Mercurial > hg > ywww

comparison goodreads/simple_html_dom.php @ 6:077b0a0a3e6d

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
remaining originals according to dependency walk
author Robert Boland <robert@markup.co.uk>
date Thu, 16 Feb 2017 22:29:02 +0000
parents
children
comparison
equal deleted inserted replaced
5:55445b456ad0 6:077b0a0a3e6d
1 <?php
2 /**
3 * Website: http://sourceforge.net/projects/simplehtmldom/
4 * Acknowledge: Jose Solorzano (https://sourceforge.net/projects/php-html/)
5 * Contributions by:
6 * Yousuke Kumakura (Attribute filters)
7 * Vadim Voituk (Negative indexes supports of "find" method)
8 * Antcs (Constructor with automatically load contents either text or file/url)
9 *
10 * all affected sections have comments starting with "PaperG"
11 *
12 * Paperg - Added case insensitive testing of the value of the selector.
13 * Paperg - Added tag_start for the starting index of tags - NOTE: This works but not accurately.
14 * This tag_start gets counted AFTER \r\n have been crushed out, and after the remove_noice calls so it will not reflect the REAL position of the tag in the source,
15 * it will almost always be smaller by some amount.
16 * We use this to determine how far into the file the tag in question is. This "percentage will never be accurate as the $dom->size is the "real" number of bytes the dom was created from.
17 * but for most purposes, it's a really good estimation.
18 * Paperg - Added the forceTagsClosed to the dom constructor. Forcing tags closed is great for malformed html, but it CAN lead to parsing errors.
19 * Allow the user to tell us how much they trust the html.
20 * Paperg add the text and plaintext to the selectors for the find syntax. plaintext implies text in the innertext of a node. text implies that the tag is a text node.
21 * This allows for us to find tags based on the text they contain.
22 * Create find_ancestor_tag to see if a tag is - at any level - inside of another specific tag.
23 * Paperg: added parse_charset so that we know about the character set of the source document.
24 * NOTE: If the user's system has a routine called get_last_retrieve_url_contents_content_type availalbe, we will assume it's returning the content-type header from the
25 * last transfer or curl_exec, and we will parse that and use it in preference to any other method of charset detection.
26 *
27 * Licensed under The MIT License
28 * Redistributions of files must retain the above copyright notice.
29 *
30 * @author S.C. Chen <me578022@gmail.com>
31 * @author John Schlick
32 * @author Rus Carroll
33 * @version 1.11 ($Rev: 184 $)
34 * @package PlaceLocalInclude
35 * @subpackage simple_html_dom
36 */
37
38 /**
39 * All of the Defines for the classes below.
40 * @author S.C. Chen <me578022@gmail.com>
41 */
42 define('HDOM_TYPE_ELEMENT', 1);
43 define('HDOM_TYPE_COMMENT', 2);
44 define('HDOM_TYPE_TEXT', 3);
45 define('HDOM_TYPE_ENDTAG', 4);
46 define('HDOM_TYPE_ROOT', 5);
47 define('HDOM_TYPE_UNKNOWN', 6);
48 define('HDOM_QUOTE_DOUBLE', 0);
49 define('HDOM_QUOTE_SINGLE', 1);
50 define('HDOM_QUOTE_NO', 3);
51 define('HDOM_INFO_BEGIN', 0);
52 define('HDOM_INFO_END', 1);
53 define('HDOM_INFO_QUOTE', 2);
54 define('HDOM_INFO_SPACE', 3);
55 define('HDOM_INFO_TEXT', 4);
56 define('HDOM_INFO_INNER', 5);
57 define('HDOM_INFO_OUTER', 6);
58 define('HDOM_INFO_ENDSPACE',7);
59 define('DEFAULT_TARGET_CHARSET', 'UTF-8');
60 define('DEFAULT_BR_TEXT', "\r\n");
61 // helper functions
62 // -----------------------------------------------------------------------------
63 // get html dom from file
64 // $maxlen is defined in the code as PHP_STREAM_COPY_ALL which is defined as -1.
65 function file_get_html($url, $use_include_path = false, $context=null, $offset = -1, $maxLen=-1, $lowercase = true, $forceTagsClosed=true, $target_charset = DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT)
66 {
67 // We DO force the tags to be terminated.
68 $dom = new simple_html_dom(null, $lowercase, $forceTagsClosed, $target_charset, $defaultBRText);
69 // For sourceforge users: uncomment the next line and comment the retreive_url_contents line 2 lines down if it is not already done.
70 $contents = file_get_contents($url, $use_include_path, $context, $offset);
71 // Paperg - use our own mechanism for getting the contents as we want to control the timeout.
72 // $contents = retrieve_url_contents($url);
73 if (empty($contents))
74 {
75 return false;
76 }
77 // The second parameter can force the selectors to all be lowercase.
78 $dom->load($contents, $lowercase, $stripRN);
79 return $dom;
80 }
81
82 // get html dom from string
83 function str_get_html($str, $lowercase=true, $forceTagsClosed=true, $target_charset = DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT)
84 {
85 $dom = new simple_html_dom(null, $lowercase, $forceTagsClosed, $target_charset, $defaultBRText);
86 if (empty($str))
87 {
88 $dom->clear();
89 return false;
90 }
91 $dom->load($str, $lowercase, $stripRN);
92 return $dom;
93 }
94
95 // dump html dom tree
96 function dump_html_tree($node, $show_attr=true, $deep=0)
97 {
98 $node->dump($node);
99 }
100
101 /**
102 * simple html dom node
103 * PaperG - added ability for "find" routine to lowercase the value of the selector.
104 * PaperG - added $tag_start to track the start position of the tag in the total byte index
105 *
106 * @package PlaceLocalInclude
107 */
108 class simple_html_dom_node {
109 public $nodetype = HDOM_TYPE_TEXT;
110 public $tag = 'text';
111 public $attr = array();
112 public $children = array();
113 public $nodes = array();
114 public $parent = null;
115 public $_ = array();
116 public $tag_start = 0;
117 private $dom = null;
118
119 function __construct($dom)
120 {
121 $this->dom = $dom;
122 $dom->nodes[] = $this;
123 }
124
125 function __destruct()
126 {
127 $this->clear();
128 }
129
130 function __toString()
131 {
132 return $this->outertext();
133 }
134
135 // clean up memory due to php5 circular references memory leak...
136 function clear()
137 {
138 $this->dom = null;
139 $this->nodes = null;
140 $this->parent = null;
141 $this->children = null;
142 }
143
144 // dump node's tree
145 function dump($show_attr=true, $deep=0)
146 {
147 $lead = str_repeat(' ', $deep);
148
149 echo $lead.$this->tag;
150 if ($show_attr && count($this->attr)>0)
151 {
152 echo '(';
153 foreach ($this->attr as $k=>$v)
154 echo "[$k]=>\"".$this->$k.'", ';
155 echo ')';
156 }
157 echo "\n";
158
159 foreach ($this->nodes as $c)
160 $c->dump($show_attr, $deep+1);
161 }
162
163
164 // Debugging function to dump a single dom node with a bunch of information about it.
165 function dump_node()
166 {
167 echo $this->tag;
168 if (count($this->attr)>0)
169 {
170 echo '(';
171 foreach ($this->attr as $k=>$v)
172 {
173 echo "[$k]=>\"".$this->$k.'", ';
174 }
175 echo ')';
176 }
177 if (count($this->attr)>0)
178 {
179 echo ' $_ (';
180 foreach ($this->_ as $k=>$v)
181 {
182 if (is_array($v))
183 {
184 echo "[$k]=>(";
185 foreach ($v as $k2=>$v2)
186 {
187 echo "[$k2]=>\"".$v2.'", ';
188 }
189 echo ")";
190 } else {
191 echo "[$k]=>\"".$v.'", ';
192 }
193 }
194 echo ")";
195 }
196
197 if (isset($this->text))
198 {
199 echo " text: (" . $this->text . ")";
200 }
201
202 echo " children: " . count($this->children);
203 echo " nodes: " . count($this->nodes);
204 echo " tag_start: " . $this->tag_start;
205 echo "\n";
206
207 }
208
209 // returns the parent of node
210 function parent()
211 {
212 return $this->parent;
213 }
214
215 // returns children of node
216 function children($idx=-1)
217 {
218 if ($idx===-1) return $this->children;
219 if (isset($this->children[$idx])) return $this->children[$idx];
220 return null;
221 }
222
223 // returns the first child of node
224 function first_child()
225 {
226 if (count($this->children)>0) return $this->children[0];
227 return null;
228 }
229
230 // returns the last child of node
231 function last_child()
232 {
233 if (($count=count($this->children))>0) return $this->children[$count-1];
234 return null;
235 }
236
237 // returns the next sibling of node
238 function next_sibling()
239 {
240 if ($this->parent===null) return null;
241 $idx = 0;
242 $count = count($this->parent->children);
243 while ($idx<$count && $this!==$this->parent->children[$idx])
244 ++$idx;
245 if (++$idx>=$count) return null;
246 return $this->parent->children[$idx];
247 }
248
249 // returns the previous sibling of node
250 function prev_sibling()
251 {
252 if ($this->parent===null) return null;
253 $idx = 0;
254 $count = count($this->parent->children);
255 while ($idx<$count && $this!==$this->parent->children[$idx])
256 ++$idx;
257 if (--$idx<0) return null;
258 return $this->parent->children[$idx];
259 }
260
261 // function to locate a specific ancestor tag in the path to the root.
262 function find_ancestor_tag($tag)
263 {
264 global $debugObject;
265 if (is_object($debugObject))
266 {
267 $debugObject->debugLogEntry(1);
268 }
269
270 // Start by including ourselves in the comparison.
271 $returnDom = $this;
272
273 while (!is_null($returnDom))
274 {
275 if (is_object($debugObject))
276 {
277 $debugObject->debugLog(2, "Current tag is: " . $returnDom->tag);
278 }
279
280 if ($returnDom->tag == $tag)
281 {
282 break;
283 }
284 $returnDom = $returnDom->parent;
285 }
286 return $returnDom;
287 }
288
289 // get dom node's inner html
290 function innertext()
291 {
292 if (isset($this->_[HDOM_INFO_INNER])) return $this->_[HDOM_INFO_INNER];
293 if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
294
295 $ret = '';
296 foreach ($this->nodes as $n)
297 $ret .= $n->outertext();
298 return $ret;
299 }
300
301 // get dom node's outer text (with tag)
302 function outertext()
303 {
304 global $debugObject;
305 if (is_object($debugObject))
306 {
307 $text = '';
308 if ($this->tag == 'text')
309 {
310 if (!empty($this->text))
311 {
312 $text = " with text: " . $this->text;
313 }
314 }
315 $debugObject->debugLog(1, 'Innertext of tag: ' . $this->tag . $text);
316 }
317
318 if ($this->tag==='root') return $this->innertext();
319
320 // trigger callback
321 if ($this->dom && $this->dom->callback!==null)
322 {
323 call_user_func_array($this->dom->callback, array($this));
324 }
325
326 if (isset($this->_[HDOM_INFO_OUTER])) return $this->_[HDOM_INFO_OUTER];
327 if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
328
329 // render begin tag
330 if ($this->dom && $this->dom->nodes[$this->_[HDOM_INFO_BEGIN]])
331 {
332 $ret = $this->dom->nodes[$this->_[HDOM_INFO_BEGIN]]->makeup();
333 } else {
334 $ret = "";
335 }
336
337 // render inner text
338 if (isset($this->_[HDOM_INFO_INNER]))
339 {
340 // If it's a br tag... don't return the HDOM_INNER_INFO that we may or may not have added.
341 if ($this->tag != "br")
342 {
343 $ret .= $this->_[HDOM_INFO_INNER];
344 }
345 } else {
346 if ($this->nodes)
347 {
348 foreach ($this->nodes as $n)
349 {
350 $ret .= $this->convert_text($n->outertext());
351 }
352 }
353 }
354
355 // render end tag
356 if (isset($this->_[HDOM_INFO_END]) && $this->_[HDOM_INFO_END]!=0)
357 $ret .= '</'.$this->tag.'>';
358 return $ret;
359 }
360
361 // get dom node's plain text
362 function text()
363 {
364 if (isset($this->_[HDOM_INFO_INNER])) return $this->_[HDOM_INFO_INNER];
365 switch ($this->nodetype)
366 {
367 case HDOM_TYPE_TEXT: return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
368 case HDOM_TYPE_COMMENT: return '';
369 case HDOM_TYPE_UNKNOWN: return '';
370 }
371 if (strcasecmp($this->tag, 'script')===0) return '';
372 if (strcasecmp($this->tag, 'style')===0) return '';
373
374 $ret = '';
375 // In rare cases, (always node type 1 or HDOM_TYPE_ELEMENT - observed for some span tags, and some p tags) $this->nodes is set to NULL.
376 // NOTE: This indicates that there is a problem where it's set to NULL without a clear happening.
377 // WHY is this happening?
378 if (!is_null($this->nodes))
379 {
380 foreach ($this->nodes as $n)
381 {
382 $ret .= $this->convert_text($n->text());
383 }
384 }
385 return $ret;
386 }
387
388 function xmltext()
389 {
390 $ret = $this->innertext();
391 $ret = str_ireplace('<![CDATA[', '', $ret);
392 $ret = str_replace(']]>', '', $ret);
393 return $ret;
394 }
395
396 // build node's text with tag
397 function makeup()
398 {
399 // text, comment, unknown
400 if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
401
402 $ret = '<'.$this->tag;
403 $i = -1;
404
405 foreach ($this->attr as $key=>$val)
406 {
407 ++$i;
408
409 // skip removed attribute
410 if ($val===null || $val===false)
411 continue;
412
413 $ret .= $this->_[HDOM_INFO_SPACE][$i][0];
414 //no value attr: nowrap, checked selected...
415 if ($val===true)
416 $ret .= $key;
417 else {
418 switch ($this->_[HDOM_INFO_QUOTE][$i])
419 {
420 case HDOM_QUOTE_DOUBLE: $quote = '"'; break;
421 case HDOM_QUOTE_SINGLE: $quote = '\''; break;
422 default: $quote = '';
423 }
424 $ret .= $key.$this->_[HDOM_INFO_SPACE][$i][1].'='.$this->_[HDOM_INFO_SPACE][$i][2].$quote.$val.$quote;
425 }
426 }
427 $ret = $this->dom->restore_noise($ret);
428 return $ret . $this->_[HDOM_INFO_ENDSPACE] . '>';
429 }
430
431 // find elements by css selector
432 //PaperG - added ability for find to lowercase the value of the selector.
433 function find($selector, $idx=null, $lowercase=false)
434 {
435 $selectors = $this->parse_selector($selector);
436 if (($count=count($selectors))===0) return array();
437 $found_keys = array();
438
439 // find each selector
440 for ($c=0; $c<$count; ++$c)
441 {
442 // The change on the below line was documented on the sourceforge code tracker id 2788009
443 // used to be: if (($levle=count($selectors[0]))===0) return array();
444 if (($levle=count($selectors[$c]))===0) return array();
445 if (!isset($this->_[HDOM_INFO_BEGIN])) return array();
446
447 $head = array($this->_[HDOM_INFO_BEGIN]=>1);
448
449 // handle descendant selectors, no recursive!
450 for ($l=0; $l<$levle; ++$l)
451 {
452 $ret = array();
453 foreach ($head as $k=>$v)
454 {
455 $n = ($k===-1) ? $this->dom->root : $this->dom->nodes[$k];
456 //PaperG - Pass this optional parameter on to the seek function.
457 $n->seek($selectors[$c][$l], $ret, $lowercase);
458 }
459 $head = $ret;
460 }
461
462 foreach ($head as $k=>$v)
463 {
464 if (!isset($found_keys[$k]))
465 $found_keys[$k] = 1;
466 }
467 }
468
469 // sort keys
470 ksort($found_keys);
471
472 $found = array();
473 foreach ($found_keys as $k=>$v)
474 $found[] = $this->dom->nodes[$k];
475
476 // return nth-element or array
477 if (is_null($idx)) return $found;
478 else if ($idx<0) $idx = count($found) + $idx;
479 return (isset($found[$idx])) ? $found[$idx] : null;
480 }
481
482 // seek for given conditions
483 // PaperG - added parameter to allow for case insensitive testing of the value of a selector.
484 protected function seek($selector, &$ret, $lowercase=false)
485 {
486 global $debugObject;
487 if (is_object($debugObject))
488 {
489 $debugObject->debugLogEntry(1);
490 }
491
492 list($tag, $key, $val, $exp, $no_key) = $selector;
493
494 // xpath index
495 if ($tag && $key && is_numeric($key))
496 {
497 $count = 0;
498 foreach ($this->children as $c)
499 {
500 if ($tag==='*' || $tag===$c->tag) {
501 if (++$count==$key) {
502 $ret[$c->_[HDOM_INFO_BEGIN]] = 1;
503 return;
504 }
505 }
506 }
507 return;
508 }
509
510 $end = (!empty($this->_[HDOM_INFO_END])) ? $this->_[HDOM_INFO_END] : 0;
511 if ($end==0) {
512 $parent = $this->parent;
513 while (!isset($parent->_[HDOM_INFO_END]) && $parent!==null) {
514 $end -= 1;
515 $parent = $parent->parent;
516 }
517 $end += $parent->_[HDOM_INFO_END];
518 }
519
520 for ($i=$this->_[HDOM_INFO_BEGIN]+1; $i<$end; ++$i) {
521 $node = $this->dom->nodes[$i];
522
523 $pass = true;
524
525 if ($tag==='*' && !$key) {
526 if (in_array($node, $this->children, true))
527 $ret[$i] = 1;
528 continue;
529 }
530
531 // compare tag
532 if ($tag && $tag!=$node->tag && $tag!=='*') {$pass=false;}
533 // compare key
534 if ($pass && $key) {
535 if ($no_key) {
536 if (isset($node->attr[$key])) $pass=false;
537 } else {
538 if (($key != "plaintext") && !isset($node->attr[$key])) $pass=false;
539 }
540 }
541 // compare value
542 if ($pass && $key && $val && $val!=='*') {
543 // If they have told us that this is a "plaintext" search then we want the plaintext of the node - right?
544 if ($key == "plaintext") {
545 // $node->plaintext actually returns $node->text();
546 $nodeKeyValue = $node->text();
547 } else {
548 // this is a normal search, we want the value of that attribute of the tag.
549 $nodeKeyValue = $node->attr[$key];
550 }
551 if (is_object($debugObject)) {$debugObject->debugLog(2, "testing node: " . $node->tag . " for attribute: " . $key . $exp . $val . " where nodes value is: " . $nodeKeyValue);}
552
553 //PaperG - If lowercase is set, do a case insensitive test of the value of the selector.
554 if ($lowercase) {
555 $check = $this->match($exp, strtolower($val), strtolower($nodeKeyValue));
556 } else {
557 $check = $this->match($exp, $val, $nodeKeyValue);
558 }
559 if (is_object($debugObject)) {$debugObject->debugLog(2, "after match: " . ($check ? "true" : "false"));}
560
561 // handle multiple class
562 if (!$check && strcasecmp($key, 'class')===0) {
563 foreach (explode(' ',$node->attr[$key]) as $k) {
564 // Without this, there were cases where leading, trailing, or double spaces lead to our comparing blanks - bad form.
565 if (!empty($k)) {
566 if ($lowercase) {
567 $check = $this->match($exp, strtolower($val), strtolower($k));
568 } else {
569 $check = $this->match($exp, $val, $k);
570 }
571 if ($check) break;
572 }
573 }
574 }
575 if (!$check) $pass = false;
576 }
577 if ($pass) $ret[$i] = 1;
578 unset($node);
579 }
580 // It's passed by reference so this is actually what this function returns.
581 if (is_object($debugObject)) {$debugObject->debugLog(1, "EXIT - ret: ", $ret);}
582 }
583
584 protected function match($exp, $pattern, $value) {
585 global $debugObject;
586 if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
587
588 switch ($exp) {
589 case '=':
590 return ($value===$pattern);
591 case '!=':
592 return ($value!==$pattern);
593 case '^=':
594 return preg_match("/^".preg_quote($pattern,'/')."/", $value);
595 case '$=':
596 return preg_match("/".preg_quote($pattern,'/')."$/", $value);
597 case '*=':
598 if ($pattern[0]=='/') {
599 return preg_match($pattern, $value);
600 }
601 return preg_match("/".$pattern."/i", $value);
602 }
603 return false;
604 }
605
606 protected function parse_selector($selector_string) {
607 global $debugObject;
608 if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
609
610 // pattern of CSS selectors, modified from mootools
611 // Paperg: Add the colon to the attrbute, so that it properly finds <tag attr:ibute="something" > like google does.
612 // Note: if you try to look at this attribute, yo MUST use getAttribute since $dom->x:y will fail the php syntax check.
613 // Notice the \[ starting the attbute? and the @? following? This implies that an attribute can begin with an @ sign that is not captured.
614 // This implies that an html attribute specifier may start with an @ sign that is NOT captured by the expression.
615 // farther study is required to determine of this should be documented or removed.
616 // $pattern = "/([\w-:\*]*)(?:\#([\w-]+)|\.([\w-]+))?(?:\[@?(!?[\w-]+)(?:([!*^$]?=)[\"']?(.*?)[\"']?)?\])?([\/, ]+)/is";
617 $pattern = "/([\w-:\*]*)(?:\#([\w-]+)|\.([\w-]+))?(?:\[@?(!?[\w-:]+)(?:([!*^$]?=)[\"']?(.*?)[\"']?)?\])?([\/, ]+)/is";
618 preg_match_all($pattern, trim($selector_string).' ', $matches, PREG_SET_ORDER);
619 if (is_object($debugObject)) {$debugObject->debugLog(2, "Matches Array: ", $matches);}
620
621 $selectors = array();
622 $result = array();
623 //print_r($matches);
624
625 foreach ($matches as $m) {
626 $m[0] = trim($m[0]);
627 if ($m[0]==='' || $m[0]==='/' || $m[0]==='//') continue;
628 // for browser generated xpath
629 if ($m[1]==='tbody') continue;
630
631 list($tag, $key, $val, $exp, $no_key) = array($m[1], null, null, '=', false);
632 if (!empty($m[2])) {$key='id'; $val=$m[2];}
633 if (!empty($m[3])) {$key='class'; $val=$m[3];}
634 if (!empty($m[4])) {$key=$m[4];}
635 if (!empty($m[5])) {$exp=$m[5];}
636 if (!empty($m[6])) {$val=$m[6];}
637
638 // convert to lowercase
639 if ($this->dom->lowercase) {$tag=strtolower($tag); $key=strtolower($key);}
640 //elements that do NOT have the specified attribute
641 if (isset($key[0]) && $key[0]==='!') {$key=substr($key, 1); $no_key=true;}
642
643 $result[] = array($tag, $key, $val, $exp, $no_key);
644 if (trim($m[7])===',') {
645 $selectors[] = $result;
646 $result = array();
647 }
648 }
649 if (count($result)>0)
650 $selectors[] = $result;
651 return $selectors;
652 }
653
654 function __get($name) {
655 if (isset($this->attr[$name]))
656 {
657 return $this->convert_text($this->attr[$name]);
658 }
659 switch ($name) {
660 case 'outertext': return $this->outertext();
661 case 'innertext': return $this->innertext();
662 case 'plaintext': return $this->text();
663 case 'xmltext': return $this->xmltext();
664 default: return array_key_exists($name, $this->attr);
665 }
666 }
667
668 function __set($name, $value) {
669 switch ($name) {
670 case 'outertext': return $this->_[HDOM_INFO_OUTER] = $value;
671 case 'innertext':
672 if (isset($this->_[HDOM_INFO_TEXT])) return $this->_[HDOM_INFO_TEXT] = $value;
673 return $this->_[HDOM_INFO_INNER] = $value;
674 }
675 if (!isset($this->attr[$name])) {
676 $this->_[HDOM_INFO_SPACE][] = array(' ', '', '');
677 $this->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_DOUBLE;
678 }
679 $this->attr[$name] = $value;
680 }
681
682 function __isset($name) {
683 switch ($name) {
684 case 'outertext': return true;
685 case 'innertext': return true;
686 case 'plaintext': return true;
687 }
688 //no value attr: nowrap, checked selected...
689 return (array_key_exists($name, $this->attr)) ? true : isset($this->attr[$name]);
690 }
691
692 function __unset($name) {
693 if (isset($this->attr[$name]))
694 unset($this->attr[$name]);
695 }
696
697 // PaperG - Function to convert the text from one character set to another if the two sets are not the same.
698 function convert_text($text) {
699 global $debugObject;
700 if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
701
702 $converted_text = $text;
703
704 $sourceCharset = "";
705 $targetCharset = "";
706 if ($this->dom) {
707 $sourceCharset = strtoupper($this->dom->_charset);
708 $targetCharset = strtoupper($this->dom->_target_charset);
709 }
710 if (is_object($debugObject)) {$debugObject->debugLog(3, "source charset: " . $sourceCharset . " target charaset: " . $targetCharset);}
711
712 if (!empty($sourceCharset) && !empty($targetCharset) && (strcasecmp($sourceCharset, $targetCharset) != 0))
713 {
714 // Check if the reported encoding could have been incorrect and the text is actually already UTF-8
715 if ((strcasecmp($targetCharset, 'UTF-8') == 0) && ($this->is_utf8($text)))
716 {
717 $converted_text = $text;
718 }
719 else
720 {
721 $converted_text = iconv($sourceCharset, $targetCharset, $text);
722 }
723 }
724
725 return $converted_text;
726 }
727
728 function is_utf8($string)
729 {
730 return (utf8_encode(utf8_decode($string)) == $string);
731 }
732
733 // camel naming conventions
734 function getAllAttributes() {return $this->attr;}
735 function getAttribute($name) {return $this->__get($name);}
736 function setAttribute($name, $value) {$this->__set($name, $value);}
737 function hasAttribute($name) {return $this->__isset($name);}
738 function removeAttribute($name) {$this->__set($name, null);}
739 function getElementById($id) {return $this->find("#$id", 0);}
740 function getElementsById($id, $idx=null) {return $this->find("#$id", $idx);}
741 function getElementByTagName($name) {return $this->find($name, 0);}
742 function getElementsByTagName($name, $idx=null) {return $this->find($name, $idx);}
743 function parentNode() {return $this->parent();}
744 function childNodes($idx=-1) {return $this->children($idx);}
745 function firstChild() {return $this->first_child();}
746 function lastChild() {return $this->last_child();}
747 function nextSibling() {return $this->next_sibling();}
748 function previousSibling() {return $this->prev_sibling();}
749 }
750
751 /**
752 * simple html dom parser
753 * Paperg - in the find routine: allow us to specify that we want case insensitive testing of the value of the selector.
754 * Paperg - change $size from protected to public so we can easily access it
755 * Paperg - added ForceTagsClosed in the constructor which tells us whether we trust the html or not. Default is to NOT trust it.
756 *
757 * @package PlaceLocalInclude
758 */
759 class simple_html_dom {
760 public $root = null;
761 public $nodes = array();
762 public $callback = null;
763 public $lowercase = false;
764 public $size;
765 protected $pos;
766 protected $doc;
767 protected $char;
768 protected $cursor;
769 protected $parent;
770 protected $noise = array();
771 protected $token_blank = " \t\r\n";
772 protected $token_equal = ' =/>';
773 protected $token_slash = " />\r\n\t";
774 protected $token_attr = ' >';
775 protected $_charset = '';
776 protected $_target_charset = '';
777 protected $default_br_text = "";
778
779 // use isset instead of in_array, performance boost about 30%...
780 protected $self_closing_tags = array('img'=>1, 'br'=>1, 'input'=>1, 'meta'=>1, 'link'=>1, 'hr'=>1, 'base'=>1, 'embed'=>1, 'spacer'=>1);
781 protected $block_tags = array('root'=>1, 'body'=>1, 'form'=>1, 'div'=>1, 'span'=>1, 'table'=>1);
782 // Known sourceforge issue #2977341
783 // B tags that are not closed cause us to return everything to the end of the document.
784 protected $optional_closing_tags = array(
785 'tr'=>array('tr'=>1, 'td'=>1, 'th'=>1),
786 'th'=>array('th'=>1),
787 'td'=>array('td'=>1),
788 'li'=>array('li'=>1),
789 'dt'=>array('dt'=>1, 'dd'=>1),
790 'dd'=>array('dd'=>1, 'dt'=>1),
791 'dl'=>array('dd'=>1, 'dt'=>1),
792 'p'=>array('p'=>1),
793 'nobr'=>array('nobr'=>1),
794 'b'=>array('b'=>1),
795 );
796
797 function __construct($str=null, $lowercase=true, $forceTagsClosed=true, $target_charset=DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT) {
798 if ($str) {
799 if (preg_match("/^http:\/\//i",$str) || is_file($str))
800 $this->load_file($str);
801 else
802 $this->load($str, $lowercase, $stripRN, $defaultBRText);
803 }
804 // Forcing tags to be closed implies that we don't trust the html, but it can lead to parsing errors if we SHOULD trust the html.
805 if (!$forceTagsClosed) {
806 $this->optional_closing_array=array();
807 }
808 $this->_target_charset = $target_charset;
809 }
810
811 function __destruct() {
812 $this->clear();
813 }
814
815 // load html from string
816 function load($str, $lowercase=true, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT) {
817 global $debugObject;
818
819 // prepare
820 $this->prepare($str, $lowercase, $stripRN, $defaultBRText);
821 // strip out comments
822 $this->remove_noise("'<!--(.*?)-->'is");
823 // strip out cdata
824 $this->remove_noise("'<!\[CDATA\[(.*?)\]\]>'is", true);
825 // Per sourceforge http://sourceforge.net/tracker/?func=detail&aid=2949097&group_id=218559&atid=1044037
826 // Script tags removal now preceeds style tag removal.
827 // strip out <script> tags
828 $this->remove_noise("'<\s*script[^>]*[^/]>(.*?)<\s*/\s*script\s*>'is");
829 $this->remove_noise("'<\s*script\s*>(.*?)<\s*/\s*script\s*>'is");
830 // strip out <style> tags
831 $this->remove_noise("'<\s*style[^>]*[^/]>(.*?)<\s*/\s*style\s*>'is");
832 $this->remove_noise("'<\s*style\s*>(.*?)<\s*/\s*style\s*>'is");
833 // strip out preformatted tags
834 $this->remove_noise("'<\s*(?:code)[^>]*>(.*?)<\s*/\s*(?:code)\s*>'is");
835 // strip out server side scripts
836 $this->remove_noise("'(<\?)(.*?)(\?>)'s", true);
837 // strip smarty scripts
838 $this->remove_noise("'(\{\w)(.*?)(\})'s", true);
839
840 // parsing
841 while ($this->parse());
842 // end
843 $this->root->_[HDOM_INFO_END] = $this->cursor;
844 $this->parse_charset();
845 }
846
847 // load html from file
848 function load_file() {
849 $args = func_get_args();
850 $this->load(call_user_func_array('file_get_contents', $args), true);
851 // Per the simple_html_dom repositiry this is a planned upgrade to the codebase.
852 // Throw an error if we can't properly load the dom.
853 if (($error=error_get_last())!==null) {
854 $this->clear();
855 return false;
856 }
857 }
858
859 // set callback function
860 function set_callback($function_name) {
861 $this->callback = $function_name;
862 }
863
864 // remove callback function
865 function remove_callback() {
866 $this->callback = null;
867 }
868
869 // save dom as string
870 function save($filepath='') {
871 $ret = $this->root->innertext();
872 if ($filepath!=='') file_put_contents($filepath, $ret, LOCK_EX);
873 return $ret;
874 }
875
876 // find dom node by css selector
877 // Paperg - allow us to specify that we want case insensitive testing of the value of the selector.
878 function find($selector, $idx=null, $lowercase=false) {
879 return $this->root->find($selector, $idx, $lowercase);
880 }
881
882 // clean up memory due to php5 circular references memory leak...
883 function clear() {
884 foreach ($this->nodes as $n) {$n->clear(); $n = null;}
885 // This add next line is documented in the sourceforge repository. 2977248 as a fix for ongoing memory leaks that occur even with the use of clear.
886 if (isset($this->children)) foreach ($this->children as $n) {$n->clear(); $n = null;}
887 if (isset($this->parent)) {$this->parent->clear(); unset($this->parent);}
888 if (isset($this->root)) {$this->root->clear(); unset($this->root);}
889 unset($this->doc);
890 unset($this->noise);
891 }
892
893 function dump($show_attr=true) {
894 $this->root->dump($show_attr);
895 }
896
897 // prepare HTML data and init everything
898 protected function prepare($str, $lowercase=true, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT) {
899 $this->clear();
900
901 // set the length of content before we do anything to it.
902 $this->size = strlen($str);
903
904 //before we save the string as the doc... strip out the \r \n's if we are told to.
905 if ($stripRN) {
906 $str = str_replace("\r", " ", $str);
907 $str = str_replace("\n", " ", $str);
908 }
909
910 $this->doc = $str;
911 $this->pos = 0;
912 $this->cursor = 1;
913 $this->noise = array();
914 $this->nodes = array();
915 $this->lowercase = $lowercase;
916 $this->default_br_text = $defaultBRText;
917 $this->root = new simple_html_dom_node($this);
918 $this->root->tag = 'root';
919 $this->root->_[HDOM_INFO_BEGIN] = -1;
920 $this->root->nodetype = HDOM_TYPE_ROOT;
921 $this->parent = $this->root;
922 if ($this->size>0) $this->char = $this->doc[0];
923 }
924
925 // parse html content
926 protected function parse() {
927 if (($s = $this->copy_until_char('<'))==='')
928 return $this->read_tag();
929
930 // text
931 $node = new simple_html_dom_node($this);
932 ++$this->cursor;
933 $node->_[HDOM_INFO_TEXT] = $s;
934 $this->link_nodes($node, false);
935 return true;
936 }
937
938 // PAPERG - dkchou - added this to try to identify the character set of the page we have just parsed so we know better how to spit it out later.
939 // NOTE: IF you provide a routine called get_last_retrieve_url_contents_content_type which returns the CURLINFO_CONTENT_TYPE fromt he last curl_exec
940 // (or the content_type header fromt eh last transfer), we will parse THAT, and if a charset is specified, we will use it over any other mechanism.
941 protected function parse_charset()
942 {
943 global $debugObject;
944
945 $charset = null;
946
947 if (function_exists('get_last_retrieve_url_contents_content_type'))
948 {
949 $contentTypeHeader = get_last_retrieve_url_contents_content_type();
950 $success = preg_match('/charset=(.+)/', $contentTypeHeader, $matches);
951 if ($success)
952 {
953 $charset = $matches[1];
954 if (is_object($debugObject)) {$debugObject->debugLog(2, 'header content-type found charset of: ' . $charset);}
955 }
956
957 }
958
959 if (empty($charset))
960 {
961 $el = $this->root->find('meta[http-equiv=Content-Type]',0);
962 if (!empty($el))
963 {
964 $fullvalue = $el->content;
965 if (is_object($debugObject)) {$debugObject->debugLog(2, 'meta content-type tag found' . $fullValue);}
966
967 if (!empty($fullvalue))
968 {
969 $success = preg_match('/charset=(.+)/', $fullvalue, $matches);
970 if ($success)
971 {
972 $charset = $matches[1];
973 }
974 else
975 {
976 // If there is a meta tag, and they don't specify the character set, research says that it's typically ISO-8859-1
977 if (is_object($debugObject)) {$debugObject->debugLog(2, 'meta content-type tag couldn\'t be parsed. using iso-8859 default.');}
978 $charset = 'ISO-8859-1';
979 }
980 }
981 }
982 }
983
984 // If we couldn't find a charset above, then lets try to detect one based on the text we got...
985 if (empty($charset))
986 {
987 // Have php try to detect the encoding from the text given to us.
988 $charset = mb_detect_encoding($this->root->plaintext . "ascii", $encoding_list = array( "UTF-8", "CP1252" ) );
989 if (is_object($debugObject)) {$debugObject->debugLog(2, 'mb_detect found: ' . $charset);}
990
991 // and if this doesn't work... then we need to just wrongheadedly assume it's UTF-8 so that we can move on - cause this will usually give us most of what we need...
992 if ($charset === false)
993 {
994 if (is_object($debugObject)) {$debugObject->debugLog(2, 'since mb_detect failed - using default of utf-8');}
995 $charset = 'UTF-8';
996 }
997 }
998
999 // Since CP1252 is a superset, if we get one of it's subsets, we want it instead.
1000 if ((strtolower($charset) == strtolower('ISO-8859-1')) || (strtolower($charset) == strtolower('Latin1')) || (strtolower($charset) == strtolower('Latin-1')))
1001 {
1002 if (is_object($debugObject)) {$debugObject->debugLog(2, 'replacing ' . $charset . ' with CP1252 as its a superset');}
1003 $charset = 'CP1252';
1004 }
1005
1006 if (is_object($debugObject)) {$debugObject->debugLog(1, 'EXIT - ' . $charset);}
1007
1008 return $this->_charset = $charset;
1009 }
1010
1011 // read tag info
1012 protected function read_tag() {
1013 if ($this->char!=='<') {
1014 $this->root->_[HDOM_INFO_END] = $this->cursor;
1015 return false;
1016 }
1017 $begin_tag_pos = $this->pos;
1018 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1019
1020 // end tag
1021 if ($this->char==='/') {
1022 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1023 // This represetns the change in the simple_html_dom trunk from revision 180 to 181.
1024 // $this->skip($this->token_blank_t);
1025 $this->skip($this->token_blank);
1026 $tag = $this->copy_until_char('>');
1027
1028 // skip attributes in end tag
1029 if (($pos = strpos($tag, ' '))!==false)
1030 $tag = substr($tag, 0, $pos);
1031
1032 $parent_lower = strtolower($this->parent->tag);
1033 $tag_lower = strtolower($tag);
1034
1035 if ($parent_lower!==$tag_lower) {
1036 if (isset($this->optional_closing_tags[$parent_lower]) && isset($this->block_tags[$tag_lower])) {
1037 $this->parent->_[HDOM_INFO_END] = 0;
1038 $org_parent = $this->parent;
1039
1040 while (($this->parent->parent) && strtolower($this->parent->tag)!==$tag_lower)
1041 $this->parent = $this->parent->parent;
1042
1043 if (strtolower($this->parent->tag)!==$tag_lower) {
1044 $this->parent = $org_parent; // restore origonal parent
1045 if ($this->parent->parent) $this->parent = $this->parent->parent;
1046 $this->parent->_[HDOM_INFO_END] = $this->cursor;
1047 return $this->as_text_node($tag);
1048 }
1049 }
1050 else if (($this->parent->parent) && isset($this->block_tags[$tag_lower])) {
1051 $this->parent->_[HDOM_INFO_END] = 0;
1052 $org_parent = $this->parent;
1053
1054 while (($this->parent->parent) && strtolower($this->parent->tag)!==$tag_lower)
1055 $this->parent = $this->parent->parent;
1056
1057 if (strtolower($this->parent->tag)!==$tag_lower) {
1058 $this->parent = $org_parent; // restore origonal parent
1059 $this->parent->_[HDOM_INFO_END] = $this->cursor;
1060 return $this->as_text_node($tag);
1061 }
1062 }
1063 else if (($this->parent->parent) && strtolower($this->parent->parent->tag)===$tag_lower) {
1064 $this->parent->_[HDOM_INFO_END] = 0;
1065 $this->parent = $this->parent->parent;
1066 }
1067 else
1068 return $this->as_text_node($tag);
1069 }
1070
1071 $this->parent->_[HDOM_INFO_END] = $this->cursor;
1072 if ($this->parent->parent) $this->parent = $this->parent->parent;
1073
1074 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1075 return true;
1076 }
1077
1078 $node = new simple_html_dom_node($this);
1079 $node->_[HDOM_INFO_BEGIN] = $this->cursor;
1080 ++$this->cursor;
1081 $tag = $this->copy_until($this->token_slash);
1082 $node->tag_start = $begin_tag_pos;
1083
1084 // doctype, cdata & comments...
1085 if (isset($tag[0]) && $tag[0]==='!') {
1086 $node->_[HDOM_INFO_TEXT] = '<' . $tag . $this->copy_until_char('>');
1087
1088 if (isset($tag[2]) && $tag[1]==='-' && $tag[2]==='-') {
1089 $node->nodetype = HDOM_TYPE_COMMENT;
1090 $node->tag = 'comment';
1091 } else {
1092 $node->nodetype = HDOM_TYPE_UNKNOWN;
1093 $node->tag = 'unknown';
1094 }
1095 if ($this->char==='>') $node->_[HDOM_INFO_TEXT].='>';
1096 $this->link_nodes($node, true);
1097 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1098 return true;
1099 }
1100
1101 // text
1102 if ($pos=strpos($tag, '<')!==false) {
1103 $tag = '<' . substr($tag, 0, -1);
1104 $node->_[HDOM_INFO_TEXT] = $tag;
1105 $this->link_nodes($node, false);
1106 $this->char = $this->doc[--$this->pos]; // prev
1107 return true;
1108 }
1109
1110 if (!preg_match("/^[\w-:]+$/", $tag)) {
1111 $node->_[HDOM_INFO_TEXT] = '<' . $tag . $this->copy_until('<>');
1112 if ($this->char==='<') {
1113 $this->link_nodes($node, false);
1114 return true;
1115 }
1116
1117 if ($this->char==='>') $node->_[HDOM_INFO_TEXT].='>';
1118 $this->link_nodes($node, false);
1119 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1120 return true;
1121 }
1122
1123 // begin tag
1124 $node->nodetype = HDOM_TYPE_ELEMENT;
1125 $tag_lower = strtolower($tag);
1126 $node->tag = ($this->lowercase) ? $tag_lower : $tag;
1127
1128 // handle optional closing tags
1129 if (isset($this->optional_closing_tags[$tag_lower]) ) {
1130 while (isset($this->optional_closing_tags[$tag_lower][strtolower($this->parent->tag)])) {
1131 $this->parent->_[HDOM_INFO_END] = 0;
1132 $this->parent = $this->parent->parent;
1133 }
1134 $node->parent = $this->parent;
1135 }
1136
1137 $guard = 0; // prevent infinity loop
1138 $space = array($this->copy_skip($this->token_blank), '', '');
1139
1140 // attributes
1141 do
1142 {
1143 if ($this->char!==null && $space[0]==='') break;
1144 $name = $this->copy_until($this->token_equal);
1145 if ($guard===$this->pos) {
1146 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1147 continue;
1148 }
1149 $guard = $this->pos;
1150
1151 // handle endless '<'
1152 if ($this->pos>=$this->size-1 && $this->char!=='>') {
1153 $node->nodetype = HDOM_TYPE_TEXT;
1154 $node->_[HDOM_INFO_END] = 0;
1155 $node->_[HDOM_INFO_TEXT] = '<'.$tag . $space[0] . $name;
1156 $node->tag = 'text';
1157 $this->link_nodes($node, false);
1158 return true;
1159 }
1160
1161 // handle mismatch '<'
1162 if ($this->doc[$this->pos-1]=='<') {
1163 $node->nodetype = HDOM_TYPE_TEXT;
1164 $node->tag = 'text';
1165 $node->attr = array();
1166 $node->_[HDOM_INFO_END] = 0;
1167 $node->_[HDOM_INFO_TEXT] = substr($this->doc, $begin_tag_pos, $this->pos-$begin_tag_pos-1);
1168 $this->pos -= 2;
1169 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1170 $this->link_nodes($node, false);
1171 return true;
1172 }
1173
1174 if ($name!=='/' && $name!=='') {
1175 $space[1] = $this->copy_skip($this->token_blank);
1176 $name = $this->restore_noise($name);
1177 if ($this->lowercase) $name = strtolower($name);
1178 if ($this->char==='=') {
1179 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1180 $this->parse_attr($node, $name, $space);
1181 }
1182 else {
1183 //no value attr: nowrap, checked selected...
1184 $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_NO;
1185 $node->attr[$name] = true;
1186 if ($this->char!='>') $this->char = $this->doc[--$this->pos]; // prev
1187 }
1188 $node->_[HDOM_INFO_SPACE][] = $space;
1189 $space = array($this->copy_skip($this->token_blank), '', '');
1190 }
1191 else
1192 break;
1193 } while ($this->char!=='>' && $this->char!=='/');
1194
1195 $this->link_nodes($node, true);
1196 $node->_[HDOM_INFO_ENDSPACE] = $space[0];
1197
1198 // check self closing
1199 if ($this->copy_until_char_escape('>')==='/') {
1200 $node->_[HDOM_INFO_ENDSPACE] .= '/';
1201 $node->_[HDOM_INFO_END] = 0;
1202 }
1203 else {
1204 // reset parent
1205 if (!isset($this->self_closing_tags[strtolower($node->tag)])) $this->parent = $node;
1206 }
1207 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1208
1209 // If it's a BR tag, we need to set it's text to the default text.
1210 // This way when we see it in plaintext, we can generate formatting that the user wants.
1211 if ($node->tag == "br") {
1212 $node->_[HDOM_INFO_INNER] = $this->default_br_text;
1213 }
1214
1215 return true;
1216 }
1217
1218 // parse attributes
1219 protected function parse_attr($node, $name, &$space) {
1220 // Per sourceforge: http://sourceforge.net/tracker/?func=detail&aid=3061408&group_id=218559&atid=1044037
1221 // If the attribute is already defined inside a tag, only pay atetntion to the first one as opposed to the last one.
1222 if (isset($node->attr[$name]))
1223 {
1224 return;
1225 }
1226
1227 $space[2] = $this->copy_skip($this->token_blank);
1228 switch ($this->char) {
1229 case '"':
1230 $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_DOUBLE;
1231 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1232 $node->attr[$name] = $this->restore_noise($this->copy_until_char_escape('"'));
1233 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1234 break;
1235 case '\'':
1236 $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_SINGLE;
1237 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1238 $node->attr[$name] = $this->restore_noise($this->copy_until_char_escape('\''));
1239 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1240 break;
1241 default:
1242 $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_NO;
1243 $node->attr[$name] = $this->restore_noise($this->copy_until($this->token_attr));
1244 }
1245 // PaperG: Attributes should not have \r or \n in them, that counts as html whitespace.
1246 $node->attr[$name] = str_replace("\r", "", $node->attr[$name]);
1247 $node->attr[$name] = str_replace("\n", "", $node->attr[$name]);
1248 // PaperG: If this is a "class" selector, lets get rid of the preceeding and trailing space since some people leave it in the multi class case.
1249 if ($name == "class") {
1250 $node->attr[$name] = trim($node->attr[$name]);
1251 }
1252 }
1253
1254 // link node's parent
1255 protected function link_nodes(&$node, $is_child) {
1256 $node->parent = $this->parent;
1257 $this->parent->nodes[] = $node;
1258 if ($is_child)
1259 $this->parent->children[] = $node;
1260 }
1261
1262 // as a text node
1263 protected function as_text_node($tag) {
1264 $node = new simple_html_dom_node($this);
1265 ++$this->cursor;
1266 $node->_[HDOM_INFO_TEXT] = '</' . $tag . '>';
1267 $this->link_nodes($node, false);
1268 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1269 return true;
1270 }
1271
1272 protected function skip($chars) {
1273 $this->pos += strspn($this->doc, $chars, $this->pos);
1274 $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1275 }
1276
1277 protected function copy_skip($chars) {
1278 $pos = $this->pos;
1279 $len = strspn($this->doc, $chars, $pos);
1280 $this->pos += $len;
1281 $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1282 if ($len===0) return '';
1283 return substr($this->doc, $pos, $len);
1284 }
1285
1286 protected function copy_until($chars) {
1287 $pos = $this->pos;
1288 $len = strcspn($this->doc, $chars, $pos);
1289 $this->pos += $len;
1290 $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1291 return substr($this->doc, $pos, $len);
1292 }
1293
1294 protected function copy_until_char($char) {
1295 if ($this->char===null) return '';
1296
1297 if (($pos = strpos($this->doc, $char, $this->pos))===false) {
1298 $ret = substr($this->doc, $this->pos, $this->size-$this->pos);
1299 $this->char = null;
1300 $this->pos = $this->size;
1301 return $ret;
1302 }
1303
1304 if ($pos===$this->pos) return '';
1305 $pos_old = $this->pos;
1306 $this->char = $this->doc[$pos];
1307 $this->pos = $pos;
1308 return substr($this->doc, $pos_old, $pos-$pos_old);
1309 }
1310
1311 protected function copy_until_char_escape($char) {
1312 if ($this->char===null) return '';
1313
1314 $start = $this->pos;
1315 while (1) {
1316 if (($pos = strpos($this->doc, $char, $start))===false) {
1317 $ret = substr($this->doc, $this->pos, $this->size-$this->pos);
1318 $this->char = null;
1319 $this->pos = $this->size;
1320 return $ret;
1321 }
1322
1323 if ($pos===$this->pos) return '';
1324
1325 if ($this->doc[$pos-1]==='\\') {
1326 $start = $pos+1;
1327 continue;
1328 }
1329
1330 $pos_old = $this->pos;
1331 $this->char = $this->doc[$pos];
1332 $this->pos = $pos;
1333 return substr($this->doc, $pos_old, $pos-$pos_old);
1334 }
1335 }
1336
1337 // remove noise from html content
1338 protected function remove_noise($pattern, $remove_tag=false) {
1339 $count = preg_match_all($pattern, $this->doc, $matches, PREG_SET_ORDER|PREG_OFFSET_CAPTURE);
1340
1341 for ($i=$count-1; $i>-1; --$i) {
1342 $key = '___noise___'.sprintf('% 3d', count($this->noise)+100);
1343 $idx = ($remove_tag) ? 0 : 1;
1344 $this->noise[$key] = $matches[$i][$idx][0];
1345 $this->doc = substr_replace($this->doc, $key, $matches[$i][$idx][1], strlen($matches[$i][$idx][0]));
1346 }
1347
1348 // reset the length of content
1349 $this->size = strlen($this->doc);
1350 if ($this->size>0) $this->char = $this->doc[0];
1351 }
1352
1353 // restore noise to html content
1354 function restore_noise($text) {
1355 while (($pos=strpos($text, '___noise___'))!==false) {
1356 $key = '___noise___'.$text[$pos+11].$text[$pos+12].$text[$pos+13];
1357 if (isset($this->noise[$key]))
1358 $text = substr($text, 0, $pos).$this->noise[$key].substr($text, $pos+14);
1359 }
1360 return $text;
1361 }
1362
1363 function __toString() {
1364 return $this->root->innertext();
1365 }
1366
1367 function __get($name) {
1368 switch ($name) {
1369 case 'outertext':
1370 return $this->root->innertext();
1371 case 'innertext':
1372 return $this->root->innertext();
1373 case 'plaintext':
1374 return $this->root->text();
1375 case 'charset':
1376 return $this->_charset;
1377 case 'target_charset':
1378 return $this->_target_charset;
1379 }
1380 }
1381
1382 // camel naming conventions
1383 function childNodes($idx=-1) {return $this->root->childNodes($idx);}
1384 function firstChild() {return $this->root->first_child();}
1385 function lastChild() {return $this->root->last_child();}
1386 function getElementById($id) {return $this->find("#$id", 0);}
1387 function getElementsById($id, $idx=null) {return $this->find("#$id", $idx);}
1388 function getElementByTagName($name) {return $this->find($name, 0);}
1389 function getElementsByTagName($name, $idx=-1) {return $this->find($name, $idx);}
1390 function loadFile() {$args = func_get_args();$this->load_file($args);}
1391 }
1392
1393 ?>