API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Jan 27th, 2016
87
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 63.51 KB | None | 0 0
raw download clone embed print report
  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 $dom->size is the "real" number of bytes the dom was created from.
  17. * but for most purposes, it's a really good estimation.
  18. * Paperg - Added the forceTagsClosed to the dom constructor. Forcing tags closed is great for malformed html, but it CAN lead to parsing errors.
  19. * Allow the user to tell us how much they trust the html.
  20. * Paperg add the text and plaintext to the selectors for the find syntax. plaintext implies text in the innertext of a node. text implies that the tag is a text node.
  21. * This allows for us to find tags based on the text they contain.
  22. * Create find_ancestor_tag to see if a tag is - at any level - inside of another specific tag.
  23. * Paperg: added parse_charset so that we know about the character set of the source document.
  24. * NOTE: If the user's system has a routine called get_last_retrieve_url_contents_content_type availalbe, we will assume it's returning the content-type header from the
  25. * last transfer or curl_exec, and we will parse that and use it in preference to any other method of charset detection.
  26. *
  27. * Found infinite loop in the case of broken html in restore_noise. Rewrote to protect from that.
  28. * PaperG (John Schlick) Added get_display_size for "IMG" tags.
  29. *
  30. * Licensed under The MIT License
  31. * Redistributions of files must retain the above copyright notice.
  32. *
  33. * @author S.C. Chen <me578022@gmail.com>
  34. * @author John Schlick
  35. * @author Rus Carroll
  36. * @version 1.5 ($Rev: 196 $)
  37. * @package PlaceLocalInclude
  38. * @subpackage simple_html_dom
  39. */
  40.  
  41. /**
  42. * All of the Defines for the classes below.
  43. * @author S.C. Chen <me578022@gmail.com>
  44. */
  45. define('HDOM_TYPE_ELEMENT', 1);
  46. define('HDOM_TYPE_COMMENT', 2);
  47. define('HDOM_TYPE_TEXT', 3);
  48. define('HDOM_TYPE_ENDTAG', 4);
  49. define('HDOM_TYPE_ROOT', 5);
  50. define('HDOM_TYPE_UNKNOWN', 6);
  51. define('HDOM_QUOTE_DOUBLE', 0);
  52. define('HDOM_QUOTE_SINGLE', 1);
  53. define('HDOM_QUOTE_NO', 3);
  54. define('HDOM_INFO_BEGIN', 0);
  55. define('HDOM_INFO_END', 1);
  56. define('HDOM_INFO_QUOTE', 2);
  57. define('HDOM_INFO_SPACE', 3);
  58. define('HDOM_INFO_TEXT', 4);
  59. define('HDOM_INFO_INNER', 5);
  60. define('HDOM_INFO_OUTER', 6);
  61. define('HDOM_INFO_ENDSPACE',7);
  62. define('DEFAULT_TARGET_CHARSET', 'UTF-8');
  63. define('DEFAULT_BR_TEXT', "\r\n");
  64. define('DEFAULT_SPAN_TEXT', " ");
  65. define('MAX_FILE_SIZE', 600000);
  66. // helper functions
  67. // -----------------------------------------------------------------------------
  68. // get html dom from file
  69. // $maxlen is defined in the code as PHP_STREAM_COPY_ALL which is defined as -1.
  70. function file_get_html($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, $defaultSpanText=DEFAULT_SPAN_TEXT)
  71. {
  72. // We DO force the tags to be terminated.
  73. $dom = new simple_html_dom(null, $lowercase, $forceTagsClosed, $target_charset, $stripRN, $defaultBRText, $defaultSpanText);
  74. // For sourceforge users: uncomment the next line and comment the retreive_url_contents line 2 lines down if it is not already done.
  75. $contents = file_get_contents($url, $use_include_path, $context, $offset);
  76. // Paperg - use our own mechanism for getting the contents as we want to control the timeout.
  77. //$contents = retrieve_url_contents($url);
  78. if (empty($contents) || strlen($contents) > MAX_FILE_SIZE)
  79. {
  80. return false;
  81. }
  82. // The second parameter can force the selectors to all be lowercase.
  83. $dom->load($contents, $lowercase, $stripRN);
  84. return $dom;
  85. }
  86.  
  87. // get html dom from string
  88. function str_get_html($str, $lowercase=true, $forceTagsClosed=true, $target_charset = DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
  89. {
  90. $dom = new simple_html_dom(null, $lowercase, $forceTagsClosed, $target_charset, $stripRN, $defaultBRText, $defaultSpanText);
  91. if (empty($str) || strlen($str) > MAX_FILE_SIZE)
  92. {
  93. $dom->clear();
  94. return false;
  95. }
  96. $dom->load($str, $lowercase, $stripRN);
  97. return $dom;
  98. }
  99.  
  100. // dump html dom tree
  101. function dump_html_tree($node, $show_attr=true, $deep=0)
  102. {
  103. $node->dump($node);
  104. }
  105.  
  106.  
  107. /**
  108. * simple html dom node
  109. * PaperG - added ability for "find" routine to lowercase the value of the selector.
  110. * PaperG - added $tag_start to track the start position of the tag in the total byte index
  111. *
  112. * @package PlaceLocalInclude
  113. */
  114. class simple_html_dom_node
  115. {
  116. public $nodetype = HDOM_TYPE_TEXT;
  117. public $tag = 'text';
  118. public $attr = array();
  119. public $children = array();
  120. public $nodes = array();
  121. public $parent = null;
  122. // The "info" array - see HDOM_INFO_... for what each element contains.
  123. public $_ = array();
  124. public $tag_start = 0;
  125. private $dom = null;
  126.  
  127. function __construct($dom)
  128. {
  129. $this->dom = $dom;
  130. $dom->nodes[] = $this;
  131. }
  132.  
  133. function __destruct()
  134. {
  135. $this->clear();
  136. }
  137.  
  138. function __toString()
  139. {
  140. return $this->outertext();
  141. }
  142.  
  143. // clean up memory due to php5 circular references memory leak...
  144. function clear()
  145. {
  146. $this->dom = null;
  147. $this->nodes = null;
  148. $this->parent = null;
  149. $this->children = null;
  150. }
  151.  
  152. // dump node's tree
  153. function dump($show_attr=true, $deep=0)
  154. {
  155. $lead = str_repeat(' ', $deep);
  156.  
  157. echo $lead.$this->tag;
  158. if ($show_attr && count($this->attr)>0)
  159. {
  160. echo '(';
  161. foreach ($this->attr as $k=>$v)
  162. echo "[$k]=>\"".$this->$k.'", ';
  163. echo ')';
  164. }
  165. echo "\n";
  166.  
  167. if ($this->nodes)
  168. {
  169. foreach ($this->nodes as $c)
  170. {
  171. $c->dump($show_attr, $deep+1);
  172. }
  173. }
  174. }
  175.  
  176.  
  177. // Debugging function to dump a single dom node with a bunch of information about it.
  178. function dump_node($echo=true)
  179. {
  180.  
  181. $string = $this->tag;
  182. if (count($this->attr)>0)
  183. {
  184. $string .= '(';
  185. foreach ($this->attr as $k=>$v)
  186. {
  187. $string .= "[$k]=>\"".$this->$k.'", ';
  188. }
  189. $string .= ')';
  190. }
  191. if (count($this->_)>0)
  192. {
  193. $string .= ' $_ (';
  194. foreach ($this->_ as $k=>$v)
  195. {
  196. if (is_array($v))
  197. {
  198. $string .= "[$k]=>(";
  199. foreach ($v as $k2=>$v2)
  200. {
  201. $string .= "[$k2]=>\"".$v2.'", ';
  202. }
  203. $string .= ")";
  204. } else {
  205. $string .= "[$k]=>\"".$v.'", ';
  206. }
  207. }
  208. $string .= ")";
  209. }
  210.  
  211. if (isset($this->text))
  212. {
  213. $string .= " text: (" . $this->text . ")";
  214. }
  215.  
  216. $string .= " HDOM_INNER_INFO: '";
  217. if (isset($node->_[HDOM_INFO_INNER]))
  218. {
  219. $string .= $node->_[HDOM_INFO_INNER] . "'";
  220. }
  221. else
  222. {
  223. $string .= ' NULL ';
  224. }
  225.  
  226. $string .= " children: " . count($this->children);
  227. $string .= " nodes: " . count($this->nodes);
  228. $string .= " tag_start: " . $this->tag_start;
  229. $string .= "\n";
  230.  
  231. if ($echo)
  232. {
  233. echo $string;
  234. return;
  235. }
  236. else
  237. {
  238. return $string;
  239. }
  240. }
  241.  
  242. // returns the parent of node
  243. // If a node is passed in, it will reset the parent of the current node to that one.
  244. function parent($parent=null)
  245. {
  246. // I am SURE that this doesn't work properly.
  247. // It fails to unset the current node from it's current parents nodes or children list first.
  248. if ($parent !== null)
  249. {
  250. $this->parent = $parent;
  251. $this->parent->nodes[] = $this;
  252. $this->parent->children[] = $this;
  253. }
  254.  
  255. return $this->parent;
  256. }
  257.  
  258. // verify that node has children
  259. function has_child()
  260. {
  261. return !empty($this->children);
  262. }
  263.  
  264. // returns children of node
  265. function children($idx=-1)
  266. {
  267. if ($idx===-1)
  268. {
  269. return $this->children;
  270. }
  271. if (isset($this->children[$idx])) return $this->children[$idx];
  272. return null;
  273. }
  274.  
  275. // returns the first child of node
  276. function first_child()
  277. {
  278. if (count($this->children)>0)
  279. {
  280. return $this->children[0];
  281. }
  282. return null;
  283. }
  284.  
  285. // returns the last child of node
  286. function last_child()
  287. {
  288. if (($count=count($this->children))>0)
  289. {
  290. return $this->children[$count-1];
  291. }
  292. return null;
  293. }
  294.  
  295. // returns the next sibling of node
  296. function next_sibling()
  297. {
  298. if ($this->parent===null)
  299. {
  300. return null;
  301. }
  302.  
  303. $idx = 0;
  304. $count = count($this->parent->children);
  305. while ($idx<$count && $this!==$this->parent->children[$idx])
  306. {
  307. ++$idx;
  308. }
  309. if (++$idx>=$count)
  310. {
  311. return null;
  312. }
  313. return $this->parent->children[$idx];
  314. }
  315.  
  316. // returns the previous sibling of node
  317. function prev_sibling()
  318. {
  319. if ($this->parent===null) return null;
  320. $idx = 0;
  321. $count = count($this->parent->children);
  322. while ($idx<$count && $this!==$this->parent->children[$idx])
  323. ++$idx;
  324. if (--$idx<0) return null;
  325. return $this->parent->children[$idx];
  326. }
  327.  
  328. // function to locate a specific ancestor tag in the path to the root.
  329. function find_ancestor_tag($tag)
  330. {
  331. global $debugObject;
  332. if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
  333.  
  334. // Start by including ourselves in the comparison.
  335. $returnDom = $this;
  336.  
  337. while (!is_null($returnDom))
  338. {
  339. if (is_object($debugObject)) { $debugObject->debugLog(2, "Current tag is: " . $returnDom->tag); }
  340.  
  341. if ($returnDom->tag == $tag)
  342. {
  343. break;
  344. }
  345. $returnDom = $returnDom->parent;
  346. }
  347. return $returnDom;
  348. }
  349.  
  350. // get dom node's inner html
  351. function innertext()
  352. {
  353. if (isset($this->_[HDOM_INFO_INNER])) return $this->_[HDOM_INFO_INNER];
  354. if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
  355.  
  356. $ret = '';
  357. foreach ($this->nodes as $n)
  358. $ret .= $n->outertext();
  359. return $ret;
  360. }
  361.  
  362. // get dom node's outer text (with tag)
  363. function outertext()
  364. {
  365. global $debugObject;
  366. if (is_object($debugObject))
  367. {
  368. $text = '';
  369. if ($this->tag == 'text')
  370. {
  371. if (!empty($this->text))
  372. {
  373. $text = " with text: " . $this->text;
  374. }
  375. }
  376. $debugObject->debugLog(1, 'Innertext of tag: ' . $this->tag . $text);
  377. }
  378.  
  379. if ($this->tag==='root') return $this->innertext();
  380.  
  381. // trigger callback
  382. if ($this->dom && $this->dom->callback!==null)
  383. {
  384. call_user_func_array($this->dom->callback, array($this));
  385. }
  386.  
  387. if (isset($this->_[HDOM_INFO_OUTER])) return $this->_[HDOM_INFO_OUTER];
  388. if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
  389.  
  390. // render begin tag
  391. if ($this->dom && $this->dom->nodes[$this->_[HDOM_INFO_BEGIN]])
  392. {
  393. $ret = $this->dom->nodes[$this->_[HDOM_INFO_BEGIN]]->makeup();
  394. } else {
  395. $ret = "";
  396. }
  397.  
  398. // render inner text
  399. if (isset($this->_[HDOM_INFO_INNER]))
  400. {
  401. // If it's a br tag... don't return the HDOM_INNER_INFO that we may or may not have added.
  402. if ($this->tag != "br")
  403. {
  404. $ret .= $this->_[HDOM_INFO_INNER];
  405. }
  406. } else {
  407. if ($this->nodes)
  408. {
  409. foreach ($this->nodes as $n)
  410. {
  411. $ret .= $this->convert_text($n->outertext());
  412. }
  413. }
  414. }
  415.  
  416. // render end tag
  417. if (isset($this->_[HDOM_INFO_END]) && $this->_[HDOM_INFO_END]!=0)
  418. $ret .= '</'.$this->tag.'>';
  419. return $ret;
  420. }
  421.  
  422. // get dom node's plain text
  423. function text()
  424. {
  425. if (isset($this->_[HDOM_INFO_INNER])) return $this->_[HDOM_INFO_INNER];
  426. switch ($this->nodetype)
  427. {
  428. case HDOM_TYPE_TEXT: return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
  429. case HDOM_TYPE_COMMENT: return '';
  430. case HDOM_TYPE_UNKNOWN: return '';
  431. }
  432. if (strcasecmp($this->tag, 'script')===0) return '';
  433. if (strcasecmp($this->tag, 'style')===0) return '';
  434.  
  435. $ret = '';
  436. // In rare cases, (always node type 1 or HDOM_TYPE_ELEMENT - observed for some span tags, and some p tags) $this->nodes is set to NULL.
  437. // NOTE: This indicates that there is a problem where it's set to NULL without a clear happening.
  438. // WHY is this happening?
  439. if (!is_null($this->nodes))
  440. {
  441. foreach ($this->nodes as $n)
  442. {
  443. $ret .= $this->convert_text($n->text());
  444. }
  445.  
  446. // If this node is a span... add a space at the end of it so multiple spans don't run into each other. This is plaintext after all.
  447. if ($this->tag == "span")
  448. {
  449. $ret .= $this->dom->default_span_text;
  450. }
  451.  
  452.  
  453. }
  454. return $ret;
  455. }
  456.  
  457. function xmltext()
  458. {
  459. $ret = $this->innertext();
  460. $ret = str_ireplace('<![CDATA[', '', $ret);
  461. $ret = str_replace(']]>', '', $ret);
  462. return $ret;
  463. }
  464.  
  465. // build node's text with tag
  466. function makeup()
  467. {
  468. // text, comment, unknown
  469. if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
  470.  
  471. $ret = '<'.$this->tag;
  472. $i = -1;
  473.  
  474. foreach ($this->attr as $key=>$val)
  475. {
  476. ++$i;
  477.  
  478. // skip removed attribute
  479. if ($val===null || $val===false)
  480. continue;
  481.  
  482. $ret .= $this->_[HDOM_INFO_SPACE][$i][0];
  483. //no value attr: nowrap, checked selected...
  484. if ($val===true)
  485. $ret .= $key;
  486. else {
  487. switch ($this->_[HDOM_INFO_QUOTE][$i])
  488. {
  489. case HDOM_QUOTE_DOUBLE: $quote = '"'; break;
  490. case HDOM_QUOTE_SINGLE: $quote = '\''; break;
  491. default: $quote = '';
  492. }
  493. $ret .= $key.$this->_[HDOM_INFO_SPACE][$i][1].'='.$this->_[HDOM_INFO_SPACE][$i][2].$quote.$val.$quote;
  494. }
  495. }
  496. $ret = $this->dom->restore_noise($ret);
  497. return $ret . $this->_[HDOM_INFO_ENDSPACE] . '>';
  498. }
  499.  
  500. // find elements by css selector
  501. //PaperG - added ability for find to lowercase the value of the selector.
  502. function find($selector, $idx=null, $lowercase=false)
  503. {
  504. $selectors = $this->parse_selector($selector);
  505. if (($count=count($selectors))===0) return array();
  506. $found_keys = array();
  507.  
  508. // find each selector
  509. for ($c=0; $c<$count; ++$c)
  510. {
  511. // The change on the below line was documented on the sourceforge code tracker id 2788009
  512. // used to be: if (($levle=count($selectors[0]))===0) return array();
  513. if (($levle=count($selectors[$c]))===0) return array();
  514. if (!isset($this->_[HDOM_INFO_BEGIN])) return array();
  515.  
  516. $head = array($this->_[HDOM_INFO_BEGIN]=>1);
  517.  
  518. // handle descendant selectors, no recursive!
  519. for ($l=0; $l<$levle; ++$l)
  520. {
  521. $ret = array();
  522. foreach ($head as $k=>$v)
  523. {
  524. $n = ($k===-1) ? $this->dom->root : $this->dom->nodes[$k];
  525. //PaperG - Pass this optional parameter on to the seek function.
  526. $n->seek($selectors[$c][$l], $ret, $lowercase);
  527. }
  528. $head = $ret;
  529. }
  530.  
  531. foreach ($head as $k=>$v)
  532. {
  533. if (!isset($found_keys[$k]))
  534. $found_keys[$k] = 1;
  535. }
  536. }
  537.  
  538. // sort keys
  539. ksort($found_keys);
  540.  
  541. $found = array();
  542. foreach ($found_keys as $k=>$v)
  543. $found[] = $this->dom->nodes[$k];
  544.  
  545. // return nth-element or array
  546. if (is_null($idx)) return $found;
  547. else if ($idx<0) $idx = count($found) + $idx;
  548. return (isset($found[$idx])) ? $found[$idx] : null;
  549. }
  550.  
  551. // seek for given conditions
  552. // PaperG - added parameter to allow for case insensitive testing of the value of a selector.
  553. protected function seek($selector, &$ret, $lowercase=false)
  554. {
  555. global $debugObject;
  556. if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
  557.  
  558. list($tag, $key, $val, $exp, $no_key) = $selector;
  559.  
  560. // xpath index
  561. if ($tag && $key && is_numeric($key))
  562. {
  563. $count = 0;
  564. foreach ($this->children as $c)
  565. {
  566. if ($tag==='*' || $tag===$c->tag) {
  567. if (++$count==$key) {
  568. $ret[$c->_[HDOM_INFO_BEGIN]] = 1;
  569. return;
  570. }
  571. }
  572. }
  573. return;
  574. }
  575.  
  576. $end = (!empty($this->_[HDOM_INFO_END])) ? $this->_[HDOM_INFO_END] : 0;
  577. if ($end==0) {
  578. $parent = $this->parent;
  579. while (!isset($parent->_[HDOM_INFO_END]) && $parent!==null) {
  580. $end -= 1;
  581. $parent = $parent->parent;
  582. }
  583. $end += $parent->_[HDOM_INFO_END];
  584. }
  585.  
  586. for ($i=$this->_[HDOM_INFO_BEGIN]+1; $i<$end; ++$i) {
  587. $node = $this->dom->nodes[$i];
  588.  
  589. $pass = true;
  590.  
  591. if ($tag==='*' && !$key) {
  592. if (in_array($node, $this->children, true))
  593. $ret[$i] = 1;
  594. continue;
  595. }
  596.  
  597. // compare tag
  598. if ($tag && $tag!=$node->tag && $tag!=='*') {$pass=false;}
  599. // compare key
  600. if ($pass && $key) {
  601. if ($no_key) {
  602. if (isset($node->attr[$key])) $pass=false;
  603. } else {
  604. if (($key != "plaintext") && !isset($node->attr[$key])) $pass=false;
  605. }
  606. }
  607. // compare value
  608. if ($pass && $key && $val && $val!=='*') {
  609. // If they have told us that this is a "plaintext" search then we want the plaintext of the node - right?
  610. if ($key == "plaintext") {
  611. // $node->plaintext actually returns $node->text();
  612. $nodeKeyValue = $node->text();
  613. } else {
  614. // this is a normal search, we want the value of that attribute of the tag.
  615. $nodeKeyValue = $node->attr[$key];
  616. }
  617. if (is_object($debugObject)) {$debugObject->debugLog(2, "testing node: " . $node->tag . " for attribute: " . $key . $exp . $val . " where nodes value is: " . $nodeKeyValue);}
  618.  
  619. //PaperG - If lowercase is set, do a case insensitive test of the value of the selector.
  620. if ($lowercase) {
  621. $check = $this->match($exp, strtolower($val), strtolower($nodeKeyValue));
  622. } else {
  623. $check = $this->match($exp, $val, $nodeKeyValue);
  624. }
  625. if (is_object($debugObject)) {$debugObject->debugLog(2, "after match: " . ($check ? "true" : "false"));}
  626.  
  627. // handle multiple class
  628. if (!$check && strcasecmp($key, 'class')===0) {
  629. foreach (explode(' ',$node->attr[$key]) as $k) {
  630. // Without this, there were cases where leading, trailing, or double spaces lead to our comparing blanks - bad form.
  631. if (!empty($k)) {
  632. if ($lowercase) {
  633. $check = $this->match($exp, strtolower($val), strtolower($k));
  634. } else {
  635. $check = $this->match($exp, $val, $k);
  636. }
  637. if ($check) break;
  638. }
  639. }
  640. }
  641. if (!$check) $pass = false;
  642. }
  643. if ($pass) $ret[$i] = 1;
  644. unset($node);
  645. }
  646. // It's passed by reference so this is actually what this function returns.
  647. if (is_object($debugObject)) {$debugObject->debugLog(1, "EXIT - ret: ", $ret);}
  648. }
  649.  
  650. protected function match($exp, $pattern, $value) {
  651. global $debugObject;
  652. if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
  653.  
  654. switch ($exp) {
  655. case '=':
  656. return ($value===$pattern);
  657. case '!=':
  658. return ($value!==$pattern);
  659. case '^=':
  660. return preg_match("/^".preg_quote($pattern,'/')."/", $value);
  661. case '$=':
  662. return preg_match("/".preg_quote($pattern,'/')."$/", $value);
  663. case '*=':
  664. if ($pattern[0]=='/') {
  665. return preg_match($pattern, $value);
  666. }
  667. return preg_match("/".$pattern."/i", $value);
  668. }
  669. return false;
  670. }
  671.  
  672. protected function parse_selector($selector_string) {
  673. global $debugObject;
  674. if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
  675.  
  676. // pattern of CSS selectors, modified from mootools
  677. // Paperg: Add the colon to the attrbute, so that it properly finds <tag attr:ibute="something" > like google does.
  678. // Note: if you try to look at this attribute, yo MUST use getAttribute since $dom->x:y will fail the php syntax check.
  679. // Notice the \[ starting the attbute? and the @? following? This implies that an attribute can begin with an @ sign that is not captured.
  680. // This implies that an html attribute specifier may start with an @ sign that is NOT captured by the expression.
  681. // farther study is required to determine of this should be documented or removed.
  682. // $pattern = "/([\w-:\*]*)(?:\#([\w-]+)|\.([\w-]+))?(?:\[@?(!?[\w-]+)(?:([!*^$]?=)[\"']?(.*?)[\"']?)?\])?([\/, ]+)/is";
  683. $pattern = "/([\w-:\*]*)(?:\#([\w-]+)|\.([\w-]+))?(?:\[@?(!?[\w-:]+)(?:([!*^$]?=)[\"']?(.*?)[\"']?)?\])?([\/, ]+)/is";
  684. preg_match_all($pattern, trim($selector_string).' ', $matches, PREG_SET_ORDER);
  685. if (is_object($debugObject)) {$debugObject->debugLog(2, "Matches Array: ", $matches);}
  686.  
  687. $selectors = array();
  688. $result = array();
  689. //print_r($matches);
  690.  
  691. foreach ($matches as $m) {
  692. $m[0] = trim($m[0]);
  693. if ($m[0]==='' || $m[0]==='/' || $m[0]==='//') continue;
  694. // for browser generated xpath
  695. if ($m[1]==='tbody') continue;
  696.  
  697. list($tag, $key, $val, $exp, $no_key) = array($m[1], null, null, '=', false);
  698. if (!empty($m[2])) {$key='id'; $val=$m[2];}
  699. if (!empty($m[3])) {$key='class'; $val=$m[3];}
  700. if (!empty($m[4])) {$key=$m[4];}
  701. if (!empty($m[5])) {$exp=$m[5];}
  702. if (!empty($m[6])) {$val=$m[6];}
  703.  
  704. // convert to lowercase
  705. if ($this->dom->lowercase) {$tag=strtolower($tag); $key=strtolower($key);}
  706. //elements that do NOT have the specified attribute
  707. if (isset($key[0]) && $key[0]==='!') {$key=substr($key, 1); $no_key=true;}
  708.  
  709. $result[] = array($tag, $key, $val, $exp, $no_key);
  710. if (trim($m[7])===',') {
  711. $selectors[] = $result;
  712. $result = array();
  713. }
  714. }
  715. if (count($result)>0)
  716. $selectors[] = $result;
  717. return $selectors;
  718. }
  719.  
  720. function __get($name) {
  721. if (isset($this->attr[$name]))
  722. {
  723. return $this->convert_text($this->attr[$name]);
  724. }
  725. switch ($name) {
  726. case 'outertext': return $this->outertext();
  727. case 'innertext': return $this->innertext();
  728. case 'plaintext': return $this->text();
  729. case 'xmltext': return $this->xmltext();
  730. default: return array_key_exists($name, $this->attr);
  731. }
  732. }
  733.  
  734. function __set($name, $value) {
  735. switch ($name) {
  736. case 'outertext': return $this->_[HDOM_INFO_OUTER] = $value;
  737. case 'innertext':
  738. if (isset($this->_[HDOM_INFO_TEXT])) return $this->_[HDOM_INFO_TEXT] = $value;
  739. return $this->_[HDOM_INFO_INNER] = $value;
  740. }
  741. if (!isset($this->attr[$name])) {
  742. $this->_[HDOM_INFO_SPACE][] = array(' ', '', '');
  743. $this->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_DOUBLE;
  744. }
  745. $this->attr[$name] = $value;
  746. }
  747.  
  748. function __isset($name) {
  749. switch ($name) {
  750. case 'outertext': return true;
  751. case 'innertext': return true;
  752. case 'plaintext': return true;
  753. }
  754. //no value attr: nowrap, checked selected...
  755. return (array_key_exists($name, $this->attr)) ? true : isset($this->attr[$name]);
  756. }
  757.  
  758. function __unset($name) {
  759. if (isset($this->attr[$name]))
  760. unset($this->attr[$name]);
  761. }
  762.  
  763. // PaperG - Function to convert the text from one character set to another if the two sets are not the same.
  764. function convert_text($text)
  765. {
  766. global $debugObject;
  767. if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
  768.  
  769. $converted_text = $text;
  770.  
  771. $sourceCharset = "";
  772. $targetCharset = "";
  773.  
  774. if ($this->dom)
  775. {
  776. $sourceCharset = strtoupper($this->dom->_charset);
  777. $targetCharset = strtoupper($this->dom->_target_charset);
  778. }
  779. if (is_object($debugObject)) {$debugObject->debugLog(3, "source charset: " . $sourceCharset . " target charaset: " . $targetCharset);}
  780.  
  781. if (!empty($sourceCharset) && !empty($targetCharset) && (strcasecmp($sourceCharset, $targetCharset) != 0))
  782. {
  783. // Check if the reported encoding could have been incorrect and the text is actually already UTF-8
  784. if ((strcasecmp($targetCharset, 'UTF-8') == 0) && ($this->is_utf8($text)))
  785. {
  786. $converted_text = $text;
  787. }
  788. else
  789. {
  790. $converted_text = iconv($sourceCharset, $targetCharset, $text);
  791. }
  792. }
  793.  
  794. // Lets make sure that we don't have that silly BOM issue with any of the utf-8 text we output.
  795. if ($targetCharset == 'UTF-8')
  796. {
  797. if (substr($converted_text, 0, 3) == "\xef\xbb\xbf")
  798. {
  799. $converted_text = substr($converted_text, 3);
  800. }
  801. if (substr($converted_text, -3) == "\xef\xbb\xbf")
  802. {
  803. $converted_text = substr($converted_text, 0, -3);
  804. }
  805. }
  806.  
  807. return $converted_text;
  808. }
  809.  
  810. /**
  811. * Returns true if $string is valid UTF-8 and false otherwise.
  812. *
  813. * @param mixed $str String to be tested
  814. * @return boolean
  815. */
  816. static function is_utf8($str)
  817. {
  818. $c=0; $b=0;
  819. $bits=0;
  820. $len=strlen($str);
  821. for($i=0; $i<$len; $i++)
  822. {
  823. $c=ord($str[$i]);
  824. if($c > 128)
  825. {
  826. if(($c >= 254)) return false;
  827. elseif($c >= 252) $bits=6;
  828. elseif($c >= 248) $bits=5;
  829. elseif($c >= 240) $bits=4;
  830. elseif($c >= 224) $bits=3;
  831. elseif($c >= 192) $bits=2;
  832. else return false;
  833. if(($i+$bits) > $len) return false;
  834. while($bits > 1)
  835. {
  836. $i++;
  837. $b=ord($str[$i]);
  838. if($b < 128 || $b > 191) return false;
  839. $bits--;
  840. }
  841. }
  842. }
  843. return true;
  844. }
  845. /*
  846. function is_utf8($string)
  847. {
  848. //this is buggy
  849. return (utf8_encode(utf8_decode($string)) == $string);
  850. }
  851. */
  852.  
  853. /**
  854. * Function to try a few tricks to determine the displayed size of an img on the page.
  855. * NOTE: This will ONLY work on an IMG tag. Returns FALSE on all other tag types.
  856. *
  857. * @author John Schlick
  858. * @version April 19 2012
  859. * @return array an array containing the 'height' and 'width' of the image on the page or -1 if we can't figure it out.
  860. */
  861. function get_display_size()
  862. {
  863. global $debugObject;
  864.  
  865. $width = -1;
  866. $height = -1;
  867.  
  868. if ($this->tag !== 'img')
  869. {
  870. return false;
  871. }
  872.  
  873. // See if there is aheight or width attribute in the tag itself.
  874. if (isset($this->attr['width']))
  875. {
  876. $width = $this->attr['width'];
  877. }
  878.  
  879. if (isset($this->attr['height']))
  880. {
  881. $height = $this->attr['height'];
  882. }
  883.  
  884. // Now look for an inline style.
  885. if (isset($this->attr['style']))
  886. {
  887. // Thanks to user gnarf from stackoverflow for this regular expression.
  888. $attributes = array();
  889. preg_match_all("/([\w-]+)\s*:\s*([^;]+)\s*;?/", $this->attr['style'], $matches, PREG_SET_ORDER);
  890. foreach ($matches as $match) {
  891. $attributes[$match[1]] = $match[2];
  892. }
  893.  
  894. // If there is a width in the style attributes:
  895. if (isset($attributes['width']) && $width == -1)
  896. {
  897. // check that the last two characters are px (pixels)
  898. if (strtolower(substr($attributes['width'], -2)) == 'px')
  899. {
  900. $proposed_width = substr($attributes['width'], 0, -2);
  901. // Now make sure that it's an integer and not something stupid.
  902. if (filter_var($proposed_width, FILTER_VALIDATE_INT))
  903. {
  904. $width = $proposed_width;
  905. }
  906. }
  907. }
  908.  
  909. // If there is a width in the style attributes:
  910. if (isset($attributes['height']) && $height == -1)
  911. {
  912. // check that the last two characters are px (pixels)
  913. if (strtolower(substr($attributes['height'], -2)) == 'px')
  914. {
  915. $proposed_height = substr($attributes['height'], 0, -2);
  916. // Now make sure that it's an integer and not something stupid.
  917. if (filter_var($proposed_height, FILTER_VALIDATE_INT))
  918. {
  919. $height = $proposed_height;
  920. }
  921. }
  922. }
  923.  
  924. }
  925.  
  926. // Future enhancement:
  927. // Look in the tag to see if there is a class or id specified that has a height or width attribute to it.
  928.  
  929. // Far future enhancement
  930. // Look at all the parent tags of this image to see if they specify a class or id that has an img selector that specifies a height or width
  931. // Note that in this case, the class or id will have the img subselector for it to apply to the image.
  932.  
  933. // ridiculously far future development
  934. // If the class or id is specified in a SEPARATE css file thats not on the page, go get it and do what we were just doing for the ones on the page.
  935.  
  936. $result = array('height' => $height,
  937. 'width' => $width);
  938. return $result;
  939. }
  940.  
  941. // camel naming conventions
  942. function getAllAttributes() {return $this->attr;}
  943. function getAttribute($name) {return $this->__get($name);}
  944. function setAttribute($name, $value) {$this->__set($name, $value);}
  945. function hasAttribute($name) {return $this->__isset($name);}
  946. function removeAttribute($name) {$this->__set($name, null);}
  947. function getElementById($id) {return $this->find("#$id", 0);}
  948. function getElementsById($id, $idx=null) {return $this->find("#$id", $idx);}
  949. function getElementByTagName($name) {return $this->find($name, 0);}
  950. function getElementsByTagName($name, $idx=null) {return $this->find($name, $idx);}
  951. function parentNode() {return $this->parent();}
  952. function childNodes($idx=-1) {return $this->children($idx);}
  953. function firstChild() {return $this->first_child();}
  954. function lastChild() {return $this->last_child();}
  955. function nextSibling() {return $this->next_sibling();}
  956. function previousSibling() {return $this->prev_sibling();}
  957. function hasChildNodes() {return $this->has_child();}
  958. function nodeName() {return $this->tag;}
  959. function appendChild($node) {$node->parent($this); return $node;}
  960.  
  961. }
  962.  
  963. /**
  964. * simple html dom parser
  965. * Paperg - in the find routine: allow us to specify that we want case insensitive testing of the value of the selector.
  966. * Paperg - change $size from protected to public so we can easily access it
  967. * Paperg - added ForceTagsClosed in the constructor which tells us whether we trust the html or not. Default is to NOT trust it.
  968. *
  969. * @package PlaceLocalInclude
  970. */
  971. class simple_html_dom
  972. {
  973. public $root = null;
  974. public $nodes = array();
  975. public $callback = null;
  976. public $lowercase = false;
  977. // Used to keep track of how large the text was when we started.
  978. public $original_size;
  979. public $size;
  980. protected $pos;
  981. protected $doc;
  982. protected $char;
  983. protected $cursor;
  984. protected $parent;
  985. protected $noise = array();
  986. protected $token_blank = " \t\r\n";
  987. protected $token_equal = ' =/>';
  988. protected $token_slash = " />\r\n\t";
  989. protected $token_attr = ' >';
  990. // Note that this is referenced by a child node, and so it needs to be public for that node to see this information.
  991. public $_charset = '';
  992. public $_target_charset = '';
  993. protected $default_br_text = "";
  994. public $default_span_text = "";
  995.  
  996. // use isset instead of in_array, performance boost about 30%...
  997. protected $self_closing_tags = array('img'=>1, 'br'=>1, 'input'=>1, 'meta'=>1, 'link'=>1, 'hr'=>1, 'base'=>1, 'embed'=>1, 'spacer'=>1);
  998. protected $block_tags = array('root'=>1, 'body'=>1, 'form'=>1, 'div'=>1, 'span'=>1, 'table'=>1);
  999. // Known sourceforge issue #2977341
  1000. // B tags that are not closed cause us to return everything to the end of the document.
  1001. protected $optional_closing_tags = array(
  1002. 'tr'=>array('tr'=>1, 'td'=>1, 'th'=>1),
  1003. 'th'=>array('th'=>1),
  1004. 'td'=>array('td'=>1),
  1005. 'li'=>array('li'=>1),
  1006. 'dt'=>array('dt'=>1, 'dd'=>1),
  1007. 'dd'=>array('dd'=>1, 'dt'=>1),
  1008. 'dl'=>array('dd'=>1, 'dt'=>1),
  1009. 'p'=>array('p'=>1),
  1010. 'nobr'=>array('nobr'=>1),
  1011. 'b'=>array('b'=>1),
  1012. 'option'=>array('option'=>1),
  1013. );
  1014.  
  1015. function __construct($str=null, $lowercase=true, $forceTagsClosed=true, $target_charset=DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
  1016. {
  1017. if ($str)
  1018. {
  1019. if (preg_match("/^http:\/\//i",$str) || is_file($str))
  1020. {
  1021. $this->load_file($str);
  1022. }
  1023. else
  1024. {
  1025. $this->load($str, $lowercase, $stripRN, $defaultBRText, $defaultSpanText);
  1026. }
  1027. }
  1028. // Forcing tags to be closed implies that we don't trust the html, but it can lead to parsing errors if we SHOULD trust the html.
  1029. if (!$forceTagsClosed) {
  1030. $this->optional_closing_array=array();
  1031. }
  1032. $this->_target_charset = $target_charset;
  1033. }
  1034.  
  1035. function __destruct()
  1036. {
  1037. $this->clear();
  1038. }
  1039.  
  1040. // load html from string
  1041. function load($str, $lowercase=true, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
  1042. {
  1043. global $debugObject;
  1044.  
  1045. // prepare
  1046. $this->prepare($str, $lowercase, $stripRN, $defaultBRText, $defaultSpanText);
  1047. // strip out comments
  1048. $this->remove_noise("'<!--(.*?)-->'is");
  1049. // strip out cdata
  1050. $this->remove_noise("'<!\[CDATA\[(.*?)\]\]>'is", true);
  1051. // Per sourceforge http://sourceforge.net/tracker/?func=detail&aid=2949097&group_id=218559&atid=1044037
  1052. // Script tags removal now preceeds style tag removal.
  1053. // strip out <script> tags
  1054. $this->remove_noise("'<\s*script[^>]*[^/]>(.*?)<\s*/\s*script\s*>'is");
  1055. $this->remove_noise("'<\s*script\s*>(.*?)<\s*/\s*script\s*>'is");
  1056. // strip out <style> tags
  1057. $this->remove_noise("'<\s*style[^>]*[^/]>(.*?)<\s*/\s*style\s*>'is");
  1058. $this->remove_noise("'<\s*style\s*>(.*?)<\s*/\s*style\s*>'is");
  1059. // strip out preformatted tags
  1060. $this->remove_noise("'<\s*(?:code)[^>]*>(.*?)<\s*/\s*(?:code)\s*>'is");
  1061. // strip out server side scripts
  1062. $this->remove_noise("'(<\?)(.*?)(\?>)'s", true);
  1063. // strip smarty scripts
  1064. $this->remove_noise("'(\{\w)(.*?)(\})'s", true);
  1065.  
  1066. // parsing
  1067. while ($this->parse());
  1068. // end
  1069. $this->root->_[HDOM_INFO_END] = $this->cursor;
  1070. $this->parse_charset();
  1071.  
  1072. // make load function chainable
  1073. return $this;
  1074.  
  1075. }
  1076.  
  1077. // load html from file
  1078. function load_file()
  1079. {
  1080. $args = func_get_args();
  1081. $this->load(call_user_func_array('file_get_contents', $args), true);
  1082. // Throw an error if we can't properly load the dom.
  1083. if (($error=error_get_last())!==null) {
  1084. $this->clear();
  1085. return false;
  1086. }
  1087. }
  1088.  
  1089. // set callback function
  1090. function set_callback($function_name)
  1091. {
  1092. $this->callback = $function_name;
  1093. }
  1094.  
  1095. // remove callback function
  1096. function remove_callback()
  1097. {
  1098. $this->callback = null;
  1099. }
  1100.  
  1101. // save dom as string
  1102. function save($filepath='')
  1103. {
  1104. $ret = $this->root->innertext();
  1105. if ($filepath!=='') file_put_contents($filepath, $ret, LOCK_EX);
  1106. return $ret;
  1107. }
  1108.  
  1109. // find dom node by css selector
  1110. // Paperg - allow us to specify that we want case insensitive testing of the value of the selector.
  1111. function find($selector, $idx=null, $lowercase=false)
  1112. {
  1113. return $this->root->find($selector, $idx, $lowercase);
  1114. }
  1115.  
  1116. // clean up memory due to php5 circular references memory leak...
  1117. function clear()
  1118. {
  1119. foreach ($this->nodes as $n) {$n->clear(); $n = null;}
  1120. // This add next line is documented in the sourceforge repository. 2977248 as a fix for ongoing memory leaks that occur even with the use of clear.
  1121. if (isset($this->children)) foreach ($this->children as $n) {$n->clear(); $n = null;}
  1122. if (isset($this->parent)) {$this->parent->clear(); unset($this->parent);}
  1123. if (isset($this->root)) {$this->root->clear(); unset($this->root);}
  1124. unset($this->doc);
  1125. unset($this->noise);
  1126. }
  1127.  
  1128. function dump($show_attr=true)
  1129. {
  1130. $this->root->dump($show_attr);
  1131. }
  1132.  
  1133. // prepare HTML data and init everything
  1134. protected function prepare($str, $lowercase=true, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
  1135. {
  1136. $this->clear();
  1137.  
  1138. // set the length of content before we do anything to it.
  1139. $this->size = strlen($str);
  1140. // Save the original size of the html that we got in. It might be useful to someone.
  1141. $this->original_size = $this->size;
  1142.  
  1143. //before we save the string as the doc... strip out the \r \n's if we are told to.
  1144. if ($stripRN) {
  1145. $str = str_replace("\r", " ", $str);
  1146. $str = str_replace("\n", " ", $str);
  1147.  
  1148. // set the length of content since we have changed it.
  1149. $this->size = strlen($str);
  1150. }
  1151.  
  1152. $this->doc = $str;
  1153. $this->pos = 0;
  1154. $this->cursor = 1;
  1155. $this->noise = array();
  1156. $this->nodes = array();
  1157. $this->lowercase = $lowercase;
  1158. $this->default_br_text = $defaultBRText;
  1159. $this->default_span_text = $defaultSpanText;
  1160. $this->root = new simple_html_dom_node($this);
  1161. $this->root->tag = 'root';
  1162. $this->root->_[HDOM_INFO_BEGIN] = -1;
  1163. $this->root->nodetype = HDOM_TYPE_ROOT;
  1164. $this->parent = $this->root;
  1165. if ($this->size>0) $this->char = $this->doc[0];
  1166. }
  1167.  
  1168. // parse html content
  1169. protected function parse()
  1170. {
  1171. if (($s = $this->copy_until_char('<'))==='')
  1172. {
  1173. return $this->read_tag();
  1174. }
  1175.  
  1176. // text
  1177. $node = new simple_html_dom_node($this);
  1178. ++$this->cursor;
  1179. $node->_[HDOM_INFO_TEXT] = $s;
  1180. $this->link_nodes($node, false);
  1181. return true;
  1182. }
  1183.  
  1184. // 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.
  1185. // NOTE: IF you provide a routine called get_last_retrieve_url_contents_content_type which returns the CURLINFO_CONTENT_TYPE from the last curl_exec
  1186. // (or the content_type header from the last transfer), we will parse THAT, and if a charset is specified, we will use it over any other mechanism.
  1187. protected function parse_charset()
  1188. {
  1189. global $debugObject;
  1190.  
  1191. $charset = null;
  1192.  
  1193. if (function_exists('get_last_retrieve_url_contents_content_type'))
  1194. {
  1195. $contentTypeHeader = get_last_retrieve_url_contents_content_type();
  1196. $success = preg_match('/charset=(.+)/', $contentTypeHeader, $matches);
  1197. if ($success)
  1198. {
  1199. $charset = $matches[1];
  1200. if (is_object($debugObject)) {$debugObject->debugLog(2, 'header content-type found charset of: ' . $charset);}
  1201. }
  1202.  
  1203. }
  1204.  
  1205. if (empty($charset))
  1206. {
  1207. $el = $this->root->find('meta[http-equiv=Content-Type]',0);
  1208. if (!empty($el))
  1209. {
  1210. $fullvalue = $el->content;
  1211. if (is_object($debugObject)) {$debugObject->debugLog(2, 'meta content-type tag found' . $fullvalue);}
  1212.  
  1213. if (!empty($fullvalue))
  1214. {
  1215. $success = preg_match('/charset=(.+)/', $fullvalue, $matches);
  1216. if ($success)
  1217. {
  1218. $charset = $matches[1];
  1219. }
  1220. else
  1221. {
  1222. // If there is a meta tag, and they don't specify the character set, research says that it's typically ISO-8859-1
  1223. if (is_object($debugObject)) {$debugObject->debugLog(2, 'meta content-type tag couldn\'t be parsed. using iso-8859 default.');}
  1224. $charset = 'ISO-8859-1';
  1225. }
  1226. }
  1227. }
  1228. }
  1229.  
  1230. // If we couldn't find a charset above, then lets try to detect one based on the text we got...
  1231. if (empty($charset))
  1232. {
  1233. // Have php try to detect the encoding from the text given to us.
  1234. $charset = mb_detect_encoding($this->root->plaintext . "ascii", $encoding_list = array( "UTF-8", "CP1252" ) );
  1235. if (is_object($debugObject)) {$debugObject->debugLog(2, 'mb_detect found: ' . $charset);}
  1236.  
  1237. // and if this doesn't work... then we need to just wrongheadedly assume it's UTF-8 so that we can move on - cause this will usually give us most of what we need...
  1238. if ($charset === false)
  1239. {
  1240. if (is_object($debugObject)) {$debugObject->debugLog(2, 'since mb_detect failed - using default of utf-8');}
  1241. $charset = 'UTF-8';
  1242. }
  1243. }
  1244.  
  1245. // Since CP1252 is a superset, if we get one of it's subsets, we want it instead.
  1246. if ((strtolower($charset) == strtolower('ISO-8859-1')) || (strtolower($charset) == strtolower('Latin1')) || (strtolower($charset) == strtolower('Latin-1')))
  1247. {
  1248. if (is_object($debugObject)) {$debugObject->debugLog(2, 'replacing ' . $charset . ' with CP1252 as its a superset');}
  1249. $charset = 'CP1252';
  1250. }
  1251.  
  1252. if (is_object($debugObject)) {$debugObject->debugLog(1, 'EXIT - ' . $charset);}
  1253.  
  1254. return $this->_charset = $charset;
  1255. }
  1256.  
  1257. // read tag info
  1258. protected function read_tag()
  1259. {
  1260. if ($this->char!=='<')
  1261. {
  1262. $this->root->_[HDOM_INFO_END] = $this->cursor;
  1263. return false;
  1264. }
  1265. $begin_tag_pos = $this->pos;
  1266. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1267.  
  1268. // end tag
  1269. if ($this->char==='/')
  1270. {
  1271. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1272. // This represents the change in the simple_html_dom trunk from revision 180 to 181.
  1273. // $this->skip($this->token_blank_t);
  1274. $this->skip($this->token_blank);
  1275. $tag = $this->copy_until_char('>');
  1276.  
  1277. // skip attributes in end tag
  1278. if (($pos = strpos($tag, ' '))!==false)
  1279. $tag = substr($tag, 0, $pos);
  1280.  
  1281. $parent_lower = strtolower($this->parent->tag);
  1282. $tag_lower = strtolower($tag);
  1283.  
  1284. if ($parent_lower!==$tag_lower)
  1285. {
  1286. if (isset($this->optional_closing_tags[$parent_lower]) && isset($this->block_tags[$tag_lower]))
  1287. {
  1288. $this->parent->_[HDOM_INFO_END] = 0;
  1289. $org_parent = $this->parent;
  1290.  
  1291. while (($this->parent->parent) && strtolower($this->parent->tag)!==$tag_lower)
  1292. $this->parent = $this->parent->parent;
  1293.  
  1294. if (strtolower($this->parent->tag)!==$tag_lower) {
  1295. $this->parent = $org_parent; // restore origonal parent
  1296. if ($this->parent->parent) $this->parent = $this->parent->parent;
  1297. $this->parent->_[HDOM_INFO_END] = $this->cursor;
  1298. return $this->as_text_node($tag);
  1299. }
  1300. }
  1301. else if (($this->parent->parent) && isset($this->block_tags[$tag_lower]))
  1302. {
  1303. $this->parent->_[HDOM_INFO_END] = 0;
  1304. $org_parent = $this->parent;
  1305.  
  1306. while (($this->parent->parent) && strtolower($this->parent->tag)!==$tag_lower)
  1307. $this->parent = $this->parent->parent;
  1308.  
  1309. if (strtolower($this->parent->tag)!==$tag_lower)
  1310. {
  1311. $this->parent = $org_parent; // restore origonal parent
  1312. $this->parent->_[HDOM_INFO_END] = $this->cursor;
  1313. return $this->as_text_node($tag);
  1314. }
  1315. }
  1316. else if (($this->parent->parent) && strtolower($this->parent->parent->tag)===$tag_lower)
  1317. {
  1318. $this->parent->_[HDOM_INFO_END] = 0;
  1319. $this->parent = $this->parent->parent;
  1320. }
  1321. else
  1322. return $this->as_text_node($tag);
  1323. }
  1324.  
  1325. $this->parent->_[HDOM_INFO_END] = $this->cursor;
  1326. if ($this->parent->parent) $this->parent = $this->parent->parent;
  1327.  
  1328. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1329. return true;
  1330. }
  1331.  
  1332. $node = new simple_html_dom_node($this);
  1333. $node->_[HDOM_INFO_BEGIN] = $this->cursor;
  1334. ++$this->cursor;
  1335. $tag = $this->copy_until($this->token_slash);
  1336. $node->tag_start = $begin_tag_pos;
  1337.  
  1338. // doctype, cdata & comments...
  1339. if (isset($tag[0]) && $tag[0]==='!') {
  1340. $node->_[HDOM_INFO_TEXT] = '<' . $tag . $this->copy_until_char('>');
  1341.  
  1342. if (isset($tag[2]) && $tag[1]==='-' && $tag[2]==='-') {
  1343. $node->nodetype = HDOM_TYPE_COMMENT;
  1344. $node->tag = 'comment';
  1345. } else {
  1346. $node->nodetype = HDOM_TYPE_UNKNOWN;
  1347. $node->tag = 'unknown';
  1348. }
  1349. if ($this->char==='>') $node->_[HDOM_INFO_TEXT].='>';
  1350. $this->link_nodes($node, true);
  1351. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1352. return true;
  1353. }
  1354.  
  1355. // text
  1356. if ($pos=strpos($tag, '<')!==false) {
  1357. $tag = '<' . substr($tag, 0, -1);
  1358. $node->_[HDOM_INFO_TEXT] = $tag;
  1359. $this->link_nodes($node, false);
  1360. $this->char = $this->doc[--$this->pos]; // prev
  1361. return true;
  1362. }
  1363.  
  1364. if (!preg_match("/^[\w-:]+$/", $tag)) {
  1365. $node->_[HDOM_INFO_TEXT] = '<' . $tag . $this->copy_until('<>');
  1366. if ($this->char==='<') {
  1367. $this->link_nodes($node, false);
  1368. return true;
  1369. }
  1370.  
  1371. if ($this->char==='>') $node->_[HDOM_INFO_TEXT].='>';
  1372. $this->link_nodes($node, false);
  1373. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1374. return true;
  1375. }
  1376.  
  1377. // begin tag
  1378. $node->nodetype = HDOM_TYPE_ELEMENT;
  1379. $tag_lower = strtolower($tag);
  1380. $node->tag = ($this->lowercase) ? $tag_lower : $tag;
  1381.  
  1382. // handle optional closing tags
  1383. if (isset($this->optional_closing_tags[$tag_lower]) )
  1384. {
  1385. while (isset($this->optional_closing_tags[$tag_lower][strtolower($this->parent->tag)]))
  1386. {
  1387. $this->parent->_[HDOM_INFO_END] = 0;
  1388. $this->parent = $this->parent->parent;
  1389. }
  1390. $node->parent = $this->parent;
  1391. }
  1392.  
  1393. $guard = 0; // prevent infinity loop
  1394. $space = array($this->copy_skip($this->token_blank), '', '');
  1395.  
  1396. // attributes
  1397. do
  1398. {
  1399. if ($this->char!==null && $space[0]==='')
  1400. {
  1401. break;
  1402. }
  1403. $name = $this->copy_until($this->token_equal);
  1404. if ($guard===$this->pos)
  1405. {
  1406. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1407. continue;
  1408. }
  1409. $guard = $this->pos;
  1410.  
  1411. // handle endless '<'
  1412. if ($this->pos>=$this->size-1 && $this->char!=='>') {
  1413. $node->nodetype = HDOM_TYPE_TEXT;
  1414. $node->_[HDOM_INFO_END] = 0;
  1415. $node->_[HDOM_INFO_TEXT] = '<'.$tag . $space[0] . $name;
  1416. $node->tag = 'text';
  1417. $this->link_nodes($node, false);
  1418. return true;
  1419. }
  1420.  
  1421. // handle mismatch '<'
  1422. if ($this->doc[$this->pos-1]=='<') {
  1423. $node->nodetype = HDOM_TYPE_TEXT;
  1424. $node->tag = 'text';
  1425. $node->attr = array();
  1426. $node->_[HDOM_INFO_END] = 0;
  1427. $node->_[HDOM_INFO_TEXT] = substr($this->doc, $begin_tag_pos, $this->pos-$begin_tag_pos-1);
  1428. $this->pos -= 2;
  1429. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1430. $this->link_nodes($node, false);
  1431. return true;
  1432. }
  1433.  
  1434. if ($name!=='/' && $name!=='') {
  1435. $space[1] = $this->copy_skip($this->token_blank);
  1436. $name = $this->restore_noise($name);
  1437. if ($this->lowercase) $name = strtolower($name);
  1438. if ($this->char==='=') {
  1439. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1440. $this->parse_attr($node, $name, $space);
  1441. }
  1442. else {
  1443. //no value attr: nowrap, checked selected...
  1444. $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_NO;
  1445. $node->attr[$name] = true;
  1446. if ($this->char!='>') $this->char = $this->doc[--$this->pos]; // prev
  1447. }
  1448. $node->_[HDOM_INFO_SPACE][] = $space;
  1449. $space = array($this->copy_skip($this->token_blank), '', '');
  1450. }
  1451. else
  1452. break;
  1453. } while ($this->char!=='>' && $this->char!=='/');
  1454.  
  1455. $this->link_nodes($node, true);
  1456. $node->_[HDOM_INFO_ENDSPACE] = $space[0];
  1457.  
  1458. // check self closing
  1459. if ($this->copy_until_char_escape('>')==='/')
  1460. {
  1461. $node->_[HDOM_INFO_ENDSPACE] .= '/';
  1462. $node->_[HDOM_INFO_END] = 0;
  1463. }
  1464. else
  1465. {
  1466. // reset parent
  1467. if (!isset($this->self_closing_tags[strtolower($node->tag)])) $this->parent = $node;
  1468. }
  1469. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1470.  
  1471. // If it's a BR tag, we need to set it's text to the default text.
  1472. // This way when we see it in plaintext, we can generate formatting that the user wants.
  1473. // since a br tag never has sub nodes, this works well.
  1474. if ($node->tag == "br")
  1475. {
  1476. $node->_[HDOM_INFO_INNER] = $this->default_br_text;
  1477. }
  1478.  
  1479. return true;
  1480. }
  1481.  
  1482. // parse attributes
  1483. protected function parse_attr($node, $name, &$space)
  1484. {
  1485. // Per sourceforge: http://sourceforge.net/tracker/?func=detail&aid=3061408&group_id=218559&atid=1044037
  1486. // If the attribute is already defined inside a tag, only pay atetntion to the first one as opposed to the last one.
  1487. if (isset($node->attr[$name]))
  1488. {
  1489. return;
  1490. }
  1491.  
  1492. $space[2] = $this->copy_skip($this->token_blank);
  1493. switch ($this->char) {
  1494. case '"':
  1495. $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_DOUBLE;
  1496. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1497. $node->attr[$name] = $this->restore_noise($this->copy_until_char_escape('"'));
  1498. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1499. break;
  1500. case '\'':
  1501. $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_SINGLE;
  1502. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1503. $node->attr[$name] = $this->restore_noise($this->copy_until_char_escape('\''));
  1504. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1505. break;
  1506. default:
  1507. $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_NO;
  1508. $node->attr[$name] = $this->restore_noise($this->copy_until($this->token_attr));
  1509. }
  1510. // PaperG: Attributes should not have \r or \n in them, that counts as html whitespace.
  1511. $node->attr[$name] = str_replace("\r", "", $node->attr[$name]);
  1512. $node->attr[$name] = str_replace("\n", "", $node->attr[$name]);
  1513. // 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.
  1514. if ($name == "class") {
  1515. $node->attr[$name] = trim($node->attr[$name]);
  1516. }
  1517. }
  1518.  
  1519. // link node's parent
  1520. protected function link_nodes(&$node, $is_child)
  1521. {
  1522. $node->parent = $this->parent;
  1523. $this->parent->nodes[] = $node;
  1524. if ($is_child)
  1525. {
  1526. $this->parent->children[] = $node;
  1527. }
  1528. }
  1529.  
  1530. // as a text node
  1531. protected function as_text_node($tag)
  1532. {
  1533. $node = new simple_html_dom_node($this);
  1534. ++$this->cursor;
  1535. $node->_[HDOM_INFO_TEXT] = '</' . $tag . '>';
  1536. $this->link_nodes($node, false);
  1537. $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1538. return true;
  1539. }
  1540.  
  1541. protected function skip($chars)
  1542. {
  1543. $this->pos += strspn($this->doc, $chars, $this->pos);
  1544. $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1545. }
  1546.  
  1547. protected function copy_skip($chars)
  1548. {
  1549. $pos = $this->pos;
  1550. $len = strspn($this->doc, $chars, $pos);
  1551. $this->pos += $len;
  1552. $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1553. if ($len===0) return '';
  1554. return substr($this->doc, $pos, $len);
  1555. }
  1556.  
  1557. protected function copy_until($chars)
  1558. {
  1559. $pos = $this->pos;
  1560. $len = strcspn($this->doc, $chars, $pos);
  1561. $this->pos += $len;
  1562. $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
  1563. return substr($this->doc, $pos, $len);
  1564. }
  1565.  
  1566. protected function copy_until_char($char)
  1567. {
  1568. if ($this->char===null) return '';
  1569.  
  1570. if (($pos = strpos($this->doc, $char, $this->pos))===false) {
  1571. $ret = substr($this->doc, $this->pos, $this->size-$this->pos);
  1572. $this->char = null;
  1573. $this->pos = $this->size;
  1574. return $ret;
  1575. }
  1576.  
  1577. if ($pos===$this->pos) return '';
  1578. $pos_old = $this->pos;
  1579. $this->char = $this->doc[$pos];
  1580. $this->pos = $pos;
  1581. return substr($this->doc, $pos_old, $pos-$pos_old);
  1582. }
  1583.  
  1584. protected function copy_until_char_escape($char)
  1585. {
  1586. if ($this->char===null) return '';
  1587.  
  1588. $start = $this->pos;
  1589. while (1)
  1590. {
  1591. if (($pos = strpos($this->doc, $char, $start))===false)
  1592. {
  1593. $ret = substr($this->doc, $this->pos, $this->size-$this->pos);
  1594. $this->char = null;
  1595. $this->pos = $this->size;
  1596. return $ret;
  1597. }
  1598.  
  1599. if ($pos===$this->pos) return '';
  1600.  
  1601. if ($this->doc[$pos-1]==='\\') {
  1602. $start = $pos+1;
  1603. continue;
  1604. }
  1605.  
  1606. $pos_old = $this->pos;
  1607. $this->char = $this->doc[$pos];
  1608. $this->pos = $pos;
  1609. return substr($this->doc, $pos_old, $pos-$pos_old);
  1610. }
  1611. }
  1612.  
  1613. // remove noise from html content
  1614. // save the noise in the $this->noise array.
  1615. protected function remove_noise($pattern, $remove_tag=false)
  1616. {
  1617. global $debugObject;
  1618. if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
  1619.  
  1620. $count = preg_match_all($pattern, $this->doc, $matches, PREG_SET_ORDER|PREG_OFFSET_CAPTURE);
  1621.  
  1622. for ($i=$count-1; $i>-1; --$i)
  1623. {
  1624. $key = '___noise___'.sprintf('% 5d', count($this->noise)+1000);
  1625. if (is_object($debugObject)) { $debugObject->debugLog(2, 'key is: ' . $key); }
  1626. $idx = ($remove_tag) ? 0 : 1;
  1627. $this->noise[$key] = $matches[$i][$idx][0];
  1628. $this->doc = substr_replace($this->doc, $key, $matches[$i][$idx][1], strlen($matches[$i][$idx][0]));
  1629. }
  1630.  
  1631. // reset the length of content
  1632. $this->size = strlen($this->doc);
  1633. if ($this->size>0)
  1634. {
  1635. $this->char = $this->doc[0];
  1636. }
  1637. }
  1638.  
  1639. // restore noise to html content
  1640. function restore_noise($text)
  1641. {
  1642. global $debugObject;
  1643. if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
  1644.  
  1645. while (($pos=strpos($text, '___noise___'))!==false)
  1646. {
  1647. // Sometimes there is a broken piece of markup, and we don't GET the pos+11 etc... token which indicates a problem outside of us...
  1648. if (strlen($text) > $pos+15)
  1649. {
  1650. $key = '___noise___'.$text[$pos+11].$text[$pos+12].$text[$pos+13].$text[$pos+14].$text[$pos+15];
  1651. if (is_object($debugObject)) { $debugObject->debugLog(2, 'located key of: ' . $key); }
  1652.  
  1653. if (isset($this->noise[$key]))
  1654. {
  1655. $text = substr($text, 0, $pos).$this->noise[$key].substr($text, $pos+16);
  1656. }
  1657. else
  1658. {
  1659. // do this to prevent an infinite loop.
  1660. $text = substr($text, 0, $pos).'UNDEFINED NOISE FOR KEY: '.$key . substr($text, $pos+16);
  1661. }
  1662. }
  1663. else
  1664. {
  1665. // There is no valid key being given back to us... We must get rid of the ___noise___ or we will have a problem.
  1666. $text = substr($text, 0, $pos).'NO NUMERIC NOISE KEY' . substr($text, $pos+11);
  1667. }
  1668. }
  1669. return $text;
  1670. }
  1671.  
  1672. // Sometimes we NEED one of the noise elements.
  1673. function search_noise($text)
  1674. {
  1675. global $debugObject;
  1676. if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
  1677.  
  1678. foreach($this->noise as $noiseElement)
  1679. {
  1680. if (strpos($noiseElement, $text)!==false)
  1681. {
  1682. return $noiseElement;
  1683. }
  1684. }
  1685. }
  1686. function __toString()
  1687. {
  1688. return $this->root->innertext();
  1689. }
  1690.  
  1691. function __get($name)
  1692. {
  1693. switch ($name)
  1694. {
  1695. case 'outertext':
  1696. return $this->root->innertext();
  1697. case 'innertext':
  1698. return $this->root->innertext();
  1699. case 'plaintext':
  1700. return $this->root->text();
  1701. case 'charset':
  1702. return $this->_charset;
  1703. case 'target_charset':
  1704. return $this->_target_charset;
  1705. }
  1706. }
  1707.  
  1708. // camel naming conventions
  1709. function childNodes($idx=-1) {return $this->root->childNodes($idx);}
  1710. function firstChild() {return $this->root->first_child();}
  1711. function lastChild() {return $this->root->last_child();}
  1712. function createElement($name, $value=null) {return @str_get_html("<$name>$value</$name>")->first_child();}
  1713. function createTextNode($value) {return @end(str_get_html($value)->nodes);}
  1714. function getElementById($id) {return $this->find("#$id", 0);}
  1715. function getElementsById($id, $idx=null) {return $this->find("#$id", $idx);}
  1716. function getElementByTagName($name) {return $this->find($name, 0);}
  1717. function getElementsByTagName($name, $idx=-1) {return $this->find($name, $idx);}
  1718. function loadFile() {$args = func_get_args();$this->load_file($args);}
  1719. }
  1720.  
  1721. ?>
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!