Navigation

← Previous Changeset
Next Changeset →

Changeset 222

Timestamp:

04/29/07 22:52:18 (3 years ago)

Author:

bermiferrer

Message:

Adding new converters from and to HTML

Files:

trunk/lib/Ak.php (modified) (2 diffs)
trunk/lib/AkConverters/AkHtmlToMarkdown.php (added)
trunk/lib/AkConverters/AkHtmlToSmartypants.php (added)
trunk/lib/AkConverters/AkHtmlToTextile.php (added)
trunk/lib/AkConverters/AkMarkdownToHtml.php (added)
trunk/lib/AkConverters/AkSmartypantsToHtml.php (added)
trunk/lib/AkConverters/AkTextileToHtml.php (added)
trunk/vendor/TextParsers/html2text.php (modified) (1 diff)

Legend:

: Unmodified
: Added
: Removed
: Modified
: Copied
: Moved

trunk/lib/Ak.php

r217	r222
1348	1348	$file_name = $options['path'].DS.$converter_class_name.'.php';
1349	1349	if(!file_exists($file_name)){
1350		trigger_error(Ak::t('Could not locate %from to %to converter on %file_name',array('%from'=>$options['from'],'%to'=>$options['to'],'%file_name'=>$file_name)),E_USER_~~ERROR~~);
	1350	trigger_error(Ak::t('Could not locate %from to %to converter on %file_name',array('%from'=>$options['from'],'%to'=>$options['to'],'%file_name'=>$file_name)),E_USER_NOTICE);
1351	1351	return false;
1352	1352	}
…	…
1354	1354	}
1355	1355	if(!class_exists($converter_class_name)){
1356		trigger_error(Ak::t('Could not load %converter_class_name converter class',array('%converter_class_name'=>$converter_class_name)),E_USER_~~ERROR~~);
	1356	trigger_error(Ak::t('Could not load %converter_class_name converter class',array('%converter_class_name'=>$converter_class_name)),E_USER_NOTICE);
1357	1357	return false;
1358	1358	}

trunk/vendor/TextParsers/html2text.php

r113	r222
1	1	<?php
2	2	/**
3		* This is a port by Milian Wolff of Aaron Swartz' html2text python script.
	3	* html2text converts HTML Markup to Markdown [1]. It also supports
	4	* Markdown Extra [2] by Michel Fortin [3].
4	5	*
5		* Quote:
6		* > html2text is a Python script that converts a page of HTML into clean, easy-
7		* > to-read plain ASCII text. Better yet, that ASCII also happens to be valid
8		* > Markdown (a text-to-HTML format).
	6	* It started as a port of Aaron Swartz' [4] html2text.py [5] but
	7	* got a long way since. This is more than a mere port now!
9	8	*
10		* Support for PHP Markdown Extra [1] by Michel Fortin [2] was also added
11		*
12		* [1]: http://www.michelf.com/projects/php-markdown/extra/
13		* [2]: http://www.michelf.com/
	9	* [1]: http://daringfireball.com/projects/markdown
	10	* [2]: http://www.michelf.com/projects/php-markdown/extra/
	11	* [3]: http://www.michelf.com/
	12	* [4]: http://www.aaronsw.com/
	13	* [5]: http://www.aaronsw.com/2002/html2text/
14	14	*
15	15	* @version 1.4
16		* @license http://www.gnu.org/licenses/gpl.txt GNU GPL 2
	16	* @author Milian Wolff (mail@milianw.de,http://milianw.de)
	17	* @license LGPL, see LICENSE.txt and below
	18	* @copyright (C) 2007 Milian Wolff
17	19	*
18		* # port:
19		* @author Milian Wolff (milianwolff@gmx.net)
20		* @copyright (c) 2006 Milian Wolff
21		* # original:
22		* @author Aaron Swartz (me@aaronsw.com)
23		* @copyright (c) 2004 Aaron Swartz
	20	* This library is free software; you can redistribute it and/or
	21	* modify it under the terms of the GNU Lesser General Public
	22	* License as published by the Free Software Foundation; either
	23	* version 2.1 of the License, or (at your option) any later version.
	24	*
	25	* This library is distributed in the hope that it will be useful,
	26	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	27	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	28	* Lesser General Public License for more details.
	29	*
	30	* You should have received a copy of the GNU Lesser General Public
	31	* License along with this library; if not, write to the Free Software
	32	* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
24	33	*/
25	34	class html2text {
26		# input html
27		var $html = '';
28		# output markdown
29		var $outtext = '';
30		# some control structures
31		var $p_p = 0;
32		var $start = 1;
33		var $space = 0;
34		var $force_html = false;
35		var $force_html_start = array('tag'=>'','parents'=>0);
36		# links
37		var $a = array();
38		var $astack = array();
39		var $acount = 0;
40		# lists
41		var $list = array ();
42		var $list_depth = 0;
43		var $lastWasNL = false;
44		# indenting and appending
45		var $append = '';
46		var $indent = '';
47		# these elements will be dropped with all subelements
48		var $drop = array(
49		'script',
50		'head',
51		'style',
52		'form',
53		);
54		# these elements will be quietly ignored, their children will be parsed
55		var $ignore = array(
56		'wrapper', # important!
57		'html', # closing html tag
58		'body', # closing body tag
59		'thead',
60		'tbody',
61		'tfoot',
62		);
63		# these elements can have some attributes
64		var $has_attrs = array(
65		# tag => list of allowed attrs
66		'h1' => array('id'),
67		'h2' => array('id'),
68		'h3' => array('id'),
69		'h4' => array('id'),
70		'h5' => array('id'),
71		'h6' => array('id'),
72		'a' => array('href','title'),
73		'img' => array('src','alt','title'),
74		# tables
75		'th' => array('align'),
76		'td' => array('align'),
77		# footnotes
78		'sup' => array('id'),
79		'footnote' => array('nr'),
80		# abbrevations
81		'abbr' => array('title'),
82		'acronym' => array('title'),
83		);
84		# table
85		var $max_len = array();
86		var $align = array();
87		var $cols = array();
88		var $rows = array();
89		var $col = 0;
90		var $row = 0;
91		var $header = array();
92		# parents
93		var $parents = array();
94		# abbrevations
95		var $abbrs = array();
96		# buffer
97		var $buffer = array();
98		var $buffer_lvl = 0;
99		# options
100		var $LINKS_EACH_PARAGRAPH;
101		var $BODY_WIDTH;
102		var $KEEP_HTML;
103		# global xml parser
104		var $xml_parser;
105		/**
106		* setup the xml_parser
107		* $links_each_paragraph: if set to true, the list of links will be
108		* displayed after each paragraph, else it will be displayed on the end of
109		* the file
110		* $body_width: if set to a integer greater 0 the output text will be
111		* wrapped to that width (in characters)
112		* $keep_html: if set to true, all unrecognized html tags will be kept, else
113		* they'll be removed
114		*
115		* @param bool $links_each_paragraph default true
116		* @param integer $body_width default 0
117		* @param bool $keep_html default true
118		* @return void
119		*/
120		function html2text($links_each_paragraph = true,$body_width = 0,$keep_html = true) {
121		$this->LINKS_EACH_PARAGRAPH = $links_each_paragraph;
122		$this->BODY_WIDTH = $body_width;
123		$this->KEEP_HTML = $keep_html;
124		$this->xml_parser = xml_parser_create();
125		xml_set_object($this->xml_parser, $this);
126		xml_parser_set_option($this->xml_parser, XML_OPTION_CASE_FOLDING, 0);
127		xml_set_element_handler($this->xml_parser, 'starttag', 'endtag');
128		xml_set_character_data_handler($this->xml_parser, 'handle_data');
129		xml_set_default_handler($this->xml_parser,'handle_default');
130		}
131		/**
132		* parse a html string to text
133		*
134		* @param string $html
135		* @return string
136		*/
137		function load_string($html) {
138		$html = trim($html);
139		if(empty($html)){
140		return '';
141		}
142		# use unix style newlines
143		$html = str_replace("\r","\n",str_replace("\r\n","\n",$html));
144		# remove doctype and xml tags
145		$html = preg_replace('#^.<body[^>]>#Us','<html><body>',$html);
146		/*
147		* cope with bad html
148		*/
149		$html = preg_replace('#&(?!amp;)#','&',$html);
150		$html = str_replace('<','<',$html);
151		$html = preg_replace('#<([a-z]+[^>]*) ?/>#Us','<$1 />',$html);
152		# unmatched tags (poor performance)
153		preg_match_all('#<(([a-z]\|h[1-6])+)(?= \|>)#',$html,$matches);
154		foreach($matches[1] as $tag){
155		$html = preg_replace('#<'.$tag.'( \|>)(.*)</'.$tag.'>#Us','<'.$tag.'$1$2</'.$tag.'>',$html,1);
156		}
157		# encode < to < and & to & inside <pre>\|<code>
158		$html = preg_replace_callback('#(<pre[^>]>\s<code[^>]>\|<code[^>]>\|<pre[^>]>)(.)(</pre>\s*</code>\|</code>\|</pre>)#Us',
159		create_function(
160		'$matches',
161		'return $matches[1].str_replace(\'<\',\'<\',$matches[2]).$matches[3];'
162		),$html);
163		# handle empty attributes (e.g. <input checked>)
164		$html = preg_replace_callback('#<([a-z]+)(?>[^>]* [^=]+(?> [^>]*)?) ?/?>#s',array(&$this,'parse_empty_attribs'),$html);
165		# fake wrapper
166		$html = '<wrapper>'.$html.'</wrapper>';
167		# footnotes
168		$html = preg_replace_callback('#<div class="footnotes">\s<hr />\s<ol>\s(<li id="fn:\d+">.+</li>)\s</ol>\s*</div>#Us',array(&$this,'footnotes'),$html);
169		# last newline inside <pre> should not be parsed
170		$html = preg_replace('#\n</code></pre>#s','</code></pre>',$html);
171		# some html elements should not be parsed if their children wont be parsed:
172		if($this->KEEP_HTML){
173		# <ul\|ol><li class="asdf">, complex because we need to handle nested lists
174		if(preg_match('#<li [^>]+>#',$html)){
175		preg_match_all('#(?:<li [^>]+>\|</?(?:ul\|ol)[^>]*>)#',$html,$matches,PREG_OFFSET_CAPTURE);
176		$lists = array();
177		$offset = 0;
178		$ins = ' forcehtml="1"';
179		$add = strlen($ins);
180		foreach($matches[0] as $k => $a){
181		if(substr($a[0],0,3) == '<li'){
182		$list = &$lists[count($lists)-1];
183		if(!$list['forced']){
184		$list['forced'] = true;
185		$html = substr_replace($html,$ins,$list['offset']+$offset,0);
186		$offset += $add;
187		}
188		} else {
189		if(substr($a[0],0,2) == '</'){ # close tag
190		array_pop($lists);
191		} else { # open tag
192		array_push($lists,array(
193		'offset' => $a[1]+3,
194		'forced' => strstr($a[0],'forcehtml='),
195		));
196		}
197		}
198		}
199		}
200		# <pre><code class="asdf">
201		$html = preg_replace('#(?><pre>)\s*(<code .+>)#Us','<pre forcehtml="1">$1',$html);
202		}
203		$this->html = $html;
204		# ok, now lets start parsing!
205		#echo dump($html);
206		$this->parse();
207		return $this->close();
208		}
209		/**
210		* clean up footnotes
211		*
212		* @param array $matches
213		* @return string
214		*/
215		function footnotes($matches){
216		# remove footnote link
217		$matches = preg_replace('@<a href="#fnref:\d+" rev="footnote"[^>]*>&#8617;</a>@U','',$matches[1]);
218		# remove empty paragraph
219		$matches = str_replace('<p></p>','',$matches);
220		# wrap in footnotes tag
221		$matches = '<footnotes>'.$matches.'</footnotes>';
222		# <li id="fn:1">...</li> -> <footnote nr="1">...</footnote>
223		$matches = str_replace('<li id="fn:','<footnote nr="',$matches);
224		return preg_replace('#</li>\s*(<footnote\|</footnotes)#s','</footnote>$1',$matches);
225		}
226		/**
227		* @param array $matches
228		* @return string
229		*/
230		function parse_empty_attribs($matches){
231		if(preg_match('#^<[a-z]+(?: [a-z]+=(?:"[^"]"\|\'[^\']\'))+ ?/?>$#s',$matches[0])){
232		# mismatch, this tag is correct
233		return $matches[0];
234		}
235		echo dump($matches[0]);
236		die();
237		$rep = $this->KEEP_HTML ? '$1="$1"' : '';
238		return '<'.$matches[1].preg_replace('#(?<= )([^ =>]{2,})(?= \|$)#Us',$rep,$matches[2]).'>';
239		}
240		/**
241		* parse a html file to text
242		*
243		* @param string $file
244		* @return string
245		*/
246		function load_file($file) {
247		$contents = file_get_contents($file);
248		if(!$contents){
249		trigger_error('could not open XML input',E_USER_WARNING);
250		return false;
251		}
252		return $this->load_string($contents);
253		}
254		/**
255		* start parsing html to text
256		*
257		* @param void
258		* @return void
259		*/
260		function parse() {
261		$html = explode("\n", $this->html);
262		foreach ($html as $line) {
263		if (!xml_parse($this->xml_parser, $line . "\n")) {
264		$errcode = xml_get_error_code($this->xml_parser);
265		trigger_error(sprintf("XML error #%d: %s at line %d:<br /><pre><code>%s</code></pre>", $errcode,xml_error_string($errcode), xml_get_current_line_number($this->xml_parser),htmlspecialchars($line)),E_USER_WARNING);
266		#return;
267		}
268		}
269		}
270		/**
271		* close parser and return text
272		*
273		* @param void
274		* @return string
275		*/
276		function close() {
277		xml_parser_free($this->xml_parser);
278		$this->pbr();
279		$this->o('', false, 'end');
280		$this->out("\n");
281		$this->links();
282		# blockquotes
283		$this->outtext = preg_replace_callback('#^(\s*)((> )+)#m',array(&$this,'cleanup_bq'),$this->outtext);
284		# cleanup
285		$this->outtext = str_replace('&','&',str_replace('<','<',str_replace('>','>',$this->outtext)));
286		# empty lines (not preformatted)
287		$this->outtext = preg_replace('#^\s{1,4}$#m','',$this->outtext);
288		# empty quoted lines
289		$this->outtext = preg_replace('#^(>+)\s{1,5}$#m','$1',$this->outtext);
290		return rtrim($this->optwrap($this->outtext));
291		}
292		/**
293		* replace "> > > " with ">>> "
294		*
295		* @param array $m matches
296		* @return string
297		*/
298		function cleanup_bq($m){
299		return $m[1].str_repeat('>',strlen($m[2])/2).' ';
300		}
301		/**
302		* handles html comments
303		*
304		* @param resource $parser
305		* @param string $data
306		* @return void
307		*/
308		function handle_default($parser,$data){
309		if(substr($data,0,4) == '<!--' && substr($data,-3) == '-->'){
310		$this->outtext .= "\n\n".$data."\n";
311		}
312		}
313		/**
314		* adds pure data to the output (e.g. <p>DATA</p>)
315		*
316		* @param resource $parser
317		* @param string $data
318		* @return void
319		*/
320		function handle_data($parser, $data) {
321		$this->o($data, true);
322		}
323		/**
324		* start tags (e.g. <p>)
325		*
326		* @param resource $parser
327		* @param string $tag
328		* @param array $attrs
329		* @return void
330		*/
331		function starttag($parser, $tag, $attrs) {
332		$this->handle_tag($tag, $attrs, true);
333		}
334		/**
335		* end tags (e.g. </p>)
336		*
337		* @param resource $parser
338		* @param string $tag
339		* @return void
340		*/
341		function endtag($parser, $tag) {
342		$this->handle_tag($tag, null, false);
343		}
344		/**
345		* force html output of all children
346		*
347		* @param $tag
348		* @return void
349		*/
350		function force_html($tag){
351		$this->force_html = true;
352		$this->force_html_start = array(
353		'tag' => $tag,
354		'parents' => isset($this->parents[$tag]) ? strlen($this->parents[$tag]) : 0
355		);
356		}
357		/**
358		* parsing logic based on tag name
359		*
360		* @param string $tag
361		* @param array $attrs
362		* @param bool $start
363		* @return void
364		*/
365		function handle_tag($tag, $attrs, $start) {
366		if(in_array($tag,$this->drop)){ # drop tags with content
367		if($start){
368		$this->buffer();
369		} else {
370		$this->unbuffer();
371		}
372		return;
373		}
374		if(in_array($tag,$this->ignore)){ # drop tags but keep content
375		return;
376		}
377		# keeping the original html
378		if($this->KEEP_HTML){
379		if($start){
380		# is the force html attr set?
381		if(!$this->force_html && isset($attrs['forcehtml'])){
382		$this->force_html($tag);
383		}
384		# we'll have to keep this tag
385		if($this->force_html) {
386		$this->keep_tag($tag,$attrs,$start,true);
387		return;
388		} else {
389		# tag has attrs which can't be converted
390		if(!empty($attrs) && $this->keep_tag($tag,$attrs,$start)){
391		return;
392		}
393		}
394		} else {
395		if($this->force_html){
396		$this->keep_tag($tag,$attrs,$start,true);
397		if($tag == $this->force_html_start['tag'] && strlen($this->parents[$tag]) == $this->force_html_start['parents']){
398		$this->force_html = false;
399		}
400		return;
401		} elseif($this->parent($tag,'kept') && $this->keep_tag($tag,$attrs,$start)) {
402		return;
403		}
404		}
405		}
406		switch ($tag) {
407		case 'h1' :
408		case 'h2' :
409		case 'h3' :
410		case 'h4' :
411		case 'h5' :
412		case 'h6' :
413		$this->p();
414		if ($start) {
415		$this->o(str_repeat('#', intval($tag[1])) . ' ');
416		if(!empty($attrs['id'])){
417		$this->append = ' {#'.$attrs['id'].'}';
418		}
419		} else {
420		$this->out($this->append);
421		$this->append = '';
422		}
423		break;
424		case 'div' :
425		$this->p();
426		break;
427		case 'p' :
428		$this->p();
429		break;
430		case 'br' :
431		if ($start) {
432		$this->o(" \n");
433		}
434		break;
435		case 'hr' :
436		if ($start) {
437		$this->p();
438		$this->o('* * *');
439		$this->p();
440		}
441		break;
442		case 'blockquote' :
443		$this->indent('> ',$start);
444		if ($start) {
445		$this->start = true;
446		$this->out("\n\n".$this->indent);
447		}
448		break;
449		case 'em' :
450		case 'i' :
451		case 'u' :
452		$this->o('_');
453		break;
454		case 'strong' :
455		case 'b' :
456		$this->o('**');
457		break;
458		# footnotes
459		case 'sup':
460		if($start){
461		if(count($attrs) != 1 \|\| !isset($attrs['id']) \|\| !preg_match('#^fnref:(\d+)$#',$attrs['id'],$matches)){
462		# keep tag
463		$this->keep_tag($tag,$attrs,$start,true);
464		return;
465		}
466		# parse footnote
467		$this->out('[^'.$matches[1].']');
468		# omit output of link (<a href="#fn:1" rel="footnote">1</a>)
469		$this->buffer();
470		} else {
471		# last sup was not parsed -> keep tag
472		if(!$this->parent('sup')){
473		$this->keep_tag($tag,$attrs,$start);
474		return;
475		}
476		# sup was parsed -> reset buffer
477		$this->unbuffer();
478		}
479		break;
480		case 'footnotes':
481		$this->p();
482		break;
483		case 'footnote':
484		if($start){
485		$this->o('[^'.$attrs['nr']."]:\n".$this->indent.' ');
486		$this->start = true;
487		}
488		$this->indent(' ',$start);
489		break;
490		case 'a':
491		if($start) {
492		# buffer to check for inline links like <foo@bar.com> and the like
493		if (isset ($attrs['href'])) {
494		$this->buffer();
495		array_push($this->astack, $attrs);
496		} else {
497		array_push($this->astack, null);
498		}
499		} else {
500		if($this->astack) {
501		$a = array_pop($this->astack);
502		if ($a) {
503		# for emails
504		$a['href'] = $this->decode($a['href']);
505		$buffer = $this->unbuffer();
506		$buffer_check = $this->decode(trim($buffer));
507		if((substr($a['href'],0,7) == 'mailto:' && 'mailto:'.$buffer_check == $a['href']) \|\| $a['href'] == $buffer_check){
508		# inline link
509		$this->out('<'.$buffer_check.'>',true);
510		} else {
511		# block link
512		$this->previousIndex($a);
513		$this->out('['.$buffer.']['.$a['count'].']',true);
514		}
515		}
516		}
517		}
518		break;
519		# abbrevations
520		case 'abbr':
521		case 'acronym':
522		if($start){
523		$this->buffer();
524		array_push($this->abbrs,isset($attrs['title'])?$attrs['title']:'');
525		} else {
526		$abbr = $this->unbuffer();
527		$def = array_pop($this->abbrs);
528		# only add abbr if its not already defined
529		if(!isset($this->abbrs[$abbr])){
530		$this->abbrs[$abbr] = $def;
531		}
532		$this->o($abbr);
533		}
534		break;
535		case 'img' :
536		if ($start) {
537		if (isset ($attrs['src'])) {
538		$attrs['href'] = $attrs['src'];
539		$alt = '';
540		if (isset ($attrs['alt'])) {
541		$alt = $attrs['alt'];
542		} elseif(isset($attrs['title'])){
543		$alt = $attrs['title'];
544		}
545		$this->previousIndex($attrs);
546		$this->o('!['.$alt.'][' . $attrs['count'] . ']');
547		}
548		}
549		break;
550		case 'code':
551		# do we have to keep this tag?
552		# or is a parent <pre> element existing?
553		if($this->keep_tag($tag,$attrs,$start) \|\| $this->parent('pre')){
554		return;
555		}
556		# convert to `code` and handle backticks inside code block
557		# <code>foo`bar</code> has to get ``foo`bar`` and so forth
558		if($start){
559		$this->buffer();
560		} else {
561		$str = $this->unbuffer();
562		preg_match_all('#`+#',$str,$matches);
563		if(!empty($matches[0])){
564		rsort($matches[0]);
565		$len = strlen($matches[0][0])+1;
566		} else {
567		$len = 1;
568		}
569		$ticks = str_repeat('`',$len);
570		$this->out($ticks.$str.$ticks);
571		}
572		break;
573		case 'dl' :
574		# note: if <dl> gets parsed, its direct children (<dd> and <dt>) will be parsed as well
575		if ($start) {
576		$this->p();
577		}
578		break;
579		case 'dd' :
580		# is the parent dl parsed?
581		if(!$this->parent('dl')){
582		$this->keep_tag($tag,$attrs,$start,true);
583		return;
584		}
585		if ($start) {
586		$this->o(': ');
587		$this->start = true;
588		} else {
589		$this->outtext .= "\n";
590		$this->pbr();
591		}
592		$this->indent(' ',$start);
593		break;
594		case 'dt' :
595		# is the parent dl parsed?
596		if(!$this->parent('dl')){
597		$this->keep_tag($tag,$attrs,$start,true);
598		return;
599		}
600		if (!$start) {
601		$this->pbr();
602		}
603		break;
604		case 'ol' :
605		case 'ul' :
606		# note: if this element gets parsed, its direct children <li>s will be parsed as well
607		if ($start) {
608		array_push($this->list, array (
609		'name' => $tag,
610		'num' => 0
611		));
612		} else {
613		array_pop($this->list);
614		$this->pbr();
615		}
616		break;
617		case 'li' :
618		if ($this->list) {
619		$li = &$this->list[count($this->list) - 1];
620		}
621		# not inside a list or the list tag was not parsed
622		if(!isset($li) \|\| !$this->parent($li['name'])){
623		$this->keep_tag($tag,$attrs,$start,true);
624		return;
625		}
626		if ($start) {
627		$this->pbr();
628		if($li['name'] == 'ul'){
629		$this->o('* ');
630		} else {
631		$li['num']++;
632		/**
633		* @todo line up <ol><li>s > 9 correctly.
634		*/
635		$this->o($li['num'].'. ');
636		}
637		$this->start = true;
638		$this->indent(' ',$start);
639		} else {
640		$this->indent(' ',$start);
641		}
642		break;
643		case 'table':
644		# NOTE: if the <table> tag gets parsed, all its children will be as well!
	35	# input html
	36	var $html = '';
	37	# output markdown
	38	var $outtext = '';
	39	# some control structures
	40	var $p_p = 0;
	41	var $start = 1;
	42	var $space = 0;
	43	var $force_html = false;
	44	var $force_html_start = array('tag'=>'','parents'=>0);
	45	# links
	46	var $a = array();
	47	var $astack = array();
	48	var $acount = 0;
	49	# lists
	50	var $list = array ();
	51	var $list_depth = 0;
	52	var $lastWasNL = false;
	53	# indenting and appending
	54	var $append = '';
	55	var $indent = '';
	56	# these elements will be dropped with all subelements
	57	var $drop = array(
	58	'script',
	59	'head',
	60	'style',
	61	'form',
	62	);
	63	# these elements will be quietly ignored, their children will be parsed
	64	var $ignore = array(
	65	'wrapper', # important!
	66	'html', # closing html tag
	67	'body', # closing body tag
	68	'thead',
	69	'tbody',
	70	'tfoot',
	71	);
	72	# these elements can have some attributes
	73	var $has_attrs = array(
	74	# tag => list of allowed attrs
	75	'h1' => array('id'),
	76	'h2' => array('id'),
	77	'h3' => array('id'),
	78	'h4' => array('id'),
	79	'h5' => array('id'),
	80	'h6' => array('id'),
	81	'a' => array('href','title'),
	82	'img' => array('src','alt','title'),
	83	# tables
	84	'th' => array('align'),
	85	'td' => array('align'),
	86	# footnotes
	87	'sup' => array('id'),
	88	'footnote' => array('nr'),
	89	# abbrevations
	90	'abbr' => array('title'),
	91	'acronym' => array('title'),
	92	);
	93	# table
	94	var $max_len = array();
	95	var $align = array();
	96	var $cols = array();
	97	var $rows = array();
	98	var $col = 0;
	99	var $row = 0;
	100	var $header = array();
	101	# parents
	102	var $parents = array();
	103	# abbrevations
	104	var $abbrs = array();
	105	# buffer
	106	var $buffer = array();
	107	var $buffer_lvl = 0;
	108	# options
	109	var $LINKS_EACH_PARAGRAPH;
	110	var $BODY_WIDTH;
	111	var $KEEP_HTML;
	112	# global xml parser
	113	var $xml_parser;
	114	/**
	115	* setup the xml_parser
	116	* $links_each_paragraph: if set to true, the list of links will be
	117	* displayed after each paragraph, else it will be displayed on the end of
	118	* the file
	119	* $body_width: if set to a integer greater 0 the output text will be
	120	* wrapped to that width (in characters)
	121	* $keep_html: if set to true, all unrecognized html tags will be kept, else
	122	* they'll be removed
	123	*
	124	* @param bool $links_each_paragraph default true
	125	* @param integer $body_width default 0
	126	* @param bool $keep_html default true
	127	* @return void
	128	*/
	129	function html2text($links_each_paragraph = true,$body_width = 0,$keep_html = true) {
	130	$this->LINKS_EACH_PARAGRAPH = $links_each_paragraph;
	131	$this->BODY_WIDTH = $body_width;
	132	$this->KEEP_HTML = $keep_html;
	133	$this->xml_parser = xml_parser_create();
	134	xml_set_object($this->xml_parser, $this);
	135	xml_parser_set_option($this->xml_parser, XML_OPTION_CASE_FOLDING, 0);
	136	xml_set_element_handler($this->xml_parser, 'starttag', 'endtag');
	137	xml_set_character_data_handler($this->xml_parser, 'handle_data');
	138	xml_set_default_handler($this->xml_parser,'handle_default');
	139	}
	140	/**
	141	* parse a html string to text
	142	*
	143	* @param string $html
	144	* @return string
	145	*/
	146	function load_string($html) {
	147	$html = trim($html);
	148	if(empty($html)){
	149	return '';
	150	}
	151	# use unix style newlines
	152	$html = str_replace("\r","\n",str_replace("\r\n","\n",$html));
	153	# remove doctype and xml tags
	154	$html = preg_replace('#^.<body[^>]>#Us','<html><body>',$html);
	155	/*
	156	* cope with bad html
	157	*/
	158	$html = preg_replace('/<img(?!.\/>)([^>])>/Us','<img$1 />',$html);
	159	$html = preg_replace('#&(?!amp;)#','&',$html);
	160	$html = str_replace('<','<',$html);
	161	$html = preg_replace('#<([a-z]+[^>]*) ?/>#Us','<$1 />',$html);
	162	# unmatched tags (poor performance)
	163	preg_match_all('#<(([a-z]\|h[1-6])+)(?= \|>)#',$html,$matches);
	164	foreach($matches[1] as $tag){
	165	$html = preg_replace('#<'.$tag.'( \|>)(.*)</'.$tag.'>#Us','<'.$tag.'$1$2</'.$tag.'>',$html,1);
	166	}
	167	# encode < to < and & to & inside <pre>\|<code>
	168	$html = preg_replace_callback('#(<pre[^>]>\s<code[^>]>\|<code[^>]>\|<pre[^>]>)(.)(</pre>\s*</code>\|</code>\|</pre>)#Us',
	169	create_function(
	170	'$matches',
	171	'return $matches[1].str_replace(\'<\',\'<\',$matches[2]).$matches[3];'
	172	),$html);
	173	# handle empty attributes (e.g. <input checked>)
	174	$html = preg_replace_callback('#<([a-z]+)(?>[^>]* [^=]+(?> [^>]*)?) ?/?>#s',array(&$this,'parse_empty_attribs'),$html);
	175	# fake wrapper
	176	$html = '<wrapper>'.$html.'</wrapper>';
	177	# footnotes
	178	$html = preg_replace_callback('#<div class="footnotes">\s<hr />\s<ol>\s(<li id="fn:\d+">.+</li>)\s</ol>\s*</div>#Us',array(&$this,'footnotes'),$html);
	179	# last newline inside <pre> should not be parsed
	180	$html = preg_replace('#\n</code></pre>#s','</code></pre>',$html);
	181	# some html elements should not be parsed if their children wont be parsed:
	182	if($this->KEEP_HTML){
	183	# <ul\|ol><li class="asdf">, complex because we need to handle nested lists
	184	if(preg_match('#<li [^>]+>#',$html)){
	185	preg_match_all('#(?:<li [^>]+>\|</?(?:ul\|ol)[^>]*>)#',$html,$matches,PREG_OFFSET_CAPTURE);
	186	$lists = array();
	187	$offset = 0;
	188	$ins = ' forcehtml="1"';
	189	$add = strlen($ins);
	190	foreach($matches[0] as $k => $a){
	191	if(substr($a[0],0,3) == '<li'){
	192	$list = &$lists[count($lists)-1];
	193	if(!$list['forced']){
	194	$list['forced'] = true;
	195	$html = substr_replace($html,$ins,$list['offset']+$offset,0);
	196	$offset += $add;
	197	}
	198	} else {
	199	if(substr($a[0],0,2) == '</'){ # close tag
	200	array_pop($lists);
	201	} else { # open tag
	202	array_push($lists,array(
	203	'offset' => $a[1]+3,
	204	'forced' => strstr($a[0],'forcehtml='),
	205	));
	206	}
	207	}
	208	}
	209	}
	210	# <pre><code class="asdf">
	211	$html = preg_replace('#(?><pre>)\s*(<code .+>)#Us','<pre forcehtml="1">$1',$html);
	212	}
	213	$this->html = $html;
	214	# ok, now lets start parsing!
	215	#echo dump($html);
	216	$this->parse();
	217	return $this->close();
	218	}
	219	/**
	220	* clean up footnotes
	221	*
	222	* @param array $matches
	223	* @return string
	224	*/
	225	function footnotes($matches){
	226	# remove footnote link
	227	$matches = preg_replace('@<a href="#fnref:\d+" rev="footnote"[^>]*>&#8617;</a>@U','',$matches[1]);
	228	# remove empty paragraph
	229	$matches = str_replace('<p></p>','',$matches);
	230	# wrap in footnotes tag
	231	$matches = '<footnotes>'.$matches.'</footnotes>';
	232	# <li id="fn:1">...</li> -> <footnote nr="1">...</footnote>
	233	$matches = str_replace('<li id="fn:','<footnote nr="',$matches);
	234	return preg_replace('#</li>\s*(<footnote\|</footnotes)#s','</footnote>$1',$matches);
	235	}
	236	/**
	237	* @param array $matches
	238	* @return string
	239	*/
	240	function parse_empty_attribs($matches){
	241	if(preg_match('#^<[a-z]+(?: [a-z]+=(?:"[^"]"\|\'[^\']\'))+ ?/?>$#s',$matches[0])){
	242	# mismatch, this tag is correct
	243	return $matches[0];
	244	}
	245	echo dump($matches[0]);
	246	die();
	247	$rep = $this->KEEP_HTML ? '$1="$1"' : '';
	248	return '<'.$matches[1].preg_replace('#(?<= )([^ =>]{2,})(?= \|$)#Us',$rep,$matches[2]).'>';
	249	}
	250	/**
	251	* parse a html file to text
	252	*
	253	* @param string $file
	254	* @return string
	255	*/
	256	function load_file($file) {
	257	$contents = file_get_contents($file);
	258	if(!$contents){
	259	trigger_error('could not open XML input',E_USER_WARNING);
	260	return false;
	261	}
	262	return $this->load_string($contents);
	263	}
	264	/**
	265	* start parsing html to text
	266	*
	267	* @param void
	268	* @return void
	269	*/
	270	function parse() {
	271	$html = explode("\n", $this->html);
	272	foreach ($html as $line) {
	273	if (!xml_parse($this->xml_parser, $line . "\n")) {
	274	$errcode = xml_get_error_code($this->xml_parser);
	275	trigger_error(sprintf("XML error #%d: %s at line %d:<br /><pre><code>%s</code></pre>", $errcode,xml_error_string($errcode), xml_get_current_line_number($this->xml_parser),htmlspecialchars($line)),E_USER_WARNING);
	276	#return;
	277	}
	278	}
	279	}
	280	/**
	281	* close parser and return text
	282	*
	283	* @param void
	284	* @return string
	285	*/
	286	function close() {
	287	xml_parser_free($this->xml_parser);
	288	$this->pbr();
	289	$this->o('', false, 'end');
	290	$this->out("\n");
	291	$this->links();
	292	# blockquotes
	293	$this->outtext = preg_replace_callback('#^(\s*)((> )+)#m',array(&$this,'cleanup_bq'),$this->outtext);
	294	# cleanup
	295	$this->outtext = str_replace('&','&',str_replace('<','<',str_replace('>','>',$this->outtext)));
	296	# empty lines (not preformatted)
	297	$this->outtext = preg_replace('#^\s{1,4}$#m','',$this->outtext);
	298	# empty quoted lines
	299	$this->outtext = preg_replace('#^(>+)\s{1,5}$#m','$1',$this->outtext);
	300	return rtrim($this->optwrap($this->outtext));
	301	}
	302	/**
	303	* replace "> > > " with ">>> "
	304	*
	305	* @param array $m matches
	306	* @return string
	307	*/
	308	function cleanup_bq($m){
	309	return $m[1].str_repeat('>',strlen($m[2])/2).' ';
	310	}
	311	/**
	312	* handles html comments
	313	*
	314	* @param resource $parser
	315	* @param string $data
	316	* @return void
	317	*/
	318	function handle_default($parser,$data){
	319	if(substr($data,0,4) == '<!--' && substr($data,-3) == '-->'){
	320	$this->outtext .= "\n\n".$data."\n";
	321	}
	322	}
	323	/**
	324	* adds pure data to the output (e.g. <p>DATA</p>)
	325	*
	326	* @param resource $parser
	327	* @param string $data
	328	* @return void
	329	*/
	330	function handle_data($parser, $data) {
	331	$this->o($data, true);
	332	}
	333	/**
	334	* start tags (e.g. <p>)
	335	*
	336	* @param resource $parser
	337	* @param string $tag
	338	* @param array $attrs
	339	* @return void
	340	*/
	341	function starttag($parser, $tag, $attrs) {
	342	$this->handle_tag($tag, $attrs, true);
	343	}
	344	/**
	345	* end tags (e.g. </p>)
	346	*
	347	* @param resource $parser
	348	* @param string $tag
	349	* @return void
	350	*/
	351	function endtag($parser, $tag) {
	352	$this->handle_tag($tag, null, false);
	353	}
	354	/**
	355	* force html output of all children
	356	*
	357	* @param $tag
	358	* @return void
	359	*/
	360	function force_html($tag){
	361	$this->force_html = true;
	362	$this->force_html_start = array(
	363	'tag' => $tag,
	364	'parents' => isset($this->parents[$tag]) ? strlen($this->parents[$tag]) : 0
	365	);
	366	}
	367	/**
	368	* parsing logic based on tag name
	369	*
	370	* @param string $tag
	371	* @param array $attrs
	372	* @param bool $start
	373	* @return void
	374	*/
	375	function handle_tag($tag, $attrs, $start) {
	376	if(in_array($tag,$this->drop)){ # drop tags with content
	377	if($start){
	378	$this->buffer();
	379	} else {
	380	$this->unbuffer();
	381	}
	382	return;
	383	}
	384	if(in_array($tag,$this->ignore)){ # drop tags but keep content
	385	return;
	386	}
	387	# keeping the original html
	388	if($this->KEEP_HTML){
	389	if($start){
	390	# is the force html attr set?
	391	if(!$this->force_html && isset($attrs['forcehtml'])){
	392	$this->force_html($tag);
	393	}
	394	# we'll have to keep this tag
	395	if($this->force_html) {
	396	$this->keep_tag($tag,$attrs,$start,true);
	397	return;
	398	} else {
	399	# tag has attrs which can't be converted
	400	if(!empty($attrs) && $this->keep_tag($tag,$attrs,$start)){
	401	return;
	402	}
	403	}
	404	} else {
	405	if($this->force_html){
	406	$this->keep_tag($tag,$attrs,$start,true);
	407	if($tag == $this->force_html_start['tag'] && strlen($this->parents[$tag]) == $this->force_html_start['parents']){
	408	$this->force_html = false;
	409	}
	410	return;
	411	} elseif($this->parent($tag,'kept') && $this->keep_tag($tag,$attrs,$start)) {
	412	return;
	413	}
	414	}
	415	}
	416	switch ($tag) {
	417	case 'h1' :
	418	case 'h2' :
	419	case 'h3' :
	420	case 'h4' :
	421	case 'h5' :
	422	case 'h6' :
	423	$this->p();
	424	if ($start) {
	425	$this->o(str_repeat('#', intval($tag[1])) . ' ');
	426	if(!empty($attrs['id'])){
	427	$this->append = ' {#'.$attrs['id'].'}';
	428	}
	429	} else {
	430	$this->out($this->append);
	431	$this->append = '';
	432	}
	433	break;
	434	case 'div' :
	435	$this->p();
	436	break;
	437	case 'p' :
	438	$this->p();
	439	break;
	440	case 'br' :
	441	if ($start) {
	442	$this->o(" \n");
	443	}
	444	break;
	445	case 'hr' :
	446	if ($start) {
	447	$this->p();
	448	$this->o('* * *');
	449	$this->p();
	450	}
	451	break;
	452	case 'blockquote' :
	453	$this->indent('> ',$start);
	454	if ($start) {
	455	$this->start = true;
	456	$this->out("\n\n".$this->indent);
	457	}
	458	break;
	459	case 'em' :
	460	case 'i' :
	461	case 'u' :
	462	$this->o('_');
	463	break;
	464	case 'strong' :
	465	case 'b' :
	466	$this->o('**');
	467	break;
	468	# footnotes
	469	case 'sup':
	470	if($start){
	471	if(count($attrs) != 1 \|\| !isset($attrs['id']) \|\| !preg_match('#^fnref:(\d+)$#',$attrs['id'],$matches)){
	472	# keep tag
	473	$this->keep_tag($tag,$attrs,$start,true);
	474	return;
	475	}
	476	# parse footnote
	477	$this->out('[^'.$matches[1].']');
	478	# omit output of link (<a href="#fn:1" rel="footnote">1</a>)
	479	$this->buffer();
	480	} else {
	481	# last sup was not parsed -> keep tag
	482	if(!$this->parent('sup')){
	483	$this->keep_tag($tag,$attrs,$start);
	484	return;
	485	}
	486	# sup was parsed -> reset buffer
	487	$this->unbuffer();
	488	}
	489	break;
	490	case 'footnotes':
	491	$this->p();
	492	break;
	493	case 'footnote':
	494	if($start){
	495	$this->o('[^'.$attrs['nr']."]:\n".$this->indent.' ');
	496	$this->start = true;
	497	}
	498	$this->indent(' ',$start);
	499	break;
	500	case 'a':
	501	if($start) {
	502	# buffer to check for inline links like <foo@bar.com> and the like
	503	if (isset ($attrs['href'])) {
	504	$this->buffer();
	505	array_push($this->astack, $attrs);
	506	} else {
	507	array_push($this->astack, null);
	508	}
	509	} else {
	510	if($this->astack) {
	511	$a = array_pop($this->astack);
	512	if ($a) {
	513	# for emails
	514	$a['href'] = $this->decode($a['href']);
	515	$buffer = $this->unbuffer();
	516	$buffer_check = $this->decode(trim($buffer));
	517	if((substr($a['href'],0,7) == 'mailto:' && 'mailto:'.$buffer_check == $a['href']) \|\| $a['href'] == $buffer_check){
	518	# inline link
	519	$this->out('<'.$buffer_check.'>',true);
	520	} else {
	521	# block link
	522	$this->previousIndex($a);
	523	$this->out('['.$buffer.']['.$a['count'].']',true);
	524	}
	525	}
	526	}
	527	}
	528	break;
	529	# abbrevations
	530	case 'abbr':
	531	case 'acronym':
	532	if($start){
	533	$this->buffer();
	534	array_push($this->abbrs,isset($attrs['title'])?$attrs['title']:'');
	535	} else {
	536	$abbr = $this->unbuffer();
	537	$def = array_pop($this->abbrs);
	538	# only add abbr if its not already defined
	539	if(!isset($this->abbrs[$abbr])){
	540	$this->abbrs[$abbr] = $def;
	541	}
	542	$this->o($abbr);
	543	}
	544	break;
	545	case 'img' :
	546	if ($start) {
	547	if (isset ($attrs['src'])) {
	548	$attrs['href'] = $attrs['src'];
	549	$alt = '';
	550	if (isset ($attrs['alt'])) {
	551	$alt = $attrs['alt'];
	552	} elseif(isset($attrs['title'])){
	553	$alt = $attrs['title'];
	554	}
	555	$this->previousIndex($attrs);
	556	$this->o('!['.$alt.'][' . $attrs['count'] . ']');
	557	}
	558	}
	559	break;
	560	case 'code':
	561	# do we have to keep this tag?
	562	# or is a parent <pre> element existing?
	563	if($this->keep_tag($tag,$attrs,$start) \|\| $this->parent('pre')){
	564	return;
	565	}
	566	# convert to `code` and handle backticks inside code block
	567	# <code>foo`bar</code> has to get ``foo`bar`` and so forth
	568	if($start){
	569	$this->buffer();
	570	} else {
	571	$str = $this->unbuffer();
	572	preg_match_all('#`+#',$str,$matches);
	573	if(!empty($matches[0])){
	574	rsort($matches[0]);
	575	$len = strlen($matches[0][0])+1;
	576	} else {
	577	$len = 1;
	578	}
	579	$ticks = str_repeat('`',$len);
	580	$this->out($ticks.$str.$ticks);
	581	}
	582	break;
	583	case 'dl' :
	584	# note: if <dl> gets parsed, its direct children (<dd> and <dt>) will be parsed as well
	585	if ($start) {
	586	$this->p();
	587	}
	588	break;
	589	case 'dd' :
	590	# is the parent dl parsed?
	591	if(!$this->parent('dl')){
	592	$this->keep_tag($tag,$attrs,$start,true);
	593	return;
	594	}
	595	if ($start) {
	596	$this->o(': ');
	597	$this->start = true;
	598	} else {
	599	$this->outtext .= "\n";
	600	$this->pbr();
	601	}
	602	$this->indent(' ',$start);
	603	break;
	604	case 'dt' :
	605	# is the parent dl parsed?
	606	if(!$this->parent('dl')){
	607	$this->keep_tag($tag,$attrs,$start,true);
	608	return;
	609	}
	610	if (!$start) {
	611	$this->pbr();
	612	}
	613	break;
	614	case 'ol' :
	615	case 'ul' :
	616	# note: if this element gets parsed, its direct children <li>s will be parsed as well
	617	if ($start) {
	618	array_push($this->list, array (
	619	'name' => $tag,
	620	'num' => 0
	621	));
	622	} else {
	623	array_pop($this->list);
	624	$this->pbr();
	625	}
	626	break;
	627	case 'li' :
	628	if ($this->list) {
	629	$li = &$this->list[count($this->list) - 1];
	630	}
	631	# not inside a list or the list tag was not parsed
	632	if(!isset($li) \|\| !$this->parent($li['name'])){
	633	$this->keep_tag($tag,$attrs,$start,true);
	634	return;
	635	}
	636	if ($start) {
	637	$this->pbr();
	638	if($li['name'] == 'ul'){
	639	$this->o('* ');
	640	} else {
	641	$li['num']++;
	642	/**
	643	* @todo line up <ol><li>s > 9 correctly.
	644	*/
	645	$this->o($li['num'].'. ');
	646	}
	647	$this->start = true;
	648	$this->indent(' ',$start);
	649	} else {
	650	$this->indent(' ',$start);
	651	}
	652	break;
	653	case 'table':
	654	# NOTE: if the <table> tag gets parsed, all its children will be as well!
645	655
646		# finally: parse the whole table
647		if(!$start){
648		$this->outtext .= "\n\n";
649		$separator = array();
650		# seperator with correct align identifikators
651		foreach($this->cols as $col => $arr){
652		$this->max_len[$col] = max($arr);
653		$left = $right = '';
654		switch($this->align[$col]){
655		case 'center':
656		$right = ':';
657		case 'left':
658		$left = ':';
659		break;
660		case 'right':
661		$right = ':';
662		break;
663		}
664		array_push($separator,$left.str_repeat('-',$this->max_len[$col]).$right);
665		}
666		$separator = '\| '.implode(' \| ',$separator).' \|';
667		# set equal width
668		array_walk($this->rows,array(&$this,'fill_td'));
669		$rows = $this->rows;
670		foreach($rows as $row => $cols){
671		$this->pbr();
672		$this->o('\| '.implode(' \| ',$cols).' \|');
673		if(in_array($row,$this->header)){
674		$this->pbr();
675		$this->o($separator);
676		}
677		}
678		$this->cols = array();
679		$this->rows = array();
680		$this->align = array();
681		$this->pbr();
682		}
683		break;
684		case 'tr':
685		# not inside a table or the parent table was not parsed
686		if(!$this->parent('table')){
687		$this->keep_tag($tag,$attrs,$start,true);
688		return;
689		}
690		if($start){
691		$this->row++;
692		} else {
693		$this->col = 0;
694		}
695		break;
696		case 'th':
697		# not inside a table or the parent table was not parsed
698		if(!$this->parent('table')){
699		$this->keep_tag($tag,$attrs,$start,true);
700		return;
701		}
702		if($start){
703		if(!in_array($this->row,$this->header)){
704		array_push($this->header,$this->row);
705		}
706		$this->col++;
707		$this->align[$this->col] = !empty($attrs['align']) ? $attrs['align'] : null;
708		}
709		break;
710		case 'td':
711		# not inside a table or the parent table was not parsed
712		if(!$this->parent('table')){
713		$this->keep_tag($tag,$attrs,$start,true);
714		return;
715		}
716		if($start){
717		$this->col++;
718		if(!empty($attrs['align']) && is_null($this->align[$this->col])){
719		$this->align[$this->col] = $attrs['align'];
720		if($attrs['align'] == 'center'){
721		$this->max_len[$this->col] +=2;
722		}
723		}
724		}
725		break;
726		case 'pre':
727		$this->indent(' ',$start,true);
728		if ($start) {
729		$this->pbr();
730		}
731		break;
732		default:
733		$this->keep_tag($tag,$attrs,$start,true);
734		return;
735		}
736		# if we want to keep all non convertible html this function has to know if some parent elemts
737		# were parsed or not (also some elements need to know if)
738		if($start){
739		if(!isset($this->parents[$tag])){
740		$this->parents[$tag] = '1';
741		} else {
742		$this->parents[$tag] .= '1';
743		}
744		} else {
745		if($this->LINKS_EACH_PARAGRAPH && in_array($tag,array('p','ul','blockquote','ol','dl','table','h1','h2','h3','h4','h5','h6'))){
746		$this->links();
747		}
748		$this->parents[$tag] = substr($this->parents[$tag],0,-1);
749		}
750		return;
751		}
752		/**
753		* adds a string to the output ($this->outtext)
754		* also copes with tables
755		*
756		* @param string $str
757		* @return void
758		*/
759		function out($str) {
760		# buffering
761		if($this->buffer_lvl){
762		$this->buffer[$this->buffer_lvl] .= $str;
763		return;
764		}
765		# this is for tables (see php markdown extra by michel fortin)
766		if(($this->parent('th') \|\| $this->parent('td'))){
767		$str = trim($str);
768		if(!isset($this->rows[$this->row][$this->col])){
769		$this->rows[$this->row][$this->col] = $str;
770		} else {
771		$this->rows[$this->row][$this->col] .= $str;
772		}
773		if(!isset($this->cols[$this->col][$this->row])){
774		$this->cols[$this->col][$this->row] = strlen($str);
775		} else {
776		$this->cols[$this->col][$this->row] += strlen($str);
777		}
778		return;
779		}
780		$this->outtext .= $str;
781		}
782		/**
783		* further parse the output and add newlines, remove whitespaces and such
784		*
785		* @param string $data
786		* @param bool $puredata
787		* @param string $force
788		* @return void
789		*/
790		function o($data, $puredata = false, $force = false) {
791		if($this->parent('table') && trim($data) == ''){ # drop whitespaces inside tables
792		return;
793		} elseif ($puredata && !$this->parent('code','both') && !$this->parent('pre','both')) { # keep whitespace for code
794		$data = preg_replace('#\s+#', ' ', $data);
795		}
796		if (!$data && !$force) {
797		return;
798		}
799		if (!empty($this->indent)) {
800		$data = str_replace("\n", "\n".$this->indent, $data);
801		}
802		if ($this->start) {
803		if($data == ' '){
804		return;
805		}
806		$this->p_p = 0;
807		$this->start = 0;
808		}
809		if ($force == 'end') {
810		# It's the end.
811		$this->p_p = 0;
812		$this->out("\n");
813		}
814		if ($this->p_p) {
815		if($data == ' '){
816		return;
817		}
818		$data = ltrim($data);
819		$this->out(str_repeat("\n".$this->indent, $this->p_p));
820		}
821		$this->p_p = 0;
822		$this->out($data);
823		if($data){
824		$this->lastWasNL = substr($data, -1) == "\n";
825		}
826		}
827		/**
828		* display block links after paragraph etc.
829		* also handle abbrs
830		*
831		* @param void
832		* @return void
833		*/
834		function links(){
835		$this->abbrs();
836		if(empty($this->a)){
837		return; # no links stored
838		}
839		$pre = '';
840		$this->out("\n\n");
841		foreach($this->a as $links){
842		/**
843		* @todo base href
844		*/
845		foreach($links as $link){
846		$a = $pre.' [' . $link['count'] . ']: ' . $link['href'];
847		if (isset ($link['title'])) {
848		$a .= ' (' . $link['title'] . ')';
849		}
850		$this->out($a."\n");
851		}
852		}
853		$this->a = array();
854		$this->out("\n");
855		$this->lastWasNL = true;
856		}
857		/**
858		* display abbr list
859		*
860		* @param void
861		* @return void
862		*/
863		function abbrs(){
864		if(empty($this->abbrs)){
865		return; # no abbrs stored
866		}
867		$this->out("\n\n");
868		foreach($this->abbrs as $abbr => $def){
869		$this->out('*['.$abbr.']: '.$def."\n");
870		}
871		$this->abbrs = array();
872		$this->out("\n");
873		$this->lastWasNL = true;
874		}
875		/**
876		* if the link is already set use its count, else increase acount
877		*
878		* @param array &$attrs link attributes
879		* @return void
880		*/
881		function previousIndex(&$attrs) {
882		# check for existing link
883		if(isset($this->a[$attrs['href']])){
884		foreach($this->a[$attrs['href']] as $a){
885		if (!empty($attrs['title']) \|\| !empty($a['title'])){
886		if($a['title'] == $attrs['title']) {
887		$attrs = $a;
888		return;
889		}
890		} else {
891		$attrs = $a;
892		return;
893		}
894		}
895		}
896		# if we come here, no matching link was found
897		$this->acount++;
898		$attrs['count'] = $this->acount;
899		if(isset($this->a[$attrs['href']])){
900		array_push($this->a[$attrs['href']],$attrs);
901		} else {
902		$this->a[$attrs['href']] = array($attrs);
903		}
904		}
905		/**
906		* handles bad html to avoid xml parse errors
907		*
908		* @param string $html
909		* @return string
910		*/
911		function handle_bad_html($html){
912		return preg_replace_callback('#<([a-z1-6]+)( [^>])?>(.(?R).*)</\\1>#Us',array(&$this,'replace_bad_html'),$html);
913		}
914		/**
915		* callback function which is used in handle_bad_html()
916		*
917		* @param array $matches
918		* @return string
919		*/
920		function replace_bad_html($matches){
921		# recursion
922		$matches[3] = $this->handle_bad_html($matches[3]);
923		return '<'.$matches[1].$matches[2].'>'.$matches[3].'</'.$matches[1].'>';
924		}
925		/**
926		* if the option BODY_WIDTH is set, this option will wrap text to the
927		* provided width
928		*
929		* @param string $text
930		* @return string
931		*
932		* @todo wrapping of code (also kept code blocks)
933		*/
934		function optwrap($text) {
935		if ($this->BODY_WIDTH < 30) {
936		return $text;
937		}
938		$result = '';
939		$split = explode("\n", $text);
940		foreach ($split as $para) {
941		if (strlen($para) > 0) {
942		if (preg_match('#^(\s*): #',$para,$indent)) { # definition lists
943		$indent = isset($indent[1]) ? $indent[1] : '';
944		$result .= wordwrap($para, $this->BODY_WIDTH - strlen($indent) - 4, "\n".$indent.' ')."\n";
945		} elseif(preg_match('#^(\s*>+)#',$para,$indent)){ # blockquote
946		$result .= wordwrap($para,$this->BODY_WIDTH - (strlen($indent[0])+1),"\n".$indent[0].' ')."\n";
947		} elseif(preg_match('#^\s*\\|#',$para)){ # table
948		$result .= $para."\n"; # dont wrap
949		} elseif(preg_match('#^(\s)\#',$para,$indent)) { # list item @todo: ol
950		$indent = isset($indent[1]) ? $indent[1] : '';
951		$indent.= ' ';
952		$result .= wordwrap($para,$this->BODY_WIDTH - strlen($indent),"\n".$indent). "\n";
953		} elseif(preg_match('#^ \[[^\]]+\]:#',$para)){ # block links
954		# don't wrap at the moment
955		$result .= $para."\n";
956		continue;
957		} else { # something else
958		preg_match('#^\s+#',$para,$indent);
959		$indent = isset($indent[0]) ? $indent[0] : '';
960		$result .= wordwrap($para,$this->BODY_WIDTH - strlen($indent),"\n".$indent). "\n";
961		}
962		} else {
963		$result .= "\n";
964		}
965		}
966		return $result;
967		}
968		/**
969		* handles html tags which are not represented by the parser logic
970		* if $this->KEEP_HTML is set to true, the tag will be appended to the
971		* output and `markdown="1"` added to its attributes
972		*
973		* @param string $tag
974		* @param array $attrs
975		* @param bool $start
976		* @param array $known_attrs these attrs can be handled by markdown
977		* @return bool
978		*/
979		function keep_tag($tag,$attrs,$start,$force = false){
980		if(!$force && !$this->KEEP_HTML){
981		return false;
982		}
	656	# finally: parse the whole table
	657	if(!$start){
	658	$this->outtext .= "\n\n";
	659	$separator = array();
	660	# seperator with correct align identifikators
	661	foreach($this->cols as $col => $arr){
	662	$this->max_len[$col] = max($arr);
	663	$left = $right = '';
	664	switch($this->align[$col]){
	665	case 'center':
	666	$right = ':';
	667	case 'left':
	668	$left = ':';
	669	break;
	670	case 'right':
	671	$right = ':';
	672	break;
	673	}
	674	array_push($separator,$left.str_repeat('-',$this->max_len[$col]).$right);
	675	}
	676	$separator = '\| '.implode(' \| ',$separator).' \|';
	677	# set equal width
	678	array_walk($this->rows,array(&$this,'fill_td'));
	679	$rows = $this->rows;
	680	foreach($rows as $row => $cols){
	681	$this->pbr();
	682	$this->o('\| '.implode(' \| ',$cols).' \|');
	683	if(in_array($row,$this->header)){
	684	$this->pbr();
	685	$this->o($separator);
	686	}
	687	}
	688	$this->cols = array();
	689	$this->rows = array();
	690	$this->align = array();
	691	$this->pbr();
	692	}
	693	break;
	694	case 'tr':
	695	# not inside a table or the parent table was not parsed
	696	if(!$this->parent('table')){
	697	$this->keep_tag($tag,$attrs,$start,true);
	698	return;
	699	}
	700	if($start){
	701	$this->row++;
	702	} else {
	703	$this->col = 0;
	704	}
	705	break;
	706	case 'th':
	707	# not inside a table or the parent table was not parsed
	708	if(!$this->parent('table')){
	709	$this->keep_tag($tag,$attrs,$start,true);
	710	return;
	711	}
	712	if($start){
	713	if(!in_array($this->row,$this->header)){
	714	array_push($this->header,$this->row);
	715	}
	716	$this->col++;
	717	$this->align[$this->col] = !empty($attrs['align']) ? $attrs['align'] : null;
	718	}
	719	break;
	720	case 'td':
	721	# not inside a table or the parent table was not parsed
	722	if(!$this->parent('table')){
	723	$this->keep_tag($tag,$attrs,$start,true);
	724	return;
	725	}
	726	if($start){
	727	$this->col++;
	728	if(!empty($attrs['align']) && is_null($this->align[$this->col])){
	729	$this->align[$this->col] = $attrs['align'];
	730	if($attrs['align'] == 'center'){
	731	$this->max_len[$this->col] +=2;
	732	}
	733	}
	734	}
	735	break;
	736	case 'pre':
	737	$this->indent(' ',$start,true);
	738	if ($start) {
	739	$this->pbr();
	740	}
	741	break;
	742	default:
	743	$this->keep_tag($tag,$attrs,$start,true);
	744	return;
	745	}
	746	# if we want to keep all non convertible html this function has to know if some parent elemts
	747	# were parsed or not (also some elements need to know if)
	748	if($start){
	749	if(!isset($this->parents[$tag])){
	750	$this->parents[$tag] = '1';
	751	} else {
	752	$this->parents[$tag] .= '1';
	753	}
	754	} else {
	755	if($this->LINKS_EACH_PARAGRAPH && in_array($tag,array('p','ul','blockquote','ol','dl','table','h1','h2','h3','h4','h5','h6'))){
	756	$this->links();
	757	}
	758	$this->parents[$tag] = substr($this->parents[$tag],0,-1);
	759	}
	760	return;
	761	}
	762	/**
	763	* adds a string to the output ($this->outtext)
	764	* also copes with tables
	765	*
	766	* @param string $str
	767	* @return void
	768	*/
	769	function out($str) {
	770	# buffering
	771	if($this->buffer_lvl){
	772	$this->buffer[$this->buffer_lvl] .= $str;
	773	return;
	774	}
	775	# this is for tables (see php markdown extra by michel fortin)
	776	if(($this->parent('th') \|\| $this->parent('td'))){
	777	$str = trim($str);
	778	if(!isset($this->rows[$this->row][$this->col])){
	779	$this->rows[$this->row][$this->col] = $str;
	780	} else {
	781	$this->rows[$this->row][$this->col] .= $str;
	782	}
	783	if(!isset($this->cols[$this->col][$this->row])){
	784	$this->cols[$this->col][$this->row] = strlen($str);
	785	} else {
	786	$this->cols[$this->col][$this->row] += strlen($str);
	787	}
	788	return;
	789	}
	790	$this->outtext .= $str;
	791	}
	792	/**
	793	* further parse the output and add newlines, remove whitespaces and such
	794	*
	795	* @param string $data
	796	* @param bool $puredata
	797	* @param string $force
	798	* @return void
	799	*/
	800	function o($data, $puredata = false, $force = false) {
	801	if($this->parent('table') && trim($data) == ''){ # drop whitespaces inside tables
	802	return;
	803	} elseif ($puredata && !$this->parent('code','both') && !$this->parent('pre','both')) { # keep whitespace for code
	804	$data = preg_replace('#\s+#', ' ', $data);
	805	}
	806	if (!$data && !$force) {
	807	return;
	808	}
	809	if (!empty($this->indent)) {
	810	$data = str_replace("\n", "\n".$this->indent, $data);
	811	}
	812	if ($this->start) {
	813	if($data == ' '){
	814	return;
	815	}
	816	$this->p_p = 0;
	817	$this->start = 0;
	818	}
	819	if ($force == 'end') {
	820	# It's the end.
	821	$this->p_p = 0;
	822	$this->out("\n");
	823	}
	824	if ($this->p_p) {
	825	if($data == ' '){
	826	return;
	827	}
	828	$data = ltrim($data);
	829	$this->out(str_repeat("\n".$this->indent, $this->p_p));
	830	}
	831	$this->p_p = 0;
	832	$this->out($data);
	833	if($data){
	834	$this->lastWasNL = substr($data, -1) == "\n";
	835	}
	836	}
	837	/**
	838	* display block links after paragraph etc.
	839	* also handle abbrs
	840	*
	841	* @param void
	842	* @return void
	843	*/
	844	function links(){
	845	$this->abbrs();
	846	if(empty($this->a)){
	847	return; # no links stored
	848	}
	849	$pre = '';
	850	$this->out("\n\n");
	851	foreach($this->a as $links){
	852	/**
	853	* @todo base href
	854	*/
	855	foreach($links as $link){
	856	$a = $pre.' [' . $link['count'] . ']: ' . $link['href'];
	857	if (isset ($link['title'])) {
	858	$a .= ' (' . $link['title'] . ')';
	859	}
	860	$this->out($a."\n");
	861	}
	862	}
	863	$this->a = array();
	864	$this->out("\n");
	865	$this->lastWasNL = true;
	866	}
	867	/**
	868	* display abbr list
	869	*
	870	* @param void
	871	* @return void
	872	*/
	873	function abbrs(){
	874	if(empty($this->abbrs)){
	875	return; # no abbrs stored
	876	}
	877	$this->out("\n\n");
	878	foreach($this->abbrs as $abbr => $def){
	879	$this->out('*['.$abbr.']: '.$def."\n");
	880	}
	881	$this->abbrs = array();
	882	$this->out("\n");
	883	$this->lastWasNL = true;
	884	}
	885	/**
	886	* if the link is already set use its count, else increase acount
	887	*
	888	* @param array &$attrs link attributes
	889	* @return void
	890	*/
	891	function previousIndex(&$attrs) {
	892	# check for existing link
	893	if(isset($this->a[$attrs['href']])){
	894	foreach($this->a[$attrs['href']] as $a){
	895	if (!empty($attrs['title']) \|\| !empty($a['title'])){
	896	if($a['title'] == $attrs['title']) {
	897	$attrs = $a;
	898	return;
	899	}
	900	} else {
	901	$attrs = $a;
	902	return;
	903	}
	904	}
	905	}
	906	# if we come here, no matching link was found
	907	$this->acount++;
	908	$attrs['count'] = $this->acount;
	909	if(isset($this->a[$attrs['href']])){
	910	array_push($this->a[$attrs['href']],$attrs);
	911	} else {
	912	$this->a[$attrs['href']] = array($attrs);
	913	}
	914	}
	915	/**
	916	* handles bad html to avoid xml parse errors
	917	*
	918	* @param string $html
	919	* @return string
	920	*/
	921	function handle_bad_html($html){
	922	return preg_replace_callback('#<([a-z1-6]+)( [^>])?>(.(?R).*)</\\1>#Us',array(&$this,'replace_bad_html'),$html);
	923	}
	924	/**
	925	* callback function which is used in handle_bad_html()
	926	*
	927	* @param array $matches
	928	* @return string
	929	*/
	930	function replace_bad_html($matches){
	931	# recursion
	932	$matches[3] = $this->handle_bad_html($matches[3]);
	933	return '<'.$matches[1].$matches[2].'>'.$matches[3].'</'.$matches[1].'>';
	934	}
	935	/**
	936	* if the option BODY_WIDTH is set, this option will wrap text to the
	937	* provided width
	938	*
	939	* @param string $text
	940	* @return string
	941	*
	942	* @todo wrapping of code (also kept code blocks)
	943	*/
	944	function optwrap($text) {
	945	if ($this->BODY_WIDTH < 30) {
	946	return $text;
	947	}
	948	$result = '';
	949	$split = explode("\n", $text);
	950	foreach ($split as $para) {
	951	if (strlen($para) > 0) {
	952	if (preg_match('#^(\s*): #',$para,$indent)) { # definition lists
	953	$indent = isset($indent[1]) ? $indent[1] : '';
	954	$result .= wordwrap($para, $this->BODY_WIDTH - strlen($indent) - 4, "\n".$indent.' ')."\n";
	955	} elseif(preg_match('#^(\s*>+)#',$para,$indent)){ # blockquote
	956	$result .= wordwrap($para,$this->BODY_WIDTH - (strlen($indent[0])+1),"\n".$indent[0].' ')."\n";
	957	} elseif(preg_match('#^\s*\\|#',$para)){ # table
	958	$result .= $para."\n"; # dont wrap
	959	} elseif(preg_match('#^(\s)\#',$para,$indent)) { # list item @todo: ol
	960	$indent = isset($indent[1]) ? $indent[1] : '';
	961	$indent.= ' ';
	962	$result .= wordwrap($para,$this->BODY_WIDTH - strlen($indent),"\n".$indent). "\n";
	963	} elseif(preg_match('#^ \[[^\]]+\]:#',$para)){ # block links
	964	# don't wrap at the moment
	965	$result .= $para."\n";
	966	continue;
	967	} else { # something else
	968	preg_match('#^\s+#',$para,$indent);
	969	$indent = isset($indent[0]) ? $indent[0] : '';
	970	$result .= wordwrap($para,$this->BODY_WIDTH - strlen($indent),"\n".$indent). "\n";
	971	}
	972	} else {
	973	$result .= "\n";
	974	}
	975	}
	976	return $result;
	977	}
	978	/**
	979	* handles html tags which are not represented by the parser logic
	980	* if $this->KEEP_HTML is set to true, the tag will be appended to the
	981	* output and `markdown="1"` added to its attributes
	982	*
	983	* @param string $tag
	984	* @param array $attrs
	985	* @param bool $start
	986	* @param array $known_attrs these attrs can be handled by markdown
	987	* @return bool
	988	*/
	989	function keep_tag($tag,$attrs,$start,$force = false){
	990	if(!$force && !$this->KEEP_HTML){
	991	return false;
	992	}
983	993
984		# start tag
985		if($start){
986		# if there is a attr which cannot be handled by markdown
987		# this tag will be kept.
988		if(isset($this->has_attrs[$tag])){
989		$known_attrs = $this->has_attrs[$tag];
990		} else {
991		$known_attrs = array();
992		}
993		if(!$force && count($known_attrs) >= count($attrs)){
994		if(empty($attrs) \|\| count(array_diff(array_keys($attrs),$known_attrs)) == 0){
995		# tag can be handled by markdown!
996		return false;
997		}
998		}
999		$attr = '';
1000		if(!empty($attrs)){
1001		foreach($attrs as $key => $value){
1002		if($key == 'forcehtml'){
1003		continue;
1004		}
1005		$attr.=' '.$key.'="'.$value.'"';
1006		}
1007		}
1008		if(!$force && in_array($tag,array('div','center','li','dt','dd'))){
1009		$attr.= ' markdown="1"';
1010		} elseif(!$this->force_html) {
1011		$this->force_html($tag);
1012		}
1013		$this->o('<'.$tag.$attr.'>',true);
1014		# add to list of parents:
1015		if(isset($this->parents[$tag])){
1016		$this->parents[$tag] .= '2';
1017		} else {
1018		$this->parents[$tag] = '2';
1019		}
1020		# close tag
1021		} else {
1022		if(!$force && !$this->parent($tag,'kept')){
1023		# the start tag of this element was not parsed
1024		return false;
1025		}
1026		$this->o('</'.$tag.'>');
1027		$this->parents[$tag] = substr($this->parents[$tag],0,-1);
1028		# newlines after </tag>
1029		if(in_array($tag,array('th','td','dt','dd','li','p'))){
1030		$this->o("\n");
1031		}
1032		}
1033		# newlines after <tag> and </tag>
1034		if(in_array($tag,array('div','center','table','tr','ul','ol','dl','pre'))){
1035		$this->o("\n");
1036		}
1037		return true;
1038		}
1039		/**
1040		* outputs a cell widened to the proper width
1041		*
1042		* @param array &$row
1043		* @return void
1044		*/
1045		function fill_td(&$row){
1046		$len = 0;
1047		foreach($row as $col => $cont){
1048		$width = $this->max_len[$col];
1049		switch($this->align[$col]){
1050		case 'center':
1051		$width += 2;
1052		$row[$col] = str_pad($row[$col],$width,' ',STR_PAD_BOTH);
1053		break;
1054		case 'left':
1055		$width++;
1056		default:
1057		$row[$col] = str_pad($row[$col],$width,' ');
1058		break;
1059		case 'right':
1060		$width++;
1061		$row[$col] = str_pad($row[$col],$width,' ',STR_PAD_LEFT);
1062		break;
1063		}
1064		}
1065		}
1066		/**
1067		* some sort of <br />
1068		*
1069		* @param void
1070		* @return void
1071		*/
1072		function pbr() {
1073		if ($this->p_p == 0) {
1074		$this->p_p = 1;
1075		}
1076		}
1077		/**
1078		* text <p> (e.g. newlines after output)
1079		*
1080		* @param void
1081		* @return void
1082		*/
1083		function p() {
1084		if($this->parent('table')){
1085		return;
1086		}
1087		$this->p_p = 2;
1088		}
1089		/**
1090		* add $indent before each line
1091		*
1092		* @param string $indent
1093		* @param bool $start wether it's an opening tag or a closing one
1094		* @param bool $output shall $indent be outputted? (only if $start is true)
1095		* @return void
1096		*/
1097		function indent($indent,$start,$output=false){
1098		if($start){
1099		if($output){
1100		$this->o($indent);
1101		}
1102		$this->indent .= $indent;
1103		} else {
1104		$len = strlen($indent);
1105		if($len >= strlen($this->indent)){
1106		$this->indent = '';
1107		} else {
1108		$this->indent = substr($this->indent,0,-$len);
1109		}
1110		}
1111		}
1112		/**
1113		* checks if a parent element exists
1114		* use $type to check for a parsed parent element or a kept element
1115		* @param string $parent name of the parent tag
1116		* @param string $type either 'parsed' or 'kept' or 'both'
1117		* @return bool
1118		*/
1119		function parent($parent,$type = 'parsed'){
1120		if(!isset($this->parents[$parent])){
1121		return false;
1122		}
1123		if($type != 'both'){
1124		$type = $type == 'parsed' ? '1' : '2';
1125		return substr($this->parents[$parent],-1) === $type;
1126		} else {
1127		return !empty($this->parents[$parent]);
1128		}
1129		}
1130		/**
1131		* start buffer
1132		*
1133		* @param void
1134		* @return void
1135		*/
1136		function buffer(){
1137		if($this->p_p){
1138		$this->out(str_repeat("\n".$this->indent, $this->p_p));
1139		$this->p_p = 0;
1140		}
1141		$this->buffer_lvl++;
1142		$this->buffer[$this->buffer_lvl] = '';
1143		}
1144		/**
1145		* end buffer and return buffered output
1146		*
1147		* @param void
1148		* @return string
1149		*/
1150		function unbuffer(){
1151		$out = $this->buffer[$this->buffer_lvl];
1152		unset($this->buffer[$this->buffer_lvl]);
1153		$this->buffer_lvl--;
1154		return $out;
1155		}
1156		/**
1157		* decode email
1158		*
1159		* @author derernst@gmx.ch <http://www.php.net/manual/en/function.html-entity-decode.php#68536>
1160		*/
1161		function decode($text,$quote_style = ENT_NOQUOTES){
1162		if (function_exists('html_entity_decode')) {
1163		$text = html_entity_decode($text, $quote_style, 'ISO-8859-1'); // NOTE: UTF-8 does not work!
1164		}
1165		else {
1166		$trans_tbl = get_html_translation_table(HTML_ENTITIES, $quote_style);
1167		$trans_tbl = array_flip($trans_tbl);
1168		$text = strtr($text, $trans_tbl);
1169		}
1170		$text = preg_replace('~&#x([0-9a-f]+);~ei', 'chr(hexdec("\\1"))', $text);
1171		$text = preg_replace('~&#([0-9]+);~e', 'chr("\\1")', $text);
1172		return $text;
1173		}
	994	# start tag
	995	if($start){
	996	# if there is a attr which cannot be handled by markdown
	997	# this tag will be kept.
	998	if(isset($this->has_attrs[$tag])){
	999	$known_attrs = $this->has_attrs[$tag];
	1000	} else {
	1001	$known_attrs = array();
	1002	}
	1003	if(!$force && count($known_attrs) >= count($attrs)){
	1004	if(empty($attrs) \|\| count(array_diff(array_keys($attrs),$known_attrs)) == 0){
	1005	# tag can be handled by markdown!
	1006	return false;
	1007	}
	1008	}
	1009	$attr = '';
	1010	if(!empty($attrs)){
	1011	foreach($attrs as $key => $value){
	1012	if($key == 'forcehtml'){
	1013	continue;
	1014	}
	1015	$attr.=' '.$key.'="'.$value.'"';
	1016	}
	1017	}
	1018	if(!$force && in_array($tag,array('div','center','li','dt','dd'))){
	1019	$attr.= ' markdown="1"';
	1020	} elseif(!$this->force_html) {
	1021	$this->force_html($tag);
	1022	}
	1023	$this->o('<'.$tag.$attr.'>',true);
	1024	# add to list of parents:
	1025	if(isset($this->parents[$tag])){
	1026	$this->parents[$tag] .= '2';
	1027	} else {
	1028	$this->parents[$tag] = '2';
	1029	}
	1030	# close tag
	1031	} else {
	1032	if(!$force && !$this->parent($tag,'kept')){
	1033	# the start tag of this element was not parsed
	1034	return false;
	1035	}
	1036	$this->o('</'.$tag.'>');
	1037	$this->parents[$tag] = substr($this->parents[$tag],0,-1);
	1038	# newlines after </tag>
	1039	if(in_array($tag,array('th','td','dt','dd','li','p'))){
	1040	$this->o("\n");
	1041	}
	1042	}
	1043	# newlines after <tag> and </tag>
	1044	if(in_array($tag,array('div','center','table','tr','ul','ol','dl','pre'))){
	1045	$this->o("\n");
	1046	}
	1047	return true;
	1048	}
	1049	/**
	1050	* outputs a cell widened to the proper width
	1051	*
	1052	* @param array &$row
	1053	* @return void
	1054	*/
	1055	function fill_td(&$row){
	1056	$len = 0;
	1057	foreach($row as $col => $cont){
	1058	$width = $this->max_len[$col];
	1059	switch($this->align[$col]){
	1060	case 'center':
	1061	$width += 2;
	1062	$row[$col] = str_pad($row[$col],$width,' ',STR_PAD_BOTH);
	1063	break;
	1064	case 'left':
	1065	$width++;
	1066	default:
	1067	$row[$col] = str_pad($row[$col],$width,' ');
	1068	break;
	1069	case 'right':
	1070	$width++;
	1071	$row[$col] = str_pad($row[$col],$width,' ',STR_PAD_LEFT);
	1072	break;
	1073	}
	1074	}
	1075	}
	1076	/**
	1077	* some sort of <br />
	1078	*
	1079	* @param void
	1080	* @return void
	1081	*/
	1082	function pbr() {
	1083	if ($this->p_p == 0) {
	1084	$this->p_p = 1;
	1085	}
	1086	}
	1087	/**
	1088	* text <p> (e.g. newlines after output)
	1089	*
	1090	* @param void
	1091	* @return void
	1092	*/
	1093	function p() {
	1094	if($this->parent('table')){
	1095	return;
	1096	}
	1097	$this->p_p = 2;
	1098	}
	1099	/**
	1100	* add $indent before each line
	1101	*
	1102	* @param string $indent
	1103	* @param bool $start wether it's an opening tag or a closing one
	1104	* @param bool $output shall $indent be outputted? (only if $start is true)
	1105	* @return void
	1106	*/
	1107	function indent($indent,$start,$output=false){
	1108	if($start){
	1109	if($output){
	1110	$this->o($indent);
	1111	}
	1112	$this->indent .= $indent;
	1113	} else {
	1114	$len = strlen($indent);
	1115	if($len >= strlen($this->indent)){
	1116	$this->indent = '';
	1117	} else {
	1118	$this->indent = substr($this->indent,0,-$len);
	1119	}
	1120	}
	1121	}
	1122	/**
	1123	* checks if a parent element exists
	1124	* use $type to check for a parsed parent element or a kept element
	1125	* @param string $parent name of the parent tag
	1126	* @param string $type either 'parsed' or 'kept' or 'both'
	1127	* @return bool
	1128	*/
	1129	function parent($parent,$type = 'parsed'){
	1130	if(!isset($this->parents[$parent])){
	1131	return false;
	1132	}
	1133	if($type != 'both'){
	1134	$type = $type == 'parsed' ? '1' : '2';
	1135	return substr($this->parents[$parent],-1) === $type;
	1136	} else {
	1137	return !empty($this->parents[$parent]);
	1138	}
	1139	}
	1140	/**
	1141	* start buffer
	1142	*
	1143	* @param void
	1144	* @return void
	1145	*/
	1146	function buffer(){
	1147	if($this->p_p){
	1148	$this->out(str_repeat("\n".$this->indent, $this->p_p));
	1149	$this->p_p = 0;
	1150	}
	1151	$this->buffer_lvl++;
	1152	$this->buffer[$this->buffer_lvl] = '';
	1153	}
	1154	/**
	1155	* end buffer and return buffered output
	1156	*
	1157	* @param void
	1158	* @return string
	1159	*/
	1160	function unbuffer(){
	1161	$out = $this->buffer[$this->buffer_lvl];
	1162	unset($this->buffer[$this->buffer_lvl]);
	1163	$this->buffer_lvl--;
	1164	return $out;
	1165	}
	1166	/**
	1167	* decode email
	1168	*
	1169	* @author derernst@gmx.ch <http://www.php.net/manual/en/function.html-entity-decode.php#68536>
	1170	*/
	1171	function decode($text,$quote_style = ENT_NOQUOTES){
	1172	if (function_exists('html_entity_decode')) {
	1173	$text = html_entity_decode($text, $quote_style, 'ISO-8859-1'); // NOTE: UTF-8 does not work!
	1174	}
	1175	else {
	1176	$trans_tbl = get_html_translation_table(HTML_ENTITIES, $quote_style);
	1177	$trans_tbl = array_flip($trans_tbl);
	1178	$text = strtr($text, $trans_tbl);
	1179	}
	1180	$text = preg_replace('~&#x([0-9a-f]+);~ei', 'chr(hexdec("\\1"))', $text);
	1181	$text = preg_replace('~&#([0-9]+);~e', 'chr("\\1")', $text);
	1182	return $text;
	1183	}
1174	1184	}
1175	1185	?>

Navigation

Changeset 222

Legend:

trunk/lib/Ak.php

trunk/vendor/TextParsers/html2text.php

Download in other formats: