile)) {
189 return array(
190 'editFile' => $SW_editFile,
191 'relEditFile' => $SW_relpath,
192 'contentField' => trim($SW_p[1]),
193 'markerField' => trim($SW_p[2]),
194 'loadFromFileField' => trim($SW_p[3]),
195 'statusField' => trim($SW_p[4])
196 );
197 } else return "ERROR: Editfile '".$SW_relpath."' did not exist";
198 } else return "ERROR: Edit file name could not be found or was bad.";
199 } else return "ERROR: staticFileEditPath was not set, not set correctly or did not exist!";
200 }
201 }
202
203
204
205
206
207
208
209
210
211
212
213
214
215 /**********************************************
216 *
217 * Main function
218 *
219 **********************************************/
220
221 /**
222 * Transform value for RTE based on specConf in the direction specified by $direction (rte/db)
223 * This is the main function called from tcemain and transfer data classes
224 *
225 * @param string Input value
226 * @param array Special configuration for a field; This is coming from the types-configuration of the field in the TCA. In the types-configuration you can setup features for the field rendering and in particular the RTE takes al its major configuration options from there!
227 * @param string Direction of the transformation. Two keywords are allowed; "db" or "rte". If "db" it means the transformation will clean up content coming from the Rich Text Editor and goes into the database. The other direction, "rte", is of course when content is coming from database and must be transformed to fit the RTE.
228 * @param array Parsed TypoScript content configuring the RTE, probably coming from Page TSconfig.
229 * @return string Output value
230 * @see t3lib_TCEmain::fillInFieldArray(), t3lib_transferData::renderRecord_typesProc()
231 */
232 function RTE_transform($value,$specConf,$direction='rte',$thisConfig=array()) {
233
234 // Init:
235 $this->procOptions = $thisConfig['proc.'];
236 $this->preserveTags = strtoupper(implode(',',t3lib_div::trimExplode(',',$this->procOptions['preserveTags'])));
237
238 // Get parameters for rte_transformation:
239 $p = $this->rte_p = t3lib_BEfunc::getSpecConfParametersFromArray($specConf['rte_transform']['parameters']);
240
241 // Setting modes:
242 if (strcmp($this->procOptions['overruleMode'],'')) {
243 $modes = array_unique(t3lib_div::trimExplode(',',$this->procOptions['overruleMode']));
244 } else {
245 $modes = array_unique(t3lib_div::trimExplode('-',$p['mode']));
246 }
247 $revmodes = array_flip($modes);
248
249 // Find special modes and extract them:
250 if (isset($revmodes['ts'])) {
251 $modes[$revmodes['ts']] = 'ts_transform,ts_preserve,ts_images,ts_links';
252 }
253 // Find special modes and extract them:
254 if (isset($revmodes['ts_css'])) {
255 $modes[$revmodes['ts_css']] = 'css_transform,ts_images,ts_links';
256 }
257
258 // Make list unique
259 $modes = array_unique(t3lib_div::trimExplode(',',implode(',',$modes),1));
260
261 // Reverse order if direction is "rte"
262 if ($direction=='rte') {
263 $modes = array_reverse($modes);
264 }
265
266 // 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:
267 $entry_HTMLparser = $this->procOptions['entryHTMLparser_'.$direction] ? $this->HTMLparserConfig($this->procOptions['entryHTMLparser_'.$direction.'.']) : '';
268 $exit_HTMLparser = $this->procOptions['exitHTMLparser_'.$direction] ? $this->HTMLparserConfig($this->procOptions['exitHTMLparser_'.$direction.'.']) : '';
269
270 // Line breaks of content is unified into char-10 only (removing char 13)
271 if (!$this->procOptions['disableUnifyLineBreaks']) {
272 $value = str_replace(chr(13).chr(10),chr(10),$value);
273 }
274
275 // In an entry-cleaner was configured, pass value through the HTMLcleaner with that:
276 if (is_array($entry_HTMLparser)) {
277 $value = $this->HTMLcleaner($value,$entry_HTMLparser[0],$entry_HTMLparser[1],$entry_HTMLparser[2],$entry_HTMLparser[3]);
278 }
279
280 // Traverse modes:
281 foreach($modes as $cmd) {
282 // ->DB
283 if ($direction=='db') {
284 // Checking for user defined transformation:
285 if ($_classRef = $GLOBALS['TYPO3_CONF_VARS']['SC_OPTIONS']['t3lib/class.t3lib_parsehtml_proc.php']['transformation'][$cmd]) {
286 $_procObj = &t3lib_div::getUserObj($_classRef);
287 $_procObj->pObj = &$this;
288 $_procObj->transformationKey = $cmd;
289 $value = $_procObj->transform_db($value,$this);
290 } else { // ... else use defaults:
291 switch($cmd) {
292 case 'ts_images':
293 $value = $this->TS_images_db($value);
294 break;
295 case 'ts_reglinks':
296 $value = $this->TS_reglinks($value,'db');
297 break;
298 case 'ts_links':
299 $value = $this->TS_links_db($value);
300 break;
301 case 'ts_preserve':
302 $value = $this->TS_preserve_db($value);
303 break;
304 case 'ts_transform':
305 case 'css_transform':
306 $value = str_replace(chr(13),'',$value); // Has a very disturbing effect, so just remove all '13' - depend on '10'
307 $this->allowedClasses = t3lib_div::trimExplode(',',strtoupper($this->procOptions['allowedClasses']),1);
308 $value = $this->TS_transform_db($value,$cmd=='css_transform');
309 break;
310 case 'ts_strip':
311 $value = $this->TS_strip_db($value);
312 break;
313 default:
314 break;
315 }
316 }
317 }
318 // ->RTE
319 if ($direction=='rte') {
320 // Checking for user defined transformation:
321 if ($_classRef = $GLOBALS['TYPO3_CONF_VARS']['SC_OPTIONS']['t3lib/class.t3lib_parsehtml_proc.php']['transformation'][$cmd]) {
322 $_procObj = &t3lib_div::getUserObj($_classRef);
323 $_procObj->pObj = &$this;
324 $value = $_procObj->transform_rte($value,$this);
325 } else { // ... else use defaults:
326 switch($cmd) {
327 case 'ts_images':
328 $value = $this->TS_images_rte($value);
329 break;
330 case 'ts_reglinks':
331 $value = $this->TS_reglinks($value,'rte');
332 break;
333 case 'ts_links':
334 $value = $this->TS_links_rte($value);
335 break;
336 case 'ts_preserve':
337 $value = $this->TS_preserve_rte($value);
338 break;
339 case 'ts_transform':
340 case 'css_transform':
341 $value = str_replace(chr(13),'',$value); // Has a very disturbing effect, so just remove all '13' - depend on '10'
342 $value = $this->TS_transform_rte($value,$cmd=='css_transform');
343 break;
344 default:
345 break;
346 }
347 }
348 }
349 }
350
351 // In an exit-cleaner was configured, pass value through the HTMLcleaner with that:
352 if (is_array($exit_HTMLparser)) {
353 $value = $this->HTMLcleaner($value,$exit_HTMLparser[0],$exit_HTMLparser[1],$exit_HTMLparser[2],$exit_HTMLparser[3]);
354 }
355
356 // Final clean up of linebreaks:
357 if (!$this->procOptions['disableUnifyLineBreaks']) {
358 $value = str_replace(chr(13).chr(10),chr(10),$value); // Make sure no \r\n sequences has entered in the meantime...
359 $value = str_replace(chr(10),chr(13).chr(10),$value); // ... and then change all \n into \r\n
360 }
361
362 // Return value:
363 return $value;
364 }
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381 /************************************
382 *
383 * Specific RTE TRANSFORMATION functions
384 *
385 *************************************/
386
387 /**
388 * Transformation handler: 'ts_images' / direction: "db"
389 * Processing images inserted in the RTE.
390 * This is used when content goes from the RTE to the database.
391 * Images inserted in the RTE has an absolute URL applied to the src attribute. This URL is converted to a relative URL
392 * If it turns out that the URL is from another website than the current the image is read from that external URL and moved to the local server.
393 * Also "magic" images are processed here.
394 *
395 * @param string The content from RTE going to Database
396 * @return string Processed content
397 */
398 function TS_images_db($value) {
399
400 // Split content by <img> tags and traverse the resulting array for processing:
401 $imgSplit = $this->splitTags('img',$value);
402 foreach($imgSplit as $k => $v) {
403 if ($k%2) { // image found, do processing:
404
405 // Init
406 $attribArray = $this->get_tag_attributes_classic($v,1);
407 $siteUrl = $this->siteUrl();
408 $sitePath = str_replace (t3lib_div::getIndpEnv('TYPO3_REQUEST_HOST'), '', $siteUrl);
409
410 $absRef = trim($attribArray['src']); // It's always a absolute URL coming from the RTE into the Database.
411
412 // make path absolute if it is relative and we have a site path wich is not '/'
413 $pI=pathinfo($absRef);
414 if($sitePath AND !$pI['scheme'] && t3lib_div::isFirstPartOfStr($absRef,$sitePath)) {
415 // if site is in a subpath (eg. /~user_jim/) this path needs to be removed because it will be added with $siteUrl
416 $absRef = substr($absRef,strlen($sitePath));
417 $absRef = $siteUrl.$absRef;
418 }
419
420 // External image from another URL? In that case, fetch image (unless disabled feature).
421 if (!t3lib_div::isFirstPartOfStr($absRef,$siteUrl) && !$this->procOptions['dontFetchExtPictures']) {
422 $externalFile = $this->getUrl($absRef); // Get it
423 if ($externalFile) {
424 $pU = parse_url($absRef);
425 $pI=pathinfo($pU['path']);
426
427 if (t3lib_div::inList('gif,png,jpeg,jpg',strtolower($pI['extension']))) {
428 $filename = t3lib_div::shortMD5($absRef).'.'.$pI['extension'];
429 $origFilePath = PATH_site.$this->rteImageStorageDir().'RTEmagicP_'.$filename;
430 $C_origFilePath = PATH_site.$this->rteImageStorageDir().'RTEmagicC_'.$filename.'.'.$pI['extension'];
431 if (!@is_file($origFilePath)) {
432 t3lib_div::writeFile($origFilePath,$externalFile);
433 t3lib_div::writeFile($C_origFilePath,$externalFile);
434 }
435 $absRef = $siteUrl.$this->rteImageStorageDir().'RTEmagicC_'.$filename.'.'.$pI['extension'];
436
437 $attribArray['src']=$absRef;
438 $params = t3lib_div::implodeAttributes($attribArray,1);
439 $imgSplit[$k] = '<img '.$params.' />';
440 }
441 }
442 }
443
444 // Check image as local file (siteURL equals the one of the image)
445 if (t3lib_div::isFirstPartOfStr($absRef,$siteUrl)) {
446 $path = rawurldecode(substr($absRef,strlen($siteUrl))); // Rel-path, rawurldecoded for special characters.
447 $filepath = t3lib_div::getFileAbsFileName($path); // Abs filepath, locked to relative path of this project.
448
449 // Check file existence (in relative dir to this installation!)
450 if ($filepath && @is_file($filepath)) {
451
452 // If "magic image":
453 $pathPre=$this->rteImageStorageDir().'RTEmagicC_';
454 if (t3lib_div::isFirstPartOfStr($path,$pathPre)) {
455 // Find original file:
456 $pI=pathinfo(substr($path,strlen($pathPre)));
457 $filename = substr($pI['basename'],0,-strlen('.'.$pI['extension']));
458 $origFilePath = PATH_site.$this->rteImageStorageDir().'RTEmagicP_'.$filename;
459 if (@is_file($origFilePath)) {
460 $imgObj = t3lib_div::makeInstance('t3lib_stdGraphic');
461 $imgObj->init();
462 $imgObj->mayScaleUp=0;
463 $imgObj->tempPath=PATH_site.$imgObj->tempPath;
464
465 $curInfo = $imgObj->getImageDimensions($filepath); // Image dimensions of the current image
466 $curWH = $this->getWHFromAttribs($attribArray); // Image dimensions as set in the image tag
467 // Compare dimensions:
468 if ($curWH[0]!=$curInfo[0] || $curWH[1]!=$curInfo[1]) {
469 $origImgInfo = $imgObj->getImageDimensions($origFilePath); // Image dimensions of the current image
470 $cW = $curWH[0];
471 $cH = $curWH[1];
472 $cH = 1000; // Make the image based on the width solely...
473 $imgI = $imgObj->imageMagickConvert($origFilePath,$pI['extension'],$cW.'m',$cH.'m');
474 if ($imgI[3]) {
475 $fI=pathinfo($imgI[3]);
476 @copy($imgI[3],$filepath); // Override the child file
477 unset($attribArray['style']);
478 $attribArray['width']=$imgI[0];
479 $attribArray['height']=$imgI[1];
480 if (!$attribArray['border']) $attribArray['border']=0;
481 $params = t3lib_div::implodeAttributes($attribArray,1);
482 $imgSplit[$k]='<img '.$params.' />';
483 }
484 }
485 }
486
487 } elseif ($this->procOptions['plainImageMode']) { // If "plain image" has been configured:
488
489 // Image dimensions as set in the image tag
490 $curWH = $this->getWHFromAttribs($attribArray);
491 $attribArray['width'] = $curWH[0];
492 $attribArray['height'] = $curWH[1];
493
494 // Forcing values for style and border:
495 unset($attribArray['style']);
496 if (!$attribArray['border']) $attribArray['border'] = 0;
497
498 // Finding dimensions of image file:
499 $fI = @getimagesize($filepath);
500
501 // Perform corrections to aspect ratio based on configuration:
502 switch((string)$this->procOptions['plainImageMode']) {
503 case 'lockDimensions':
504 $attribArray['width']=$fI[0];
505 $attribArray['height']=$fI[1];
506 break;
507 case 'lockRatioWhenSmaller': // If the ratio has to be smaller, then first set the width...:
508 if ($attribArray['width']>$fI[0]) $attribArray['width'] = $fI[0];
509 case 'lockRatio':
510 if ($fI[0]>0) {
511 $attribArray['height']=round($attribArray['width']*($fI[1]/$fI[0]));
512 }
513 break;
514 }
515
516 // Compile the image tag again:
517 $params = t3lib_div::implodeAttributes($attribArray,1);
518 $imgSplit[$k]='<img '.$params.' />';
519 }
520 } else { // Remove image if it was not found in a proper position on the server!
521
522 // Commented out; removing the image tag might not be that logical...
523 #$imgSplit[$k]='';
524 }
525 }
526
527 // Convert abs to rel url
528 if ($imgSplit[$k]) {
529 $attribArray=$this->get_tag_attributes_classic($imgSplit[$k],1);
530 $absRef = trim($attribArray['src']);
531 if (t3lib_div::isFirstPartOfStr($absRef,$siteUrl)) {
532 $attribArray['src'] = $this->relBackPath.substr($absRef,strlen($siteUrl));
533 if (!isset($attribArray['alt'])) $attribArray['alt']=''; // Must have alt-attribute for XHTML compliance.
534 $imgSplit[$k]='<img '.t3lib_div::implodeAttributes($attribArray,1,1).' />';
535 }
536 }
537 }
538 }
539 return implode('',$imgSplit);
540 }
541
542 /**
543 * Transformation handler: 'ts_images' / direction: "rte"
544 * Processing images from database content going into the RTE.
545 * Processing includes converting the src attribute to an absolute URL.
546 *
547 * @param string Content input
548 * @return string Content output
549 */
550 function TS_images_rte($value) {
551
552 $siteUrl = $this->siteUrl();
553 $sitePath = str_replace (t3lib_div::getIndpEnv('TYPO3_REQUEST_HOST'), '', $siteUrl);
554
555 // Split content by <img> tags and traverse the resulting array for processing:
556 $imgSplit = $this->splitTags('img',$value);
557 foreach($imgSplit as $k => $v) {
558 if ($k%2) { // image found:
559
560 // Init
561 $attribArray=$this->get_tag_attributes_classic($v,1);
562 $absRef = trim($attribArray['src']);
563
564 // Unless the src attribute is already pointing to an external URL:
565 if (strtolower(substr($absRef,0,4))!='http') {
566 $attribArray['src'] = substr($attribArray['src'],strlen($this->relBackPath));
567 // if site is in a subpath (eg. /~user_jim/) this path needs to be removed because it will be added with $siteUrl
568 $attribArray['src'] = preg_replace('#^'.preg_quote($sitePath,'#').'#','',$attribArray['src']);
569 $attribArray['src'] = $siteUrl.$attribArray['src'];
570 if (!isset($attribArray['alt'])) $attribArray['alt']='';
571 $params = t3lib_div::implodeAttributes($attribArray);
572 $imgSplit[$k]='<img '.$params.' />';
573 }
574 }
575 }
576
577 // return processed content:
578 return implode('',$imgSplit);
579 }
580
581 /**
582 * Transformation handler: 'ts_reglinks' / direction: "db"+"rte" depending on $direction variable.
583 * Converting <A>-tags to/from abs/rel
584 *
585 * @param string Content input
586 * @param string Direction of conversion; "rte" (from database to RTE) or "db" (from RTE to database)
587 * @return string Content output
588 */
589 function TS_reglinks($value,$direction) {
590 $retVal = '';
591
592 switch($direction) {
593 case 'rte':
594 $retVal = $this->TS_AtagToAbs($value,1);
595 break;
596 case 'db':
597 $siteURL = $this->siteUrl();
598 $blockSplit = $this->splitIntoBlock('A',$value);
599 reset($blockSplit);
600 while(list($k,$v)=each($blockSplit)) {
601 if ($k%2) { // block:
602 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v),1);
603 // If the url is local, remove url-prefix
604 if ($siteURL && substr($attribArray['href'],0,strlen($siteURL))==$siteURL) {
605 $attribArray['href']=$this->relBackPath.substr($attribArray['href'],strlen($siteURL));
606 }
607 $bTag='<a '.t3lib_div::implodeAttributes($attribArray,1).'>';
608 $eTag='</a>';
609 $blockSplit[$k] = $bTag.$this->TS_reglinks($this->removeFirstAndLastTag($blockSplit[$k]),$direction).$eTag;
610 }
611 }
612 $retVal = implode('',$blockSplit);
613 break;
614 }
615 return $retVal;
616 }
617
618 /**
619 * Transformation handler: 'ts_links' / direction: "db"
620 * Converting <A>-tags to <link tags>
621 *
622 * @param string Content input
623 * @return string Content output
624 * @see TS_links_rte()
625 */
626 function TS_links_db($value) {
627
628 // Split content into <a> tag blocks and process:
629 $blockSplit = $this->splitIntoBlock('A',$value);
630 foreach($blockSplit as $k => $v) {
631 if ($k%2) { // If an A-tag was found:
632 $attribArray = $this->get_tag_attributes_classic($this->getFirstTag($v),1);
633 $info = $this->urlInfoForLinkTags($attribArray['href']);
634
635 // Check options:
636 $attribArray_copy = $attribArray;
637 unset($attribArray_copy['href']);
638 unset($attribArray_copy['target']);
639 unset($attribArray_copy['class']);
640 unset($attribArray_copy['title']);
641 if ($attribArray_copy['rteerror']) { // Unset "rteerror" and "style" attributes if "rteerror" is set!
642 unset($attribArray_copy['style']);
643 unset($attribArray_copy['rteerror']);
644 }
645 if (!count($attribArray_copy)) { // Only if href, target and class are the only attributes, we can alter the link!
646 // Creating the TYPO3 pseudo-tag "<LINK>" for the link (includes href/url, target and class attributes):
647 $bTag='<link '.$info['url'].($attribArray['target']?' '.$attribArray['target']:(($attribArray['class'] || $attribArray['title'])?' -':'')).($attribArray['class']?' '.$attribArray['class']:($attribArray['title']?' -':'')).($attribArray['title']?' "'.$attribArray['title'].'"':'').'>';
648 $eTag='</link>';
649 $blockSplit[$k] = $bTag.$this->TS_links_db($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
650 } else { // ... otherwise store the link as a-tag.
651 // Unsetting 'rtekeep' attribute if that had been set.
652 unset($attribArray['rtekeep']);
653 // If the url is local, remove url-prefix
654 $siteURL = $this->siteUrl();
655 if ($siteURL && substr($attribArray['href'],0,strlen($siteURL))==$siteURL) {
656 $attribArray['href']=$this->relBackPath.substr($attribArray['href'],strlen($siteURL));
657 }
658 $bTag='<a '.t3lib_div::implodeAttributes($attribArray,1).'>';
659 $eTag='</a>';
660 $blockSplit[$k] = $bTag.$this->TS_links_db($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
661 }
662 }
663 }
664 return implode('',$blockSplit);
665 }
666
667 /**
668 * Transformation handler: 'ts_links' / direction: "rte"
669 * Converting <link tags> to <A>-tags
670 *
671 * @param string Content input
672 * @return string Content output
673 * @see TS_links_rte()
674 */
675 function TS_links_rte($value) {
676 $value = $this->TS_AtagToAbs($value);
677
678 // Split content by the TYPO3 pseudo tag "<link>":
679 $blockSplit = $this->splitIntoBlock('link',$value,1);
680 foreach($blockSplit as $k => $v) {
681 $error = '';
682 if ($k%2) { // block:
683 $tagCode = t3lib_div::unQuoteFilenames(trim(substr($this->getFirstTag($v),0,-1)),true);
684 $link_param = $tagCode[1];
685 $href = '';
686 $siteUrl = $this->siteUrl();
687 // Parsing the typolink data. This parsing is roughly done like in tslib_content->typolink()
688 if(strstr($link_param,'@')) { // mailadr
689 $href = 'mailto:'.eregi_replace('^mailto:','',$link_param);
690 } elseif (substr($link_param,0,1)=='#') { // check if anchor
691 $href = $siteUrl.$link_param;
692 } else {
693 $fileChar=intval(strpos($link_param, '/'));
694 $urlChar=intval(strpos($link_param, '.'));
695
696 // Detects if a file is found in site-root OR is a simulateStaticDocument.
697 list($rootFileDat) = explode('?',$link_param);
698 $rFD_fI = pathinfo($rootFileDat);
699 if (trim($rootFileDat) && !strstr($link_param,'/') && (@is_file(PATH_site.$rootFileDat) || t3lib_div::inList('php,html,htm',strtolower($rFD_fI['extension'])))) {
700 $href = $siteUrl.$link_param;
701 } elseif($urlChar && (strstr($link_param,'//') || !$fileChar || $urlChar<$fileChar)) { // url (external): If doubleSlash or if a '.' comes before a '/'.
702 if (!ereg('^[a-z]*://',trim(strtolower($link_param)))) {$scheme='http://';} else {$scheme='';}
703 $href = $scheme.$link_param;
704 } elseif($fileChar) { // file (internal)
705 $href = $siteUrl.$link_param;
706 } else { // integer or alias (alias is without slashes or periods or commas, that is 'nospace,alphanum_x,lower,unique' according to tables.php!!)
707 $link_params_parts = explode('#',$link_param);
708 $idPart = trim($link_params_parts[0]); // Link-data del
709 if (!strcmp($idPart,'')) { $idPart=$this->recPid; } // If no id or alias is given, set it to class record pid
710
711 // FIXME commented because useless - what is it for?
712 // if ($link_params_parts[1] && !$sectionMark) {
713 // $sectionMark = '#'.trim($link_params_parts[1]);
714 // }
715
716 // Splitting the parameter by ',' and if the array counts more than 1 element it's a id/type/? pair
717 $pairParts = t3lib_div::trimExplode(',',$idPart);
718 if (count($pairParts)>1) {
719 $idPart = $pairParts[0];
720 // Type ? future support for?
721 }
722 // Checking if the id-parameter is an alias.
723 if (!t3lib_div::testInt($idPart)) {
724 list($idPartR) = t3lib_BEfunc::getRecordsByField('pages','alias',$idPart);
725 $idPart = intval($idPartR['uid']);
726 }
727 $page = t3lib_BEfunc::getRecord('pages', $idPart);
728 if (is_array($page)) { // Page must exist...
729 $href = $siteUrl.'?id='.$link_param;
730 } else {
731 #$href = '';
732 $href = $siteUrl.'?id='.$link_param;
733 $error = 'No page found: '.$idPart;
734 }
735 }
736 }
737
738 // Setting the A-tag:
739 $bTag = '<a href="'.htmlspecialchars($href).'"'.
740 ($tagCode[2]&&$tagCode[2]!='-' ? ' target="'.htmlspecialchars($tagCode[2]).'"' : '').
741 ($tagCode[3]&&$tagCode[3]!='-' ? ' class="'.htmlspecialchars($tagCode[3]).'"' : '').
742 ($tagCode[4] ? ' title="'.htmlspecialchars($tagCode[4]).'"' : '').
743 ($error ? ' rteerror="'.htmlspecialchars($error).'" style="background-color: yellow; border:2px red solid; color: black;"' : ''). // Should be OK to add the style; the transformation back to databsae will remove it...
744 '>';
745 $eTag = '</a>';
746 $blockSplit[$k] = $bTag.$this->TS_links_rte($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
747 }
748 }
749
750 // Return content:
751 return implode('',$blockSplit);
752 }
753
754 /**
755 * Preserve special tags
756 *
757 * @param string Content input
758 * @return string Content output
759 */
760 function TS_preserve_db($value) {
761 if (!$this->preserveTags) return $value;
762
763 // Splitting into blocks for processing (span-tags are used for special tags)
764 $blockSplit = $this->splitIntoBlock('span',$value);
765 foreach($blockSplit as $k => $v) {
766 if ($k%2) { // block:
767 $attribArray=$this->get_tag_attributes_classic($this->getFirstTag($v));
768 if ($attribArray['specialtag']) {
769 $theTag = rawurldecode($attribArray['specialtag']);
770 $theTagName = $this->getFirstTagName($theTag);
771 $blockSplit[$k] = $theTag.$this->removeFirstAndLastTag($blockSplit[$k]).'</'.$theTagName.'>';
772 }
773 }
774 }
775 return implode('',$blockSplit);
776 }
777
778 /**
779 * Preserve special tags
780 *
781 * @param string Content input
782 * @return string Content output
783 */
784 function TS_preserve_rte($value) {
785 if (!$this->preserveTags) return $value;
786
787 $blockSplit = $this->splitIntoBlock($this->preserveTags,$value);
788 foreach($blockSplit as $k => $v) {
789 if ($k%2) { // block:
790 $blockSplit[$k] = '<span specialtag="'.rawurlencode($this->getFirstTag($v)).'">'.$this->removeFirstAndLastTag($blockSplit[$k]).'</span>';
791 }
792 }
793 return implode('',$blockSplit);
794 }
795
796 /**
797 * Transformation handler: 'ts_transform' + 'css_transform' / direction: "db"
798 * Cleaning (->db) for standard content elements (ts)
799 *
800 * @param string Content input
801 * @param boolean If true, the transformation was "css_transform", otherwise "ts_transform"
802 * @return string Content output
803 * @see TS_transform_rte()
804 */
805 function TS_transform_db($value,$css=FALSE) {
806
807 // safety... so forever loops are avoided (they should not occur, but an error would potentially do this...)
808 $this->TS_transform_db_safecounter--;
809 if ($this->TS_transform_db_safecounter<0) return $value;
810
811 // Split the content from RTE by the occurence of these blocks:
812 $blockSplit = $this->splitIntoBlock('TABLE,BLOCKQUOTE,'.$this->headListTags,$value);
813
814 $cc=0;
815 $aC = count($blockSplit);
816
817 // Avoid superfluous linebreaks by transform_db after ending headListTag
818 while($aC && !strcmp(trim($blockSplit[$aC-1]),'')) {
819 unset($blockSplit[$aC-1]);
820 $aC = count($blockSplit);
821 }
822
823 // Traverse the blocks
824 foreach($blockSplit as $k => $v) {
825 $cc++;
826 $lastBR = $cc==$aC ? '' : chr(10);
827
828 if ($k%2) { // Inside block:
829
830 // Init:
831 $tag=$this->getFirstTag($v);
832 $tagName=strtolower($this->getFirstTagName($v));
833
834 // Process based on the tag:
835 switch($tagName) {
836 case 'blockquote': // Keep blockquotes, but clean the inside recursively in the same manner as the main code
837 $blockSplit[$k]='<'.$tagName.'>'.$this->TS_transform_db($this->removeFirstAndLastTag($blockSplit[$k]),$css).'</'.$tagName.'>'.$lastBR;
838 break;
839 case 'ol':
840 case 'ul': // Transform lists into <typolist>-tags:
841 if (!$css) {
842 if (!isset($this->procOptions['typolist']) || $this->procOptions['typolist']) {
843 $parts = $this->getAllParts($this->splitIntoBlock('LI',$this->removeFirstAndLastTag($blockSplit[$k])),1,0);
844 while(list($k2)=each($parts)) {
845 $parts[$k2]=preg_replace('/['.preg_quote(chr(10).chr(13)).']+/','',$parts[$k2]); // remove all linesbreaks!
846 $parts[$k2]=$this->defaultTStagMapping($parts[$k2],'db');
847 $parts[$k2]=$this->cleanFontTags($parts[$k2],0,0,0);
848 $parts[$k2] = $this->HTMLcleaner_db($parts[$k2],strtolower($this->procOptions['allowTagsInTypolists']?$this->procOptions['allowTagsInTypolists']:'br,font,b,i,u,a,img,span,strong,em'));
849 }
850 if ($tagName=='ol') { $params=' type="1"'; } else { $params=''; }
851 $blockSplit[$k]='<typolist'.$params.'>'.chr(10).implode(chr(10),$parts).chr(10).'</typolist>'.$lastBR;
852 }
853 } else {
854 $blockSplit[$k]=preg_replace('/['.preg_quote(chr(10).chr(13)).']+/',' ',$this->transformStyledATags($blockSplit[$k])).$lastBR;
855 }
856 break;
857 case 'table': // Tables are NOT allowed in any form (unless preserveTables is set or CSS is the mode)
858 if (!$this->procOptions['preserveTables'] && !$css) {
859 $blockSplit[$k]=$this->TS_transform_db($this->removeTables($blockSplit[$k]));
860 } else {
861 $blockSplit[$k]=preg_replace('/['.preg_quote(chr(10).chr(13)).']+/',' ',$this->transformStyledATags($blockSplit[$k])).$lastBR;
862 }
863 break;
864 case 'h1':
865 case 'h2':
866 case 'h3':
867 case 'h4':
868 case 'h5':
869 case 'h6':
870 if (!$css) {
871 $attribArray=$this->get_tag_attributes_classic($tag);
872 // Processing inner content here:
873 $innerContent = $this->HTMLcleaner_db($this->removeFirstAndLastTag($blockSplit[$k]));
874
875 if (!isset($this->procOptions['typohead']) || $this->procOptions['typohead']) {
876 $type = intval(substr($tagName,1));
877 $blockSplit[$k]='<typohead'.
878 ($type!=6?' type="'.$type.'"':'').
879 ($attribArray['align']?' align="'.$attribArray['align'].'"':'').
880 ($attribArray['class']?' class="'.$attribArray['class'].'"':'').
881 '>'.
882 $innerContent.
883 '</typohead>'.
884 $lastBR;
885 } else {
886 $blockSplit[$k]='<'.$tagName.
887 ($attribArray['align']?' align="'.htmlspecialchars($attribArray['align']).'"':'').
888 ($attribArray['class']?' class="'.htmlspecialchars($attribArray['class']).'"':'').
889 '>'.
890 $innerContent.
891 '</'.$tagName.'>'.
892 $lastBR;
893 }
894 } else {
895 // Eliminate true linebreaks inside Hx tags
896 $blockSplit[$k]=preg_replace('/['.preg_quote(chr(10).chr(13)).']+/',' ',$this->transformStyledATags($blockSplit[$k])).$lastBR;
897 }
898 break;
899 default:
900 // Eliminate true linebreaks inside other headlist tags and after hr tag
901 $blockSplit[$k]=preg_replace('/['.preg_quote(chr(10).chr(13)).']+/',' ',$this->transformStyledATags($blockSplit[$k])).$lastBR;
902 break;
903 }
904 } else { // NON-block:
905 if (strcmp(trim($blockSplit[$k]),'')) {
906 $blockSplit[$k]=$this->divideIntoLines(preg_replace('/['.preg_quote(chr(10).chr(13)).']+/',' ',$blockSplit[$k])).$lastBR;
907 $blockSplit[$k]=$this->transformStyledATags($blockSplit[$k]);
908 } else unset($blockSplit[$k]);
909 }
910 }
911 $this->TS_transform_db_safecounter++;
912
913 return implode('',$blockSplit);
914 }
915
916 /**
917 * Wraps a-tags that contain a style attribute with a span-tag
918 *
919 * @param string Content input
920 * @return string Content output
921 */
922 function transformStyledATags($value) {
923 $blockSplit = $this->splitIntoBlock('A',$value);
924 foreach($blockSplit as $k => $v) {
925 if ($k%2) { // If an A-tag was found:
926 $attribArray = $this->get_tag_attributes_classic($this->getFirstTag($v),1);
927 if ($attribArray['style']) { // If "style" attribute is set!
928 $attribArray_copy['style'] = $attribArray['style'];
929 unset($attribArray['style']);
930 $bTag='<span '.t3lib_div::implodeAttributes($attribArray_copy,1).'><a '.t3lib_div::implodeAttributes($attribArray,1).'>';
931 $eTag='</a></span>';
932 $blockSplit[$k] = $bTag.$this->removeFirstAndLastTag($blockSplit[$k]).$eTag;
933 }
934 }
935 }
936 return implode('',$blockSplit);
937 }
938
939 /**
940 * Transformation handler: 'ts_transform' + 'css_transform' / direction: "rte"
941 * Set (->rte) for standard content elements (ts)
942 *
943 * @param string Content input
944 * @param boolean If true, the transformation was "css_transform", otherwise "ts_transform"
945 * @return string Content output
946 * @see TS_transform_db()
947 */
948 function TS_transform_rte($value,$css=0) {
949
950 // Split the content from Database by the occurence of these blocks:
951 $blockSplit = $this->splitIntoBlock('TABLE,BLOCKQUOTE,TYPOLIST,TYPOHEAD,'.$this->headListTags,$value);
952
953 // Traverse the blocks
954 foreach($blockSplit as $k => $v) {
955 if ($k%2) { // Inside one of the blocks:
956
957 // Init:
958 $tag = $this->getFirstTag($v);
959 $tagName = strtolower($this->getFirstTagName($v));
960 $attribArray = $this->get_tag_attributes_classic($tag);
961
962 // Based on tagname, we do transformations:
963 switch($tagName) {
964 case 'blockquote': // Keep blockquotes:
965 $blockSplit[$k] = $tag.
966 $this->TS_transform_rte($this->removeFirstAndLastTag($blockSplit[$k]),$css).
967 '</'.$tagName.'>';
968 break;
969 case 'typolist': // Transform typolist blocks into OL/UL lists. Type 1 is expected to be numerical block
970 if (!isset($this->procOptions['typolist']) || $this->procOptions['typolist']) {
971 $tListContent = $this->removeFirstAndLastTag($blockSplit[$k]);
972 $tListContent = ereg_replace('^[ ]*'.chr(10),'',$tListContent);
973 $tListContent = ereg_replace(chr(10).'[ ]*$','',$tListContent);
974 $lines = explode(chr(10),$tListContent);
975 $typ = $attribArray['type']==1 ? 'ol' : 'ul';
976 $blockSplit[$k] = '<'.$typ.'>'.chr(10).
977 '<li>'.implode('</li>'.chr(10).'<li>',$lines).'</li>'.
978 '</'.$typ.'>';
979 }
980 break;
981 case 'typohead': // Transform typohead into Hx tags.
982 if (!isset($this->procOptions['typohead']) || $this->procOptions['typohead']) {
983 $tC = $this->removeFirstAndLastTag($blockSplit[$k]);
984 $typ = t3lib_div::intInRange($attribArray['type'],0,6);
985 if (!$typ) $typ=6;
986 $align = $attribArray['align']?' align="'.$attribArray['align'].'"': '';
987 $class = $attribArray['class']?' class="'.$attribArray['class'].'"': '';
988 $blockSplit[$k] = '<h'.$typ.$align.$class.'>'.
989 $tC.
990 '</h'.$typ.'>';
991 }
992 break;
993 }
994 $blockSplit[$k+1] = ereg_replace('^[ ]*'.chr(10),'',$blockSplit[$k+1]); // Removing linebreak if typohead
995 } else { // NON-block:
996 $nextFTN = $this->getFirstTagName($blockSplit[$k+1]);
997 $singleLineBreak = $blockSplit[$k]==chr(10);
998 if (t3lib_div::inList('TABLE,BLOCKQUOTE,TYPOLIST,TYPOHEAD,'.$this->headListTags,$nextFTN)) { // Removing linebreak if typolist/typohead
999 $blockSplit[$k] = ereg_replace(chr(10).'[ ]*$','',$blockSplit[$k]);
1000 }
1001 // If $blockSplit[$k] is blank then unset the line. UNLESS the line happend to be a single line break.
1002 if (!strcmp($blockSplit[$k],'') && !$singleLineBreak) {
1003 unset($blockSplit[$k]);
1004 } else {
1005 $blockSplit[$k] = $this->setDivTags($blockSplit[$k],($this->procOptions['useDIVasParagraphTagForRTE']?'div':'p'));
1006 }
1007 }
1008 }
1009 return implode(chr(10),$blockSplit);
1010 }
1011
1012 /**
1013 * Transformation handler: 'ts_strip' / direction: "db"
1014 * Removing all non-allowed tags
1015 *
1016 * @param string Content input
1017 * @return string Content output
1018 */
1019 function TS_strip_db($value) {
1020 $value = strip_tags($value,'<'.implode('><',explode(',','b,i,u,a,img,br,div,center,pre,font,hr,sub,sup,p,strong,em,li,ul,ol,blockquote')).'>');
1021 return $value;
1022 }
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037 /***************************************************************
1038 *
1039 * Generic RTE transformation, analysis and helper functions
1040 *
1041 **************************************************************/
1042
1043 /**
1044 * Reads the file or url $url and returns the content
1045 *
1046 * @param string Filepath/URL to read
1047 * @return string The content from the resource given as input.
1048 * @see t3lib_div::getURL()
1049 */
1050 function getURL($url) {
1051 return t3lib_div::getURL($url);
1052 }
1053
1054 /**
1055 * Function for cleaning content going into the database.
1056 * Content is cleaned eg. by removing unallowed HTML and ds-HSC content
1057 * It is basically calling HTMLcleaner from the parent class with some preset configuration specifically set up for cleaning content going from the RTE into the db
1058 *
1059 * @param string Content to clean up
1060 * @param string Comma list of tags to specifically allow. Default comes from getKeepTags and is ""
1061 * @return string Clean content
1062 * @see getKeepTags()
1063 */
1064 function HTMLcleaner_db($content,$tagList='') {
1065 if (!$tagList) {
1066 $keepTags = $this->getKeepTags('db');
1067 } else {
1068 $keepTags = $this->getKeepTags('db',$tagList);
1069 }
1070 $kUknown = $this->procOptions['dontRemoveUnknownTags_db'] ? 1 : 0; // Default: remove unknown tags.
1071 $hSC = $this->procOptions['dontUndoHSC_db'] ? 0 : -1; // Default: re-convert literals to characters (that is &lt; to <)
1072
1073 // Create additional configuration in order to honor the setting RTE.default.proc.HTMLparser_db.xhtml_cleaning=1
1074 $addConfig=array();
1075 if ((is_array($this->procOptions['HTMLparser_db.']) && $this->procOptions['HTMLparser_db.']['xhtml_cleaning']) || (is_array($this->procOptions['entryHTMLparser_db.']) && $this->procOptions['entryHTMLparser_db.']['xhtml_cleaning']) || (is_array($this->procOptions['exitHTMLparser_db.']) && $this->procOptions['exitHTMLparser_db.']['xhtml_cleaning'])) {
1076 $addConfig['xhtml']=1;
1077 }
1078
1079 return $this->HTMLcleaner($content,$keepTags,$kUknown,$hSC,$addConfig);
1080 }
1081
1082 /**
1083 * Creates an array of configuration for the HTMLcleaner function based on whether content go TO or FROM the Rich Text Editor ($direction)
1084 * Unless "tagList" is given, the function will cache the configuration for next time processing goes on. (In this class that is the case only if we are processing a bulletlist)
1085 *
1086 * @param string The direction of the content being processed by the output configuration; "db" (content going into the database FROM the rte) or "rte" (content going into the form)
1087 * @param string Comma list of tags to keep (overriding default which is to keep all + take notice of internal configuration)
1088 * @return array Configuration array
1089 * @see HTMLcleaner_db()
1090 */
1091 function getKeepTags($direction='rte',$tagList='') {
1092 if (!is_array($this->getKeepTags_cache[$direction]) || $tagList) {
1093
1094 // Setting up allowed tags:
1095 if (strcmp($tagList,'')) { // If the $tagList input var is set, this will take precedence
1096 $keepTags = array_flip(t3lib_div::trimExplode(',',$tagList,1));
1097 } else { // Default is to get allowed/denied tags from internal array of processing options:
1098 // Construct default list of tags to keep:
1099 $typoScript_list = 'b,i,u,a,img,br,div,center,pre,font,hr,sub,sup,p,strong,em,li,ul,ol,blockquote,strike,span';
1100 $keepTags = array_flip(t3lib_div::trimExplode(',',$typoScript_list.','.strtolower($this->procOptions['allowTags']),1));
1101
1102 // For tags to deny, remove them from $keepTags array:
1103 $denyTags = t3lib_div::trimExplode(',',$this->procOptions['denyTags'],1);
1104 foreach($denyTags as $dKe) {
1105 unset($keepTags[$dKe]);
1106 }
1107 }
1108
1109 // Based on the direction of content, set further options:
1110 switch ($direction) {
1111
1112 // GOING from database to Rich Text Editor:
1113 case 'rte':
1114 // Transform bold/italics tags to strong/em
1115 if (isset($keepTags['b'])) {$keepTags['b']=array('remap'=>'STRONG');}
1116 if (isset($keepTags['i'])) {$keepTags['i']=array('remap'=>'EM');}
1117
1118 // 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.
1119 list($keepTags) = $this->HTMLparserConfig($this->procOptions['HTMLparser_rte.'],$keepTags);
1120 break;
1121
1122 // GOING from RTE to database:
1123 case 'db':
1124 // Transform strong/em back to bold/italics:
1125 if (isset($keepTags['strong'])) { $keepTags['strong']=array('remap'=>'b'); }
1126 if (isset($keepTags['em'])) { $keepTags['em']=array('remap'=>'i'); }
1127
1128 // Setting up span tags if they are allowed:
1129 if (isset($keepTags['span'])) {
1130 $classes=array_merge(array(''),$this->allowedClasses);
1131 $keepTags['span']=array(
1132 'allowedAttribs' => 'class,style,xml:lang',
1133 'fixAttrib' => Array(
1134 'class' => Array (
1135 'list' => $classes,
1136 'removeIfFalse' => 1
1137 )
1138 ),
1139 'rmTagIfNoAttrib' => 1
1140 );
1141 if (!$this->procOptions['allowedClasses']) unset($keepTags['span']['fixAttrib']['class']['list']);
1142 }
1143
1144 // Setting up font tags if they are allowed:
1145 if (isset($keepTags['font'])) {
1146 $colors=array_merge(array(''),t3lib_div::trimExplode(',',$this->procOptions['allowedFontColors'],1));
1147 $keepTags['font']=array(
1148 'allowedAttribs'=>'face,color,size',
1149 'fixAttrib' => Array(
1150 'face' => Array (
1151 'removeIfFalse' => 1
1152 ),
1153 'color' => Array (
1154 'removeIfFalse' => 1,
1155 'list'=>$colors
1156 ),
1157 'size' => Array (
1158 'removeIfFalse' => 1,
1159 )
1160 ),
1161 'rmTagIfNoAttrib' => 1
1162 );
1163 if (!$this->procOptions['allowedFontColors']) unset($keepTags['font']['fixAttrib']['color']['list']);
1164 }
1165
1166 // Setting further options, getting them from the processiong options:
1167 $TSc = $this->procOptions['HTMLparser_db.'];
1168 if (!$TSc['globalNesting']) $TSc['globalNesting']='b,i,u,a,center,font,sub,sup,strong,em,strike,span';
1169 if (!$TSc['noAttrib']) $TSc['noAttrib']='b,i,u,br,center,hr,sub,sup,strong,em,li,ul,ol,blockquote,strike';
1170
1171 // Transforming the array from TypoScript to regular array:
1172 list($keepTags) = $this->HTMLparserConfig($TSc,$keepTags);
1173 break;
1174 }
1175
1176 // Caching (internally, in object memory) the result unless tagList is set:
1177 if (!$tagList) {
1178 $this->getKeepTags_cache[$direction] = $keepTags;
1179 } else {
1180 return $keepTags;
1181 }
1182 }
1183
1184 // Return result:
1185 return $this->getKeepTags_cache[$direction];
1186 }
1187
1188 /**
1189 * This resolves the $value into parts based on <div></div>-sections and <p>-sections and <br />-tags. These are returned as lines separated by chr(10).
1190 * This point is to resolve the HTML-code returned from RTE into ordinary lines so it's 'human-readable'
1191 * The function ->setDivTags does the opposite.
1192 * This function processes content to go into the database.
1193 *
1194 * @param string Value to process.
1195 * @param integer Recursion brake. Decremented on each recursion down to zero. Default is 5 (which equals the allowed nesting levels of p/div tags).
1196 * @param boolean If true, an array with the lines is returned, otherwise a string of the processed input value.
1197 * @return string Processed input value.
1198 * @see setDivTags()
1199 */
1200 function divideIntoLines($value,$count=5,$returnArray=FALSE) {
1201
1202 // Internalize font tags (move them from OUTSIDE p/div to inside it that is the case):
1203 if ($this->procOptions['internalizeFontTags']) {$value = $this->internalizeFontTags($value);}
1204
1205 // Setting configuration for processing:
1206 $allowTagsOutside = t3lib_div::trimExplode(',',strtolower($this->procOptions['allowTagsOutside']?$this->procOptions['allowTagsOutside']:'img'),1);
1207 $remapParagraphTag = strtoupper($this->procOptions['remapParagraphTag']);
1208 $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...?
1209
1210 if ($this->procOptions['keepPDIVattribs']) {
1211 $keepAttribListArr = t3lib_div::trimExplode(',',strtolower($this->procOptions['keepPDIVattribs']),1);
1212 } else {
1213 $keepAttribListArr = array();
1214 }
1215
1216 // Returns plainly the value if there was no div/p sections in it
1217 if (count($divSplit)<=1 || $count<=0) {
1218 return $value;
1219 }
1220
1221 // Traverse the splitted sections:
1222 foreach($divSplit as $k => $v) {
1223 if ($k%2) { // Inside
1224 $v=$this->removeFirstAndLastTag($v);
1225
1226 // Fetching 'sub-lines' - which will explode any further p/div nesting...
1227 $subLines = $this->divideIntoLines($v,$count-1,1);
1228 if (is_array($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')
1229 // No noting.
1230 } else { //... but if NO subsection was found, we process it as a TRUE line without erronous content:
1231 $subLines = array($subLines);
1232 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...
1233 $subLines = spliti('<br[[:space:]]*[\/]?>',$v);
1234 }
1235
1236 // Traverse sublines (there is typically one, except if <br/> has been converted to lines as well!)
1237 reset($subLines);
1238 while(list($sk)=each($subLines)) {
1239
1240 // Clear up the subline for DB.
1241 $subLines[$sk]=$this->HTMLcleaner_db($subLines[$sk]);
1242
1243 // Get first tag, attributes etc:
1244 $fTag = $this->getFirstTag($divSplit[$k]);
1245 $tagName=strtolower($this->getFirstTagName($divSplit[$k]));
1246 $attribs=$this->get_tag_attributes($fTag);
1247
1248 // Keep attributes (lowercase)
1249 $newAttribs=array();
1250 if (count($keepAttribListArr)) {
1251 foreach($keepAttribListArr as $keepA) {
1252 if (isset($attribs[0][$keepA])) { $newAttribs[$keepA] = $attribs[0][$keepA]; }
1253 }
1254 }
1255
1256 // ALIGN attribute:
1257 if (!$this->procOptions['skipAlign'] && strcmp(trim($attribs[0]['align']),'') && strtolower($attribs[0]['align'])!='left') { // Set to value, but not 'left'
1258 $newAttribs['align']=strtolower($attribs[0]['align']);
1259 }
1260
1261 // CLASS attribute:
1262 if (!$this->procOptions['skipClass'] && strcmp(trim($attribs[0]['class']),'')) { // Set to whatever value
1263 if (!count($this->allowedClasses) || in_array(strtoupper($attribs[0]['class']),$this->allowedClasses)) {
1264 $newAttribs['class']=$attribs[0]['class'];
1265 }
1266 }
1267
1268 // Remove any line break char (10 or 13)
1269 $subLines[$sk]=ereg_replace(chr(10).'|'.chr(13),'',$subLines[$sk]);
1270
1271 // If there are any attributes or if we are supposed to remap the tag, then do so:
1272 if (count($newAttribs) && strcmp($remapParagraphTag,'1')) {
1273 if ($remapParagraphTag=='P') $tagName='p';
1274 if ($remapParagraphTag=='DIV') $tagName='div';
1275 $subLines[$sk]='<'.trim($tagName.' '.$this->compileTagAttribs($newAttribs)).'>'.$subLines[$sk].'</'.$tagName.'>';
1276 }
1277 }
1278 }
1279 // Add the processed line(s)
1280 $divSplit[$k] = implode(chr(10),$subLines);
1281
1282 // If it turns out the line is just blank (containing a &nbsp; possibly) then just make it pure blank:
1283 if (trim(strip_tags($divSplit[$k]))=='&nbsp;' && !preg_match('/\<(img)(\s[^>]*)?\/?>/si', $divSplit[$k])) {
1284 $divSplit[$k]='';
1285 }
1286 } else { // outside div:
1287 // Remove positions which are outside div/p tags and without content
1288 $divSplit[$k]=trim(strip_tags($divSplit[$k],'<'.implode('><',$allowTagsOutside).'>'));
1289 if (!strcmp($divSplit[$k],'')) unset($divSplit[$k]); // Remove part if it's empty
1290 }
1291 }
1292
1293 // Return value:
1294 return $returnArray ? $divSplit : implode(chr(10),$divSplit);
1295 }
1296
1297 /**
1298 * Converts all lines into <div></div>/<p></p>-sections (unless the line is a div-section already)
1299 * For processing of content going FROM database TO RTE.
1300 *
1301 * @param string Value to convert
1302 * @param string Tag to wrap with. Either "p" or "div" should it be. Lowercase preferably.
1303 * @return string Processed value.
1304 * @see divideIntoLines()
1305 */
1306 function setDivTags($value,$dT='p') {
1307
1308 // First, setting configuration for the HTMLcleaner function. This will process each line between the <div>/<p> section on their way to the RTE
1309 $keepTags = $this->getKeepTags('rte');
1310 $kUknown = $this->procOptions['dontProtectUnknownTags_rte'] ? 0 : 'protect'; // Default: remove unknown tags.
1311 $hSC = $this->procOptions['dontHSC_rte'] ? 0 : 1; // Default: re-convert literals to characters (that is &lt; to <)
1312 $convNBSP = !$this->procOptions['dontConvAmpInNBSP_rte']?1:0;
1313
1314 // Divide the content into lines, based on chr(10):
1315 $parts = explode(chr(10),$value);
1316 foreach($parts as $k => $v) {
1317
1318 // Processing of line content:
1319 if (!strcmp(trim($parts[$k]),'')) { // If the line is blank, set it to &nbsp;
1320 $parts[$k]='&nbsp;';
1321 } else { // Clean the line content:
1322 $parts[$k]=$this->HTMLcleaner($parts[$k],$keepTags,$kUknown,$hSC);
1323 if ($convNBSP) $parts[$k]=str_replace('&amp;nbsp;','&nbsp;',$parts[$k]);
1324 }
1325
1326 // Wrapping the line in <$dT> is not already wrapped:
1327 $testStr = strtolower(trim($parts[$k]));
1328 if (substr($testStr,0,4)!='<div' || substr($testStr,-6)!='</div>') {
1329 if (substr($testStr,0,2)!='<p' || substr($testStr,-4)!='</p>') {
1330 // Only set p-tags if there is not already div or p tags:
1331 $parts[$k]='<'.$dT.'>'.$parts[$k].'</'.$dT.'>';
1332 }
1333 }
1334 }
1335
1336 // Implode result:
1337 return implode(chr(10),$parts);
1338 }
1339
1340 /**
1341 * This splits the $value in font-tag chunks.
1342 * If there are any <P>/<DIV> sections inside of them, the font-tag is wrapped AROUND the content INSIDE of the P/DIV sections and the outer font-tag is removed.
1343 * This functions seems to be a good choice for pre-processing content if it has been pasted into the RTE from eg. star-office.
1344 * In that case the font-tags are normally on the OUTSIDE of the sections.
1345 * This function is used by eg. divideIntoLines() if the procesing option 'internalizeFontTags' is set.
1346 *
1347 * @param string Input content
1348 * @return string Output content
1349 * @see divideIntoLines()
1350 */
1351 function internalizeFontTags($value) {
1352
1353 // Splitting into font tag blocks:
1354 $fontSplit = $this->splitIntoBlock('font',$value);
1355
1356 foreach($fontSplit as $k => $v) {
1357 if ($k%2) { // Inside
1358 $fTag = $this->getFirstTag($v); // Fint font-tag
1359
1360 $divSplit_sub = $this->splitIntoBlock('div,p',$this->removeFirstAndLastTag($v),1);
1361 if (count($divSplit_sub)>1) { // If there were div/p sections inside the font-tag, do something about it...
1362 // traverse those sections:
1363 foreach($divSplit_sub as $k2 => $v2) {
1364 if ($k2%2) { // Inside
1365 $div_p = $this->getFirstTag($v2); // Fint font-tag
1366 $div_p_tagname = $this->getFirstTagName($v2); // Fint font-tag
1367 $v2=$this->removeFirstAndLastTag($v2); // ... and remove it from original.
1368 $divSplit_sub[$k2]=$div_p.$fTag.$v2.'</font>'.'</'.$div_p_tagname.'>';
1369 } elseif (trim(strip_tags($v2))) {
1370 $divSplit_sub[$k2]=$fTag.$v2.'</font>';
1371 }
1372 }
1373 $fontSplit[$k]=implode('',$divSplit_sub);
1374 }
1375 }
1376 }
1377
1378 return implode('',$fontSplit);
1379 }
1380
1381 /**
1382 * Returns SiteURL based on thisScript.
1383 *
1384 * @return string Value of t3lib_div::getIndpEnv('TYPO3_SITE_URL');
1385 * @see t3lib_div::getIndpEnv()
1386 */
1387 function siteUrl() {
1388 return t3lib_div::getIndpEnv('TYPO3_SITE_URL');
1389 }
1390
1391 /**
1392 * Return the storage folder of RTE image files.
1393 * Default is $GLOBALS['TYPO3_CONF_VARS']['BE']['RTE_imageStorageDir'] unless something else is configured in the types configuration for the RTE.
1394 *
1395 * @return string
1396 */
1397 function rteImageStorageDir() {
1398 return $this->rte_p['imgpath'] ? $this->rte_p['imgpath'] : $GLOBALS['TYPO3_CONF_VARS']['BE']['RTE_imageStorageDir'];
1399 }
1400
1401 /**
1402 * Remove all tables from incoming code
1403 * The function is trying to to this is some more or less respectfull way. The approach is to resolve each table cells content and implode it all by <br /> chars. Thus at least the content is preserved in some way.
1404 *
1405 * @param string Input value
1406 * @param string Break character to use for linebreaks.
1407 * @return string Output value
1408 */
1409 function removeTables($value,$breakChar='<br />') {
1410
1411 // Splitting value into table blocks:
1412 $tableSplit = $this->splitIntoBlock('table',$value);
1413
1414 // Traverse blocks of tables:
1415 foreach($tableSplit as $k => $v) {
1416 if ($k%2) {
1417 $tableSplit[$k]='';
1418 $rowSplit = $this->splitIntoBlock('tr',$v);
1419 foreach($rowSplit as $k2 => $v2) {
1420 if ($k2%2) {
1421 $cellSplit = $this->getAllParts($this->splitIntoBlock('td',$v2),1,0);
1422 foreach($cellSplit as $k3 => $v3) {
1423 $tableSplit[$k].=$v3.$breakChar;
1424 }
1425 }
1426 }
1427 }
1428 }
1429
1430 // Implode it all again:
1431 return implode($breakChar,$tableSplit);
1432 }
1433
1434 /**
1435 * Default tag mapping for TS
1436 *
1437 * @param string Input code to process
1438 * @param string Direction To databsae (db) or from database to RTE (rte)
1439 * @return string Processed value
1440 */
1441 function defaultTStagMapping($code,$direction='rte') {
1442 if ($direction=='db') {
1443 $code=$this->mapTags($code,array( // Map tags
1444 'strong' => 'b',
1445 'em' => 'i'
1446 ));
1447 }
1448 if ($direction=='rte') {
1449 $code=$this->mapTags($code,array( // Map tags
1450 'b' => 'strong',
1451 'i' => 'em'
1452 ));
1453 }
1454 return $code;
1455 }
1456
1457 /**
1458 * Finds width and height from attrib-array
1459 * If the width and height is found in the style-attribute, use that!
1460 *
1461 * @param array Array of attributes from tag in which to search. More specifically the content of the key "style" is used to extract "width:xxx / height:xxx" information
1462 * @return array Integer w/h in key 0/1. Zero is returned if not found.
1463 */
1464 function getWHFromAttribs($attribArray) {
1465 $style =trim($attribArray['style']);
1466 if ($style) {
1467 $regex='[[:space:]]*:[[:space:]]*([0-9]*)[[:space:]]*px';
1468 // Width
1469 $reg = array();
1470 eregi('width'.$regex,$style,$reg);
1471 $w = intval($reg[1]);
1472 // Height
1473 eregi('height'.$regex,$style,$reg);
1474 $h = intval($reg[1]);
1475 }
1476 if (!$w) {
1477 $w = $attribArray['width'];
1478 }
1479 if (!$h) {
1480 $h = $attribArray['height'];
1481 }
1482 return array(intval($w),intval($h));
1483 }
1484
1485 /**
1486 * Parse <A>-tag href and return status of email,external,file or page
1487 *
1488 * @param string URL to analyse.
1489 * @return array Information in an array about the URL
1490 */
1491 function urlInfoForLinkTags($url) {
1492 $info = array();
1493 $url = trim($url);
1494 if (substr(strtolower($url),0,7)=='mailto:') {
1495 $info['url']=trim(substr($url,7));
1496 $info['type']='email';
1497 } else {
1498 $curURL = $this->siteUrl(); // 100502, removed this: 'http://'.t3lib_div::getThisUrl(); Reason: The url returned had typo3/ in the end - should be only the site's url as far as I see...
1499 for($a=0;$a<strlen($url);$a++) {
1500 if ($url{$a}!=$curURL{$a}) {
1501 break;
1502 }
1503 }
1504
1505 $info['relScriptPath']=substr($curURL,$a);
1506 $info['relUrl']=substr($url,$a);
1507 $info['url']=$url;
1508 $info['type']='ext';
1509
1510 $siteUrl_parts = parse_url($url);
1511 $curUrl_parts = parse_url($curURL);
1512
1513 if ($siteUrl_parts['host']==$curUrl_parts['host'] // Hosts should match
1514 && (!$info['relScriptPath'] || (defined('TYPO3_mainDir') && substr($info['relScriptPath'],0,strlen(TYPO3_mainDir))==TYPO3_mainDir))) { // If the script path seems to match or is empty (FE-EDIT)
1515
1516 // New processing order 100502
1517 $uP=parse_url($info['relUrl']);
1518
1519 if (!strcmp('#'.$siteUrl_parts['fragment'],$info['relUrl'])) {
1520 $info['url']=$info['relUrl'];
1521 $info['type']='anchor';
1522 } elseif (!trim($uP['path']) || !strcmp($uP['path'],'index.php')) {
1523 $pp = explode('id=',$uP['query']);
1524 $id = trim($pp[1]);
1525 if ($id) {
1526 $info['pageid']=$id;
1527 $info['cElement']=$uP['fragment'];
1528 $info['url']=$id.($info['cElement']?'#'.$info['cElement']:'');
1529 $info['type']='page';
1530 }
1531 } else {
1532 $info['url']=$info['relUrl'];
1533 $info['type']='file';
1534 }
1535 } else {
1536 unset($info['relScriptPath']);
1537 unset($info['relUrl']);
1538 }
1539 }
1540 return $info;
1541 }
1542
1543 /**
1544 * Converting <A>-tags to absolute URLs (+ setting rtekeep attribute)
1545 *
1546 * @param string Content input
1547 * @param boolean If true, then the "rtekeep" attribute will not be set.
1548 * @return string Content output
1549 */
1550 function TS_AtagToAbs($value,$dontSetRTEKEEP=FALSE) {
1551 $blockSplit = $this->splitIntoBlock('A',$value);
1552 reset($blockSplit);
1553 while(list($k,$v)=each($blockSplit)) {
1554 if ($k%2) { // block:
1555 $attribArray = $this->get_tag_attributes_classic($this->getFirstTag($v),1);
1556
1557 // Checking if there is a scheme, and if not, prepend the current url.
1558 if (strlen($attribArray['href'])) { // ONLY do this if href has content - the <a> tag COULD be an anchor and if so, it should be preserved...
1559 $uP = parse_url(strtolower($attribArray['href']));
1560 if (!$uP['scheme']) {
1561 $attribArray['href'] = $this->siteUrl().substr($attribArray['href'],strlen($this->relBackPath));
1562 }
1563 } else {
1564 $attribArray['rtekeep'] = 1;
1565 }
1566 if (!$dontSetRTEKEEP) $attribArray['rtekeep'] = 1;
1567
1568 $bTag='<a '.t3lib_div::implodeAttributes($attribArray,1).'>';
1569 $eTag='</a>';
1570 $blockSplit[$k] = $bTag.$this->TS_AtagToAbs($this->removeFirstAndLastTag($blockSplit[$k])).$eTag;
1571 }
1572 }
1573 return implode('',$blockSplit);
1574 }
1575 }
1576
1577
1578 if (defined('TYPO3_MODE') && $TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml_proc.php']) {
1579 include_once($TYPO3_CONF_VARS[TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_parsehtml_proc.php']);
1580 }
1581 ?>