aboutsummaryrefslogblamecommitdiffhomepage
path: root/inc/Readability.php
blob: c50bf2efa1911acd3129e69cd1648e4584fd9d4b (plain) (tree)
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
















































































                                                                                              
                                                                                                       


























































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































                                                                                                                                                                                                                                                                                                                                         
<?php
/** 
* Arc90's Readability ported to PHP for FiveFilters.org
* Based on readability.js version 1.7.1 (without multi-page support)
* ------------------------------------------------------
* Original URL: http://lab.arc90.com/experiments/readability/js/readability.js
* Arc90's project URL: http://lab.arc90.com/experiments/readability/
* JS Source: http://code.google.com/p/arc90labs-readability
* Ported by: Keyvan Minoukadeh, http://www.keyvan.net
* More information: http://fivefilters.org/content-only/
* License: Apache License, Version 2.0
* Requires: PHP5
* Date: 2011-07-22
* 
* Differences between the PHP port and the original
* ------------------------------------------------------
* Arc90's Readability is designed to run in the browser. It works on the DOM 
* tree (the parsed HTML) after the page's CSS styles have been applied and 
* Javascript code executed. This PHP port does not run inside a browser. 
* We use PHP's ability to parse HTML to build our DOM tree, but we cannot 
* rely on CSS or Javascript support. As such, the results will not always 
* match Arc90's Readability. (For example, if a web page contains CSS style 
* rules or Javascript code which hide certain HTML elements from display, 
* Arc90's Readability will dismiss those from consideration but our PHP port, 
* unable to understand CSS or Javascript, will not know any better.)
* 
* Another significant difference is that the aim of Arc90's Readability is 
* to re-present the main content block of a given web page so users can 
* read it more easily in their browsers. Correct identification, clean up, 
* and separation of the content block is only a part of this process. 
* This PHP port is only concerned with this part, it does not include code 
* that relates to presentation in the browser - Arc90 already do 
* that extremely well, and for PDF output there's FiveFilters.org's 
* PDF Newspaper: http://fivefilters.org/pdf-newspaper/.
* 
* Finally, this class contains methods that might be useful for developers 
* working on HTML document fragments. So without deviating too much from 
* the original code (which I don't want to do because it makes debugging 
* and updating more difficult), I've tried to make it a little more 
* developer friendly. You should be able to use the methods here on 
* existing DOMElement objects without passing an entire HTML document to 
* be parsed.
*/

// This class allows us to do JavaScript like assignements to innerHTML
require_once(dirname(__FILE__).'/JSLikeHTMLElement.php');

// Alternative usage (for testing only!)
// uncomment the lines below and call Readability.php in your browser 
// passing it the URL of the page you'd like content from, e.g.:
// Readability.php?url=http://medialens.org/alerts/09/090615_the_guardian_climate.php

/*
if (!isset($_GET['url']) || $_GET['url'] == '') {
	die('Please pass a URL to the script. E.g. Readability.php?url=bla.com/story.html');
}
$url = $_GET['url'];
if (!preg_match('!^https?://!i', $url)) $url = 'http://'.$url;
$html = file_get_contents($url);
$r = new Readability($html, $url);
$r->init();
echo $r->articleContent->innerHTML;
*/

	
class Readability
{
	/* constants */
	const FLAG_STRIP_UNLIKELYS = 1;
	const FLAG_WEIGHT_CLASSES = 2;
	const FLAG_CLEAN_CONDITIONALLY = 4;

	public $version = '1.7.1-without-multi-page';
	public $convertLinksToFootnotes = false;
	public $revertForcedParagraphElements = true;
	public $articleTitle;
	public $articleContent;
	public $dom;
	public $url = null; // optional - URL where HTML was retrieved
	public $debug = false;
	protected $body = null; // 
	protected $bodyCache = null; // Cache the body HTML in case we need to re-use it later
	protected $flags = self::FLAG_CLEAN_CONDITIONALLY; // 1 | 2 | 4;   // Start with all flags set.
	protected $success = false; // indicates whether we were able to extract or not
	
	/**
	* All of the regular expressions in use within readability.
	* Defined up here so we don't instantiate them repeatedly in loops.
	**/
	public $regexps = array(
		'unlikelyCandidates' => '/combx|comment|comments|community|disqus|extra|foot|header|menu|remark|rss|shoutbox|sidebar|sponsor|ad-break|agegate|pagination|pager|popup|tweet|twitter/i',
		'okMaybeItsACandidate' => '/and|article|body|column|main|shadow/i',
		'positive' => '/article|body|content|entry|hentry|main|page|pagination|post|text|blog|story/i',
		'negative' => '/combx|comment|comments|com-|contact|foot|footer|footnote|masthead|media|meta|outbrain|promo|related|scroll|shoutbox|sidebar|sponsor|shopping|tags|tool|widget/i',
		'divToPElements' => '/<(a|blockquote|dl|div|ol|p|pre|table|ul)/i',
		'replaceBrs' => '/(<br[^>]*>[ \n\r\t]*){2,}/i',
		'replaceFonts' => '/<(\/?)font[^>]*>/i',
		// 'trimRe' => '/^\s+|\s+$/g', // PHP has trim()
		'normalize' => '/\s{2,}/',
		'killBreaks' => '/(<br\s*\/?>(\s|&nbsp;?)*){1,}/',
		'video' => '/http:\/\/(www\.)?(youtube|vimeo|dailymotion)\.com/i',
		'skipFootnoteLink' => '/^\s*(\[?[a-z0-9]{1,2}\]?|^|edit|citation needed)\s*$/i'
	);	
	
	/**
	* Create instance of Readability
	* @param string UTF-8 encoded string
	* @param string (optional) URL associated with HTML (used for footnotes)
	*/	
	function __construct($html, $url=null)
	{
		/* Turn all double br's into p's */
		$html = preg_replace($this->regexps['replaceBrs'], '</p><p>', $html);
		$html = preg_replace($this->regexps['replaceFonts'], '<$1span>', $html);
		$html = mb_convert_encoding($html, 'HTML-ENTITIES', "UTF-8");
		$this->dom = new DOMDocument();
		$this->dom->preserveWhiteSpace = false;
		$this->dom->registerNodeClass('DOMElement', 'JSLikeHTMLElement');
		if (trim($html) == '') $html = '<html></html>';
		@$this->dom->loadHTML($html);
		$this->url = $url;
	}

	/**
	* Get article title element
	* @return DOMElement
	*/
	public function getTitle() {
		return $this->articleTitle;
	}
	
	/**
	* Get article content element
	* @return DOMElement
	*/
	public function getContent() {
		return $this->articleContent;
	}	
	
	/**
	* Runs readability.
	* 
	* Workflow:
	*  1. Prep the document by removing script tags, css, etc.
	*  2. Build readability's DOM tree.
	*  3. Grab the article content from the current dom tree.
	*  4. Replace the current DOM tree with the new one.
	*  5. Read peacefully.
	*
	* @return boolean true if we found content, false otherwise
	**/
	public function init()
	{
		if (!isset($this->dom->documentElement)) return false;
		$this->removeScripts($this->dom);
		//die($this->getInnerHTML($this->dom->documentElement));
		
		// Assume successful outcome
		$this->success = true;

		$bodyElems = $this->dom->getElementsByTagName('body');
		if ($bodyElems->length > 0) {
			if ($this->bodyCache == null) {
				$this->bodyCache = $bodyElems->item(0)->innerHTML;
			}
			if ($this->body == null) {
				$this->body = $bodyElems->item(0);
			}
		}

		$this->prepDocument();
		
		//die($this->dom->documentElement->parentNode->nodeType);
		//$this->setInnerHTML($this->dom->documentElement, $this->getInnerHTML($this->dom->documentElement));
		//die($this->getInnerHTML($this->dom->documentElement));

		/* Build readability's DOM tree */
		$overlay        = $this->dom->createElement('div');
		$innerDiv       = $this->dom->createElement('div');
		$articleTitle   = $this->getArticleTitle();
		$articleContent = $this->grabArticle();

		if (!$articleContent) {
			$this->success = false;
			$articleContent = $this->dom->createElement('div');
			$articleContent->setAttribute('id', 'readability-content');
			$articleContent->innerHTML = '<p>Sorry, Readability was unable to parse this page for content.</p>';		
		}
		
		$overlay->setAttribute('id', 'readOverlay');
		$innerDiv->setAttribute('id', 'readInner');

		/* Glue the structure of our document together. */
		$innerDiv->appendChild($articleTitle);
		$innerDiv->appendChild($articleContent);
		$overlay->appendChild($innerDiv);
		
		/* Clear the old HTML, insert the new content. */
		$this->body->innerHTML = '';
		$this->body->appendChild($overlay);
		//document.body.insertBefore(overlay, document.body.firstChild);
		$this->body->removeAttribute('style');

		$this->postProcessContent($articleContent);
		
		// Set title and content instance variables
		$this->articleTitle = $articleTitle;
		$this->articleContent = $articleContent;
		
		return $this->success;
	}
	
	/**
	* Debug
	*/
	protected function dbg($msg) {
		if ($this->debug) echo '* ',$msg, '<br />', "\n";
	}
	
	/**
	* Run any post-process modifications to article content as necessary.
	*
	* @param DOMElement
	* @return void
	*/
	public function postProcessContent($articleContent) {
		if ($this->convertLinksToFootnotes && !preg_match('/wikipedia\.org/', @$this->url)) { 
			$this->addFootnotes($articleContent);
		}
	}
	
	/**
	* Get the article title as an H1.
	*
	* @return DOMElement
	*/
	protected function getArticleTitle() {
		$curTitle = '';
		$origTitle = '';

		try {
			$curTitle = $origTitle = $this->getInnerText($this->dom->getElementsByTagName('title')->item(0));
		} catch(Exception $e) {}
		
		if (preg_match('/ [\|\-] /', $curTitle))
		{
			$curTitle = preg_replace('/(.*)[\|\-] .*/i', '$1', $origTitle);
			
			if (count(explode(' ', $curTitle)) < 3) {
				$curTitle = preg_replace('/[^\|\-]*[\|\-](.*)/i', '$1', $origTitle);
			}
		}
		else if (strpos($curTitle, ': ') !== false)
		{
			$curTitle = preg_replace('/.*:(.*)/i', '$1', $origTitle);

			if (count(explode(' ', $curTitle)) < 3) {
				$curTitle = preg_replace('/[^:]*[:](.*)/i','$1', $origTitle);
			}
		}
		else if(strlen($curTitle) > 150 || strlen($curTitle) < 15)
		{
			$hOnes = $this->dom->getElementsByTagName('h1');
			if($hOnes->length == 1)
			{
				$curTitle = $this->getInnerText($hOnes->item(0));
			}
		}

		$curTitle = trim($curTitle);

		if (count(explode(' ', $curTitle)) <= 4) {
			$curTitle = $origTitle;
		}
		
		$articleTitle = $this->dom->createElement('h1');
		$articleTitle->innerHTML = $curTitle;
		
		return $articleTitle;
	}
	
	/**
	* Prepare the HTML document for readability to scrape it.
	* This includes things like stripping javascript, CSS, and handling terrible markup.
	* 
	* @return void
	**/
	protected function prepDocument() {
		/**
		* In some cases a body element can't be found (if the HTML is totally hosed for example)
		* so we create a new body node and append it to the document.
		*/
		if ($this->body == null)
		{
			$this->body = $this->dom->createElement('body');
			$this->dom->documentElement->appendChild($this->body);
		}
		$this->body->setAttribute('id', 'readabilityBody');

		/* Remove all style tags in head */
		$styleTags = $this->dom->getElementsByTagName('style');
		for ($i = $styleTags->length-1; $i >= 0; $i--)
		{
			$styleTags->item($i)->parentNode->removeChild($styleTags->item($i));
		}

		/* Turn all double br's into p's */
		/* Note, this is pretty costly as far as processing goes. Maybe optimize later. */
		//document.body.innerHTML = document.body.innerHTML.replace(readability.regexps.replaceBrs, '</p><p>').replace(readability.regexps.replaceFonts, '<$1span>');
		// We do this in the constructor for PHP as that's when we have raw HTML - before parsing it into a DOM tree.
		// Manipulating innerHTML as it's done in JS is not possible in PHP.
	}

	/**
	* For easier reading, convert this document to have footnotes at the bottom rather than inline links.
	* @see http://www.roughtype.com/archives/2010/05/experiments_in.php
	*
	* @return void
	**/
	public function addFootnotes($articleContent) {
		$footnotesWrapper = $this->dom->createElement('div');
		$footnotesWrapper->setAttribute('id', 'readability-footnotes');
		$footnotesWrapper->innerHTML = '<h3>References</h3>';
		
		$articleFootnotes = $this->dom->createElement('ol');
		$articleFootnotes->setAttribute('id', 'readability-footnotes-list');
		$footnotesWrapper->appendChild($articleFootnotes);
		
		$articleLinks = $articleContent->getElementsByTagName('a');
		
		$linkCount = 0;
		for ($i = 0; $i < $articleLinks->length; $i++)
		{
			$articleLink  = $articleLinks->item($i);
			$footnoteLink = $articleLink->cloneNode(true);
			$refLink      = $this->dom->createElement('a');
			$footnote     = $this->dom->createElement('li');
			$linkDomain   = @parse_url($footnoteLink->getAttribute('href'), PHP_URL_HOST);
			if (!$linkDomain && isset($this->url)) $linkDomain = @parse_url($this->url, PHP_URL_HOST);
			//linkDomain   = footnoteLink.host ? footnoteLink.host : document.location.host,
			$linkText     = $this->getInnerText($articleLink);
			
			if ((strpos($articleLink->getAttribute('class'), 'readability-DoNotFootnote') !== false) || preg_match($this->regexps['skipFootnoteLink'], $linkText)) {
				continue;
			}
			
			$linkCount++;

			/** Add a superscript reference after the article link */
			$refLink->setAttribute('href', '#readabilityFootnoteLink-' . $linkCount);
			$refLink->innerHTML = '<small><sup>[' . $linkCount . ']</sup></small>';
			$refLink->setAttribute('class', 'readability-DoNotFootnote');
			$refLink->setAttribute('style', 'color: inherit;');
			
			//TODO: does this work or should we use DOMNode.isSameNode()?
			if ($articleLink->parentNode->lastChild == $articleLink) {
				$articleLink->parentNode->appendChild($refLink);
			} else {
				$articleLink->parentNode->insertBefore($refLink, $articleLink->nextSibling);
			}

			$articleLink->setAttribute('style', 'color: inherit; text-decoration: none;');
			$articleLink->setAttribute('name', 'readabilityLink-' . $linkCount);

			$footnote->innerHTML = '<small><sup><a href="#readabilityLink-' . $linkCount . '" title="Jump to Link in Article">^</a></sup></small> ';

			$footnoteLink->innerHTML = ($footnoteLink->getAttribute('title') != '' ? $footnoteLink->getAttribute('title') : $linkText);
			$footnoteLink->setAttribute('name', 'readabilityFootnoteLink-' . $linkCount);
			
			$footnote->appendChild($footnoteLink);
			if ($linkDomain) $footnote->innerHTML = $footnote->innerHTML . '<small> (' . $linkDomain . ')</small>';
			
			$articleFootnotes->appendChild($footnote);
		}

		if ($linkCount > 0) {
			$articleContent->appendChild($footnotesWrapper);           
		}
	}

	/**
	* Reverts P elements with class 'readability-styled'
	* to text nodes - which is what they were before.
	*
	* @param DOMElement
	* @return void
	*/
	function revertReadabilityStyledElements($articleContent) {
		$xpath = new DOMXPath($articleContent->ownerDocument);
		$elems = $xpath->query('.//p[@class="readability-styled"]', $articleContent);
		//$elems = $articleContent->getElementsByTagName('p');
		for ($i = $elems->length-1; $i >= 0; $i--) {
			$e = $elems->item($i);
			$e->parentNode->replaceChild($articleContent->ownerDocument->createTextNode($e->textContent), $e);
			//if ($e->hasAttribute('class') && $e->getAttribute('class') == 'readability-styled') {
			//	$e->parentNode->replaceChild($this->dom->createTextNode($e->textContent), $e);
			//}
		}
	}
	
	/**
	* Prepare the article node for display. Clean out any inline styles,
	* iframes, forms, strip extraneous <p> tags, etc.
	*
	* @param DOMElement
	* @return void
	*/
	function prepArticle($articleContent) {
		$this->cleanStyles($articleContent);
		$this->killBreaks($articleContent);
		if ($this->revertForcedParagraphElements) {
			$this->revertReadabilityStyledElements($articleContent);
		}

		/* Clean out junk from the article content */
		$this->cleanConditionally($articleContent, 'form');
		$this->clean($articleContent, 'object');
		$this->clean($articleContent, 'h1');

		/**
		* If there is only one h2, they are probably using it
		* as a header and not a subheader, so remove it since we already have a header.
		***/
		if ($articleContent->getElementsByTagName('h2')->length == 1) {
			$this->clean($articleContent, 'h2'); 
		}
		$this->clean($articleContent, 'iframe');

		$this->cleanHeaders($articleContent);

		/* Do these last as the previous stuff may have removed junk that will affect these */
		$this->cleanConditionally($articleContent, 'table');
		$this->cleanConditionally($articleContent, 'ul');
		$this->cleanConditionally($articleContent, 'div');

		/* Remove extra paragraphs */
		$articleParagraphs = $articleContent->getElementsByTagName('p');
		for ($i = $articleParagraphs->length-1; $i >= 0; $i--)
		{
			$imgCount    = $articleParagraphs->item($i)->getElementsByTagName('img')->length;
			$embedCount  = $articleParagraphs->item($i)->getElementsByTagName('embed')->length;
			$objectCount = $articleParagraphs->item($i)->getElementsByTagName('object')->length;
			
			if ($imgCount === 0 && $embedCount === 0 && $objectCount === 0 && $this->getInnerText($articleParagraphs->item($i), false) == '')
			{
				$articleParagraphs->item($i)->parentNode->removeChild($articleParagraphs->item($i));
			}
		}

		try {
			$articleContent->innerHTML = preg_replace('/<br[^>]*>\s*<p/i', '<p', $articleContent->innerHTML);
			//articleContent.innerHTML = articleContent.innerHTML.replace(/<br[^>]*>\s*<p/gi, '<p');      
		}
		catch (Exception $e) {
			$this->dbg("Cleaning innerHTML of breaks failed. This is an IE strict-block-elements bug. Ignoring.: " . $e);
		}
	}
	
	/**
	* Initialize a node with the readability object. Also checks the
	* className/id for special names to add to its score.
	*
	* @param Element
	* @return void
	**/
	protected function initializeNode($node) {
		$readability = $this->dom->createAttribute('readability');
		$readability->value = 0; // this is our contentScore
		$node->setAttributeNode($readability);		         

		switch (strtoupper($node->tagName)) { // unsure if strtoupper is needed, but using it just in case
			case 'DIV':
				$readability->value += 5;
				break;

			case 'PRE':
			case 'TD':
			case 'BLOCKQUOTE':
				$readability->value += 3;
				break;
				
			case 'ADDRESS':
			case 'OL':
			case 'UL':
			case 'DL':
			case 'DD':
			case 'DT':
			case 'LI':
			case 'FORM':
				$readability->value -= 3;
				break;

			case 'H1':
			case 'H2':
			case 'H3':
			case 'H4':
			case 'H5':
			case 'H6':
			case 'TH':
				$readability->value -= 5;
				break;
		}
		$readability->value += $this->getClassWeight($node);
	}
	
	/***
	* grabArticle - Using a variety of metrics (content score, classname, element types), find the content that is
	*               most likely to be the stuff a user wants to read. Then return it wrapped up in a div.
	*
	* @return DOMElement
	**/
	protected function grabArticle($page=null) {
		$stripUnlikelyCandidates = $this->flagIsActive(self::FLAG_STRIP_UNLIKELYS);
		if (!$page) $page = $this->dom;
		$allElements = $page->getElementsByTagName('*');
		/**
		* First, node prepping. Trash nodes that look cruddy (like ones with the class name "comment", etc), and turn divs
		* into P tags where they have been used inappropriately (as in, where they contain no other block level elements.)
		*
		* Note: Assignment from index for performance. See http://www.peachpit.com/articles/article.aspx?p=31567&seqNum=5
		* TODO: Shouldn't this be a reverse traversal?
		**/
		$node = null;
		$nodesToScore = array();
		for ($nodeIndex = 0; ($node = $allElements->item($nodeIndex)); $nodeIndex++) {
		//for ($nodeIndex=$targetList->length-1; $nodeIndex >= 0; $nodeIndex--) {
			//$node = $targetList->item($nodeIndex);
			$tagName = strtoupper($node->tagName);
			/* Remove unlikely candidates */
			if ($stripUnlikelyCandidates) {
				$unlikelyMatchString = $node->getAttribute('class') . $node->getAttribute('id');
				if (
					preg_match($this->regexps['unlikelyCandidates'], $unlikelyMatchString) &&
					!preg_match($this->regexps['okMaybeItsACandidate'], $unlikelyMatchString) &&
					$tagName != 'BODY'
				)
				{
					$this->dbg('Removing unlikely candidate - ' . $unlikelyMatchString);
					//$nodesToRemove[] = $node;
					$node->parentNode->removeChild($node);
					$nodeIndex--;
					continue;
				}               
			}

			if ($tagName == 'P' || $tagName == 'TD' || $tagName == 'PRE') {
				$nodesToScore[] = $node;
			}

			/* Turn all divs that don't have children block level elements into p's */
			if ($tagName == 'DIV') {
				if (!preg_match($this->regexps['divToPElements'], $node->innerHTML)) {
					//$this->dbg('Altering div to p');
					$newNode = $this->dom->createElement('p');
					try {
						$newNode->innerHTML = $node->innerHTML;
						//$nodesToReplace[] = array('new'=>$newNode, 'old'=>$node);
						$node->parentNode->replaceChild($newNode, $node);
						$nodeIndex--;
						$nodesToScore[] = $node; // or $newNode?
					}
					catch(Exception $e) {
						$this->dbg('Could not alter div to p, reverting back to div.: ' . $e);
					}
				}
				else
				{
					// EXPERIMENTAL
					// TODO: change these p elements back to text nodes after processing
					for ($i = 0, $il = $node->childNodes->length; $i < $il; $i++) {
						$childNode = $node->childNodes->item($i);
						if ($childNode->nodeType == 3) { // XML_TEXT_NODE
							//$this->dbg('replacing text node with a p tag with the same content.');
							$p = $this->dom->createElement('p');
							$p->innerHTML = $childNode->nodeValue;
							$p->setAttribute('style', 'display: inline;');
							$p->setAttribute('class', 'readability-styled');
							$childNode->parentNode->replaceChild($p, $childNode);
						}
					}
				}
			}
		}
		
		/**
		* Loop through all paragraphs, and assign a score to them based on how content-y they look.
		* Then add their score to their parent node.
		*
		* A score is determined by things like number of commas, class names, etc. Maybe eventually link density.
		**/
		$candidates = array();
		for ($pt=0; $pt < count($nodesToScore); $pt++) {
			$parentNode      = $nodesToScore[$pt]->parentNode;
			// $grandParentNode = $parentNode ? $parentNode->parentNode : null;
			$grandParentNode = !$parentNode ? null : (($parentNode->parentNode instanceof DOMElement) ? $parentNode->parentNode : null);
			$innerText       = $this->getInnerText($nodesToScore[$pt]);

			if (!$parentNode || !isset($parentNode->tagName)) {
				continue;
			}

			/* If this paragraph is less than 25 characters, don't even count it. */
			if(strlen($innerText) < 25) {
				continue;
			}

			/* Initialize readability data for the parent. */
			if (!$parentNode->hasAttribute('readability')) 
			{
				$this->initializeNode($parentNode);
				$candidates[] = $parentNode;
			}

			/* Initialize readability data for the grandparent. */
			if ($grandParentNode && !$grandParentNode->hasAttribute('readability') && isset($grandParentNode->tagName))
			{
				$this->initializeNode($grandParentNode);
				$candidates[] = $grandParentNode;
			}

			$contentScore = 0;

			/* Add a point for the paragraph itself as a base. */
			$contentScore++;

			/* Add points for any commas within this paragraph */
			$contentScore += count(explode(',', $innerText));
			
			/* For every 100 characters in this paragraph, add another point. Up to 3 points. */
			$contentScore += min(floor(strlen($innerText) / 100), 3);
			
			/* Add the score to the parent. The grandparent gets half. */
			$parentNode->getAttributeNode('readability')->value += $contentScore;

			if ($grandParentNode) {
				$grandParentNode->getAttributeNode('readability')->value += $contentScore/2;             
			}
		}

		/**
		* After we've calculated scores, loop through all of the possible candidate nodes we found
		* and find the one with the highest score.
		**/
		$topCandidate = null;
		for ($c=0, $cl=count($candidates); $c < $cl; $c++)
		{
			/**
			* Scale the final candidates score based on link density. Good content should have a
			* relatively small link density (5% or less) and be mostly unaffected by this operation.
			**/
			$readability = $candidates[$c]->getAttributeNode('readability');
			$readability->value = $readability->value * (1-$this->getLinkDensity($candidates[$c]));

			$this->dbg('Candidate: ' . $candidates[$c]->tagName . ' (' . $candidates[$c]->getAttribute('class') . ':' . $candidates[$c]->getAttribute('id') . ') with score ' . $readability->value);

			if (!$topCandidate || $readability->value > (int)$topCandidate->getAttribute('readability')) {
				$topCandidate = $candidates[$c];
			}
		}

		/**
		* If we still have no top candidate, just use the body as a last resort.
		* We also have to copy the body node so it is something we can modify.
		**/
		if ($topCandidate === null || strtoupper($topCandidate->tagName) == 'BODY')
		{
			$topCandidate = $this->dom->createElement('div');
			if ($page instanceof DOMDocument) {
				if (!isset($page->documentElement)) {
					// we don't have a body either? what a mess! :)
				} else {
					$topCandidate->innerHTML = $page->documentElement->innerHTML;
					$page->documentElement->innerHTML = '';
					$page->documentElement->appendChild($topCandidate);
				}
			} else {
				$topCandidate->innerHTML = $page->innerHTML;
				$page->innerHTML = '';
				$page->appendChild($topCandidate);
			}
			$this->initializeNode($topCandidate);
		}

		/**
		* Now that we have the top candidate, look through its siblings for content that might also be related.
		* Things like preambles, content split by ads that we removed, etc.
		**/
		$articleContent        = $this->dom->createElement('div');
		$articleContent->setAttribute('id', 'readability-content');
		$siblingScoreThreshold = max(10, ((int)$topCandidate->getAttribute('readability')) * 0.2);
		$siblingNodes          = $topCandidate->parentNode->childNodes;
		if (!isset($siblingNodes)) {
			$siblingNodes = new stdClass;
			$siblingNodes->length = 0;
		}

		for ($s=0, $sl=$siblingNodes->length; $s < $sl; $s++)
		{
			$siblingNode = $siblingNodes->item($s);
			$append      = false;

			$this->dbg('Looking at sibling node: ' . $siblingNode->nodeName . (($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability')) ? (' with score ' . $siblingNode->getAttribute('readability')) : ''));

			//dbg('Sibling has score ' . ($siblingNode->readability ? siblingNode.readability.contentScore : 'Unknown'));

			if ($siblingNode === $topCandidate)
			// or if ($siblingNode->isSameNode($topCandidate))
			{
				$append = true;
			}

			$contentBonus = 0;
			/* Give a bonus if sibling nodes and top candidates have the example same classname */
			if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->getAttribute('class') == $topCandidate->getAttribute('class') && $topCandidate->getAttribute('class') != '') {
				$contentBonus += ((int)$topCandidate->getAttribute('readability')) * 0.2;
			}

			if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability') && (((int)$siblingNode->getAttribute('readability')) + $contentBonus) >= $siblingScoreThreshold)
			{
				$append = true;
			}
			
			if (strtoupper($siblingNode->nodeName) == 'P') {
				$linkDensity = $this->getLinkDensity($siblingNode);
				$nodeContent = $this->getInnerText($siblingNode);
				$nodeLength  = strlen($nodeContent);
				
				if ($nodeLength > 80 && $linkDensity < 0.25)
				{
					$append = true;
				}
				else if ($nodeLength < 80 && $linkDensity === 0 && preg_match('/\.( |$)/', $nodeContent))
				{
					$append = true;
				}
			}

			/* Look for a special classname */
			if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('class') && $siblingNode->getAttribute('class') != '')
			{
				if (preg_match($this->regexps['okMaybeItsACandidate'], $siblingNode->getAttribute('class'))) {
					$append = true;
				}
			}

			/* Look for a special classname */
			if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('id') && $siblingNode->getAttribute('id') != '')
			{
				if (preg_match($this->regexps['okMaybeItsACandidate'], $siblingNode->getAttribute('id'))) {
					$append = true;
				}
			}


			if ($append)
			{
				$this->dbg('Appending node: ' . $siblingNode->nodeName);

				$nodeToAppend = null;
				$sibNodeName = strtoupper($siblingNode->nodeName);
				if ($sibNodeName != 'DIV' && $sibNodeName != 'P') {
					/* We have a node that isn't a common block level element, like a form or td tag. Turn it into a div so it doesn't get filtered out later by accident. */
					
					$this->dbg('Altering siblingNode of ' . $sibNodeName . ' to div.');
					$nodeToAppend = $this->dom->createElement('div');
					try {
						$nodeToAppend->setAttribute('id', $siblingNode->getAttribute('id'));
						$nodeToAppend->innerHTML = $siblingNode->innerHTML;
					}
					catch(Exception $e)
					{
						$this->dbg('Could not alter siblingNode to div, reverting back to original.');
						$nodeToAppend = $siblingNode;
						$s--;
						$sl--;
					}
				} else {
					$nodeToAppend = $siblingNode;
					$s--;
					$sl--;
				}
				
				/* To ensure a node does not interfere with readability styles, remove its classnames */
				$nodeToAppend->removeAttribute('class');

				/* Append sibling and subtract from our list because it removes the node when you append to another node */
				$articleContent->appendChild($nodeToAppend);
			}
		}

		/**
		* So we have all of the content that we need. Now we clean it up for presentation.
		**/
		$this->prepArticle($articleContent);

		/**
		* Now that we've gone through the full algorithm, check to see if we got any meaningful content.
		* If we didn't, we may need to re-run grabArticle with different flags set. This gives us a higher
		* likelihood of finding the content, and the sieve approach gives us a higher likelihood of
		* finding the -right- content.
		**/
		if (strlen($this->getInnerText($articleContent, false)) < 250)
		{
			// TODO: find out why element disappears sometimes, e.g. for this URL http://www.businessinsider.com/6-hedge-fund-etfs-for-average-investors-2011-7
			// in the meantime, we check and create an empty element if it's not there.
			if (!isset($this->body->childNodes)) $this->body = $this->dom->createElement('body');
			$this->body->innerHTML = $this->bodyCache;
			
			if ($this->flagIsActive(self::FLAG_STRIP_UNLIKELYS)) {
				$this->removeFlag(self::FLAG_STRIP_UNLIKELYS);
				return $this->grabArticle($this->body);
			}
			else if ($this->flagIsActive(self::FLAG_WEIGHT_CLASSES)) {
				$this->removeFlag(self::FLAG_WEIGHT_CLASSES);
				return $this->grabArticle($this->body);              
			}
			else if ($this->flagIsActive(self::FLAG_CLEAN_CONDITIONALLY)) {
				$this->removeFlag(self::FLAG_CLEAN_CONDITIONALLY);
				return $this->grabArticle($this->body);
			}
			else {
				return false;
			}
		}
		return $articleContent;
	}
	
	/**
	* Remove script tags from document
	*
	* @param DOMElement
	* @return void
	*/
	public function removeScripts($doc) {
		$scripts = $doc->getElementsByTagName('script');
		for($i = $scripts->length-1; $i >= 0; $i--)
		{
			$scripts->item($i)->parentNode->removeChild($scripts->item($i));
		}
	}
	
	/**
	* Get the inner text of a node.
	* This also strips out any excess whitespace to be found.
	*
	* @param DOMElement $
	* @param boolean $normalizeSpaces (default: true)
	* @return string
	**/
	public function getInnerText($e, $normalizeSpaces=true) {
		$textContent = '';

		if (!isset($e->textContent) || $e->textContent == '') {
			return '';
		}

		$textContent = trim($e->textContent);

		if ($normalizeSpaces) {
			return preg_replace($this->regexps['normalize'], ' ', $textContent);
		} else {
			return $textContent;
		}
	}

	/**
	* Get the number of times a string $s appears in the node $e.
	*
	* @param DOMElement $e
	* @param string - what to count. Default is ","
	* @return number (integer)
	**/
	public function getCharCount($e, $s=',') {
		return substr_count($this->getInnerText($e), $s);
	}

	/**
	* Remove the style attribute on every $e and under.
	*
	* @param DOMElement $e
	* @return void
	*/
	public function cleanStyles($e) {
		if (!is_object($e)) return;
		$elems = $e->getElementsByTagName('*');
		foreach ($elems as $elem) {
			$elem->removeAttribute('style');
		}
	}
	
	/**
	* Get the density of links as a percentage of the content
	* This is the amount of text that is inside a link divided by the total text in the node.
	* 
	* @param DOMElement $e
	* @return number (float)
	*/
	public function getLinkDensity($e) {
		$links      = $e->getElementsByTagName('a');
		$textLength = strlen($this->getInnerText($e));
		$linkLength = 0;
		for ($i=0, $il=$links->length; $i < $il; $i++)
		{
			$linkLength += strlen($this->getInnerText($links->item($i)));
		}
		if ($textLength > 0) {
			return $linkLength / $textLength;
		} else {
			return 0;
		}
	}
	
	/**
	* Get an elements class/id weight. Uses regular expressions to tell if this 
	* element looks good or bad.
	*
	* @param DOMElement $e
	* @return number (Integer)
	*/
	public function getClassWeight($e) {
		if(!$this->flagIsActive(self::FLAG_WEIGHT_CLASSES)) {
			return 0;
		}

		$weight = 0;

		/* Look for a special classname */
		if ($e->hasAttribute('class') && $e->getAttribute('class') != '')
		{
			if (preg_match($this->regexps['negative'], $e->getAttribute('class'))) {
				$weight -= 25;
			}
			if (preg_match($this->regexps['positive'], $e->getAttribute('class'))) {
				$weight += 25;
			}
		}

		/* Look for a special ID */
		if ($e->hasAttribute('id') && $e->getAttribute('id') != '')
		{
			if (preg_match($this->regexps['negative'], $e->getAttribute('id'))) {
				$weight -= 25;
			}
			if (preg_match($this->regexps['positive'], $e->getAttribute('id'))) {
				$weight += 25;
			}
		}
		return $weight;
	}

	/**
	* Remove extraneous break tags from a node.
	*
	* @param DOMElement $node
	* @return void
	*/
	public function killBreaks($node) {
		$html = $node->innerHTML;
		$html = preg_replace($this->regexps['killBreaks'], '<br />', $html);
		$node->innerHTML = $html;
	}

	/**
	* Clean a node of all elements of type "tag".
	* (Unless it's a youtube/vimeo video. People love movies.)
	*
	* @param DOMElement $e
	* @param string $tag
	* @return void
	*/
	public function clean($e, $tag) {
		$targetList = $e->getElementsByTagName($tag);
		$isEmbed = ($tag == 'object' || $tag == 'embed');
		
		for ($y=$targetList->length-1; $y >= 0; $y--) {
			/* Allow youtube and vimeo videos through as people usually want to see those. */
			if ($isEmbed) {
				$attributeValues = '';
				for ($i=0, $il=$targetList->item($y)->attributes->length; $i < $il; $i++) {
					$attributeValues .= $targetList->item($y)->attributes->item($i)->value . '|'; // DOMAttr? (TODO: test)
				}
				
				/* First, check the elements attributes to see if any of them contain youtube or vimeo */
				if (preg_match($this->regexps['video'], $attributeValues)) {
					continue;
				}

				/* Then check the elements inside this element for the same. */
				if (preg_match($this->regexps['video'], $targetList->item($y)->innerHTML)) {
					continue;
				}
			}
			$targetList->item($y)->parentNode->removeChild($targetList->item($y));
		}
	}
	
	/**
	* Clean an element of all tags of type "tag" if they look fishy.
	* "Fishy" is an algorithm based on content length, classnames, 
	* link density, number of images & embeds, etc.
	*
	* @param DOMElement $e
	* @param string $tag
	* @return void
	*/
	public function cleanConditionally($e, $tag) {
		if (!$this->flagIsActive(self::FLAG_CLEAN_CONDITIONALLY)) {
			return;
		}

		$tagsList = $e->getElementsByTagName($tag);
		$curTagsLength = $tagsList->length;

		/**
		* Gather counts for other typical elements embedded within.
		* Traverse backwards so we can remove nodes at the same time without effecting the traversal.
		*
		* TODO: Consider taking into account original contentScore here.
		*/
		for ($i=$curTagsLength-1; $i >= 0; $i--) {
			$weight = $this->getClassWeight($tagsList->item($i));
			$contentScore = ($tagsList->item($i)->hasAttribute('readability')) ? (int)$tagsList->item($i)->getAttribute('readability') : 0;
			
			$this->dbg('Cleaning Conditionally ' . $tagsList->item($i)->tagName . ' (' . $tagsList->item($i)->getAttribute('class') . ':' . $tagsList->item($i)->getAttribute('id') . ')' . (($tagsList->item($i)->hasAttribute('readability')) ? (' with score ' . $tagsList->item($i)->getAttribute('readability')) : ''));

			if ($weight + $contentScore < 0) {
				$tagsList->item($i)->parentNode->removeChild($tagsList->item($i));
			}
			else if ( $this->getCharCount($tagsList->item($i), ',') < 10) {
				/**
				* If there are not very many commas, and the number of
				* non-paragraph elements is more than paragraphs or other ominous signs, remove the element.
				**/
				$p      = $tagsList->item($i)->getElementsByTagName('p')->length;
				$img    = $tagsList->item($i)->getElementsByTagName('img')->length;
				$li     = $tagsList->item($i)->getElementsByTagName('li')->length-100;
				$input  = $tagsList->item($i)->getElementsByTagName('input')->length;

				$embedCount = 0;
				$embeds = $tagsList->item($i)->getElementsByTagName('embed');
				for ($ei=0, $il=$embeds->length; $ei < $il; $ei++) {
					if (preg_match($this->regexps['video'], $embeds->item($ei)->getAttribute('src'))) {
					$embedCount++; 
					}
				}

				$linkDensity   = $this->getLinkDensity($tagsList->item($i));
				$contentLength = strlen($this->getInnerText($tagsList->item($i)));
				$toRemove      = false;

				if ( $img > $p ) {
					$toRemove = true;
				} else if ($li > $p && $tag != 'ul' && $tag != 'ol') {
					$toRemove = true;
				} else if ( $input > floor($p/3) ) {
					$toRemove = true; 
				} else if ($contentLength < 25 && ($img === 0 || $img > 2) ) {
					$toRemove = true;
				} else if($weight < 25 && $linkDensity > 0.2) {
					$toRemove = true;
				} else if($weight >= 25 && $linkDensity > 0.5) {
					$toRemove = true;
				} else if(($embedCount == 1 && $contentLength < 75) || $embedCount > 1) {
					$toRemove = true;
				}

				if ($toRemove) {
					$tagsList->item($i)->parentNode->removeChild($tagsList->item($i));
				}
			}
		}
	}

	/**
	* Clean out spurious headers from an Element. Checks things like classnames and link density.
	*
	* @param DOMElement $e
	* @return void
	*/
	public function cleanHeaders($e) {
		for ($headerIndex = 1; $headerIndex < 3; $headerIndex++) {
			$headers = $e->getElementsByTagName('h' . $headerIndex);
			for ($i=$headers->length-1; $i >=0; $i--) {
				if ($this->getClassWeight($headers->item($i)) < 0 || $this->getLinkDensity($headers->item($i)) > 0.33) {
					$headers->item($i)->parentNode->removeChild($headers->item($i));
				}
			}
		}
	}

	public function flagIsActive($flag) {
		return ($this->flags & $flag) > 0;
	}
	
	public function addFlag($flag) {
		$this->flags = $this->flags | $flag;
	}
	
	public function removeFlag($flag) {
		$this->flags = $this->flags & ~$flag;
	}
}
?>