├── README.md
├── Src
    └── Sunra
    │   └── PhpSimple
    │       ├── HtmlDomParser.php
    │       └── simplehtmldom_1_5
    │           └── simple_html_dom.php
└── composer.json


/README.md:
--------------------------------------------------------------------------------
 1 | php-simple-html-dom-parser
 2 | ==========================
 3 | 
 4 | Version 1.5.2
 5 | 
 6 | Adaptation for Composer and PSR-0 of:
 7 | 
 8 | A HTML DOM parser written in PHP5+ let you manipulate HTML in a very easy way!
 9 | Require PHP 5+.
10 | Supports invalid HTML.
11 | Find tags on an HTML page with selectors just like jQuery.
12 | Extract contents from HTML in a single line.
13 | 
14 | http://simplehtmldom.sourceforge.net/
15 | 
16 | 
17 | Install
18 | -------
19 | 
20 |  composer.phar
21 | ```json
22 | "require": {
23 |     "sunra/php-simple-html-dom-parser": "1.5.2"
24 |     }
25 | ```
26 | 
27 | Usage
28 | -----
29 | 
30 | ```php
31 | use Sunra\PhpSimple\HtmlDomParser;
32 | 
33 | ...
34 | $dom = HtmlDomParser::str_get_html( $str );
35 | or 
36 | $dom = HtmlDomParser::file_get_html( $file_name );
37 | 
38 | $elems = $dom->find($elem_name);
39 | ...
40 | 
41 | ```
42 | 


--------------------------------------------------------------------------------
/Src/Sunra/PhpSimple/HtmlDomParser.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | namespace Sunra\PhpSimple;
 4 | 
 5 | require 'simplehtmldom_1_5'.DIRECTORY_SEPARATOR.'simple_html_dom.php';
 6 | 
 7 | class HtmlDomParser {
 8 | 	
 9 | 	/**
10 | 	 * @return \simplehtmldom_1_5\simple_html_dom
11 | 	 */
12 | 	static public function file_get_html() {
13 | 		return call_user_func_array ( '\simplehtmldom_1_5\file_get_html' , func_get_args() );
14 | 	}
15 | 
16 | 	/**
17 | 	 * get html dom from string
18 | 	 * @return \simplehtmldom_1_5\simple_html_dom
19 | 	 */
20 | 	static public function str_get_html() {
21 | 		return call_user_func_array ( '\simplehtmldom_1_5\str_get_html' , func_get_args() );
22 | 	}
23 | }


--------------------------------------------------------------------------------
/Src/Sunra/PhpSimple/simplehtmldom_1_5/simple_html_dom.php:
--------------------------------------------------------------------------------
   1 | <?php
   2 | 
   3 | namespace simplehtmldom_1_5;
   4 | 
   5 | /**
   6 |  * Website: http://sourceforge.net/projects/simplehtmldom/
   7 |  * Acknowledge: Jose Solorzano (https://sourceforge.net/projects/php-html/)
   8 |  * Contributions by:
   9 |  *     Yousuke Kumakura (Attribute filters)
  10 |  *     Vadim Voituk (Negative indexes supports of "find" method)
  11 |  *     Antcs (Constructor with automatically load contents either text or file/url)
  12 |  *
  13 |  * all affected sections have comments starting with "PaperG"
  14 |  *
  15 |  * Paperg - Added case insensitive testing of the value of the selector.
  16 |  * Paperg - Added tag_start for the starting index of tags - NOTE: This works but not accurately.
  17 |  *  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,
  18 |  *  it will almost always be smaller by some amount.
  19 |  *  We use this to determine how far into the file the tag in question is.  This "percentage will never be accurate as the $dom->size is the "real" number of bytes the dom was created from.
  20 |  *  but for most purposes, it's a really good estimation.
  21 |  * Paperg - Added the forceTagsClosed to the dom constructor.  Forcing tags closed is great for malformed html, but it CAN lead to parsing errors.
  22 |  * Allow the user to tell us how much they trust the html.
  23 |  * Paperg add the text and plaintext to the selectors for the find syntax.  plaintext implies text in the innertext of a node.  text implies that the tag is a text node.
  24 |  * This allows for us to find tags based on the text they contain.
  25 |  * Create find_ancestor_tag to see if a tag is - at any level - inside of another specific tag.
  26 |  * Paperg: added parse_charset so that we know about the character set of the source document.
  27 |  *  NOTE:  If the user's system has a routine called get_last_retrieve_url_contents_content_type availalbe, we will assume it's returning the content-type header from the
  28 |  *  last transfer or curl_exec, and we will parse that and use it in preference to any other method of charset detection.
  29 |  *
  30 |  * Found infinite loop in the case of broken html in restore_noise.  Rewrote to protect from that.
  31 |  * PaperG (John Schlick) Added get_display_size for "IMG" tags.
  32 |  *
  33 |  * Licensed under The MIT License
  34 |  * Redistributions of files must retain the above copyright notice.
  35 |  *
  36 |  * @author S.C. Chen <me578022@gmail.com>
  37 |  * @author John Schlick
  38 |  * @author Rus Carroll
  39 |  * @version 1.5 ($Rev: 196 $)
  40 |  * @package PlaceLocalInclude
  41 |  * @subpackage simple_html_dom
  42 |  */
  43 | 
  44 | /**
  45 |  * All of the Defines for the classes below.
  46 |  * @author S.C. Chen <me578022@gmail.com>
  47 |  */
  48 | define('HDOM_TYPE_ELEMENT', 1);
  49 | define('HDOM_TYPE_COMMENT', 2);
  50 | define('HDOM_TYPE_TEXT',    3);
  51 | define('HDOM_TYPE_ENDTAG',  4);
  52 | define('HDOM_TYPE_ROOT',    5);
  53 | define('HDOM_TYPE_UNKNOWN', 6);
  54 | define('HDOM_QUOTE_DOUBLE', 0);
  55 | define('HDOM_QUOTE_SINGLE', 1);
  56 | define('HDOM_QUOTE_NO',     3);
  57 | define('HDOM_INFO_BEGIN',   0);
  58 | define('HDOM_INFO_END',     1);
  59 | define('HDOM_INFO_QUOTE',   2);
  60 | define('HDOM_INFO_SPACE',   3);
  61 | define('HDOM_INFO_TEXT',    4);
  62 | define('HDOM_INFO_INNER',   5);
  63 | define('HDOM_INFO_OUTER',   6);
  64 | define('HDOM_INFO_ENDSPACE',7);
  65 | define('DEFAULT_TARGET_CHARSET', 'UTF-8');
  66 | define('DEFAULT_BR_TEXT', "\r\n");
  67 | define('DEFAULT_SPAN_TEXT', " ");
  68 | if (!defined('MAX_FILE_SIZE'))
  69 | {
  70 |     define('MAX_FILE_SIZE', 600000);
  71 | }
  72 | // helper functions
  73 | // -----------------------------------------------------------------------------
  74 | // get html dom from file
  75 | // $maxlen is defined in the code as PHP_STREAM_COPY_ALL which is defined as -1.
  76 | function file_get_html($url, $use_include_path = false, $context=null, $offset=0, $maxLen=-1, $lowercase = true, $forceTagsClosed=true, $target_charset = DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
  77 | {
  78 |     // We DO force the tags to be terminated.
  79 |     $dom = new simple_html_dom(null, $lowercase, $forceTagsClosed, $target_charset, $stripRN, $defaultBRText, $defaultSpanText);
  80 |     // For sourceforge users: uncomment the next line and comment the retreive_url_contents line 2 lines down if it is not already done.
  81 |     $contents = file_get_contents($url, $use_include_path, $context, $offset);
  82 |     // Paperg - use our own mechanism for getting the contents as we want to control the timeout.
  83 |     //$contents = retrieve_url_contents($url);
  84 |     if (empty($contents) || strlen($contents) > MAX_FILE_SIZE)
  85 |     {
  86 |         return false;
  87 |     }
  88 |     // The second parameter can force the selectors to all be lowercase.
  89 |     $dom->load($contents, $lowercase, $stripRN);
  90 |     return $dom;
  91 | }
  92 | 
  93 | // get html dom from string
  94 | function str_get_html($str, $lowercase=true, $forceTagsClosed=true, $target_charset = DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
  95 | {
  96 |     $dom = new simple_html_dom(null, $lowercase, $forceTagsClosed, $target_charset, $stripRN, $defaultBRText, $defaultSpanText);
  97 |     if (empty($str) || strlen($str) > MAX_FILE_SIZE)
  98 |     {
  99 |         $dom->clear();
 100 |         return false;
 101 |     }
 102 |     $dom->load($str, $lowercase, $stripRN);
 103 |     return $dom;
 104 | }
 105 | 
 106 | // dump html dom tree
 107 | function dump_html_tree($node, $show_attr=true, $deep=0)
 108 | {
 109 |     $node->dump($node);
 110 | }
 111 | 
 112 | 
 113 | /**
 114 |  * simple html dom node
 115 |  * PaperG - added ability for "find" routine to lowercase the value of the selector.
 116 |  * PaperG - added $tag_start to track the start position of the tag in the total byte index
 117 |  *
 118 |  * @package PlaceLocalInclude
 119 |  */
 120 | class simple_html_dom_node
 121 | {
 122 |     public $nodetype = HDOM_TYPE_TEXT;
 123 |     public $tag = 'text';
 124 |     public $attr = array();
 125 |     /** @var simple_html_dom_node[] $children */
 126 |     public $children = array();
 127 |     public $nodes = array();
 128 |     public $parent = null;
 129 |     // The "info" array - see HDOM_INFO_... for what each element contains.
 130 |     public $_ = array();
 131 |     public $tag_start = 0;
 132 |     private $dom = null;
 133 | 
 134 |     function __construct(simple_html_dom $dom)
 135 |     {
 136 |         $this->dom = $dom;
 137 |         $dom->nodes[] = $this;
 138 |     }
 139 | 
 140 |     function __destruct()
 141 |     {
 142 |         $this->clear();
 143 |     }
 144 | 
 145 |     function __toString()
 146 |     {
 147 |         return $this->outertext();
 148 |     }
 149 | 
 150 |     // clean up memory due to php5 circular references memory leak...
 151 |     function clear()
 152 |     {
 153 |         $this->dom = null;
 154 |         $this->nodes = null;
 155 |         $this->parent = null;
 156 |         $this->children = null;
 157 |     }
 158 | 
 159 |     // dump node's tree
 160 |     function dump($show_attr=true, $deep=0)
 161 |     {
 162 |         $lead = str_repeat('    ', $deep);
 163 | 
 164 |         echo $lead.$this->tag;
 165 |         if ($show_attr && count($this->attr)>0)
 166 |         {
 167 |             echo '(';
 168 |             foreach ($this->attr as $k=>$v)
 169 |                 echo "[$k]=>\"".$this->$k.'", ';
 170 |             echo ')';
 171 |         }
 172 |         echo "\n";
 173 | 
 174 |         if ($this->nodes)
 175 |         {
 176 |             foreach ($this->nodes as $c)
 177 |             {
 178 |                 $c->dump($show_attr, $deep+1);
 179 |             }
 180 |         }
 181 |     }
 182 | 
 183 | 
 184 |     // Debugging function to dump a single dom node with a bunch of information about it.
 185 |     function dump_node($echo=true)
 186 |     {
 187 | 
 188 |         $string = $this->tag;
 189 |         if (count($this->attr)>0)
 190 |         {
 191 |             $string .= '(';
 192 |             foreach ($this->attr as $k=>$v)
 193 |             {
 194 |                 $string .= "[$k]=>\"".$this->$k.'", ';
 195 |             }
 196 |             $string .= ')';
 197 |         }
 198 |         if (count($this->_)>0)
 199 |         {
 200 |             $string .= ' $_ (';
 201 |             foreach ($this->_ as $k=>$v)
 202 |             {
 203 |                 if (is_array($v))
 204 |                 {
 205 |                     $string .= "[$k]=>(";
 206 |                     foreach ($v as $k2=>$v2)
 207 |                     {
 208 |                         $string .= "[$k2]=>\"".$v2.'", ';
 209 |                     }
 210 |                     $string .= ")";
 211 |                 } else {
 212 |                     $string .= "[$k]=>\"".$v.'", ';
 213 |                 }
 214 |             }
 215 |             $string .= ")";
 216 |         }
 217 | 
 218 |         if (isset($this->text))
 219 |         {
 220 |             $string .= " text: (" . $this->text . ")";
 221 |         }
 222 | 
 223 |         $string .= " HDOM_INNER_INFO: '";
 224 |         if (isset($node->_[HDOM_INFO_INNER]))
 225 |         {
 226 |             $string .= $node->_[HDOM_INFO_INNER] . "'";
 227 |         }
 228 |         else
 229 |         {
 230 |             $string .= ' NULL ';
 231 |         }
 232 | 
 233 |         $string .= " children: " . count($this->children);
 234 |         $string .= " nodes: " . count($this->nodes);
 235 |         $string .= " tag_start: " . $this->tag_start;
 236 |         $string .= "\n";
 237 | 
 238 |         if ($echo)
 239 |         {
 240 |             echo $string;
 241 |             return;
 242 |         }
 243 |         else
 244 |         {
 245 |             return $string;
 246 |         }
 247 |     }
 248 | 
 249 |     // returns the parent of node
 250 |     // If a node is passed in, it will reset the parent of the current node to that one.
 251 |     function parent($parent=null)
 252 |     {
 253 |         // I am SURE that this doesn't work properly.
 254 |         // It fails to unset the current node from it's current parents nodes or children list first.
 255 |         if ($parent !== null)
 256 |         {
 257 |             $this->parent = $parent;
 258 |             $this->parent->nodes[] = $this;
 259 |             $this->parent->children[] = $this;
 260 |         }
 261 | 
 262 |         return $this->parent;
 263 |     }
 264 | 
 265 |     // verify that node has children
 266 |     function has_child()
 267 |     {
 268 |         return !empty($this->children);
 269 |     }
 270 | 
 271 |     // returns children of node
 272 |     function children($idx=-1)
 273 |     {
 274 |         if ($idx===-1)
 275 |         {
 276 |             return $this->children;
 277 |         }
 278 |         if (isset($this->children[$idx])) return $this->children[$idx];
 279 |         return null;
 280 |     }
 281 | 
 282 |     // returns the first child of node
 283 |     function first_child()
 284 |     {
 285 |         if (count($this->children)>0)
 286 |         {
 287 |             return $this->children[0];
 288 |         }
 289 |         return null;
 290 |     }
 291 | 
 292 |     // returns the last child of node
 293 |     function last_child()
 294 |     {
 295 |         if (($count=count($this->children))>0)
 296 |         {
 297 |             return $this->children[$count-1];
 298 |         }
 299 |         return null;
 300 |     }
 301 | 
 302 |     // returns the next sibling of node
 303 |     function next_sibling()
 304 |     {
 305 |         if ($this->parent===null)
 306 |         {
 307 |             return null;
 308 |         }
 309 | 
 310 |         $idx = 0;
 311 |         $count = count($this->parent->children);
 312 |         while ($idx<$count && $this!==$this->parent->children[$idx])
 313 |         {
 314 |             ++$idx;
 315 |         }
 316 |         if (++$idx>=$count)
 317 |         {
 318 |             return null;
 319 |         }
 320 |         return $this->parent->children[$idx];
 321 |     }
 322 | 
 323 |     // returns the previous sibling of node
 324 |     function prev_sibling()
 325 |     {
 326 |         if ($this->parent===null) return null;
 327 |         $idx = 0;
 328 |         $count = count($this->parent->children);
 329 |         while ($idx<$count && $this!==$this->parent->children[$idx])
 330 |             ++$idx;
 331 |         if (--$idx<0) return null;
 332 |         return $this->parent->children[$idx];
 333 |     }
 334 | 
 335 |     // function to locate a specific ancestor tag in the path to the root.
 336 |     function find_ancestor_tag($tag)
 337 |     {
 338 |         global $debugObject;
 339 |         if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
 340 | 
 341 |         // Start by including ourselves in the comparison.
 342 |         $returnDom = $this;
 343 | 
 344 |         while (!is_null($returnDom))
 345 |         {
 346 |             if (is_object($debugObject)) { $debugObject->debugLog(2, "Current tag is: " . $returnDom->tag); }
 347 | 
 348 |             if ($returnDom->tag == $tag)
 349 |             {
 350 |                 break;
 351 |             }
 352 |             $returnDom = $returnDom->parent;
 353 |         }
 354 |         return $returnDom;
 355 |     }
 356 | 
 357 |     // get dom node's inner html
 358 |     function innertext()
 359 |     {
 360 |         if (isset($this->_[HDOM_INFO_INNER])) return $this->_[HDOM_INFO_INNER];
 361 |         if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
 362 | 
 363 |         $ret = '';
 364 |         foreach ($this->nodes as $n)
 365 |             $ret .= $n->outertext();
 366 |         return $ret;
 367 |     }
 368 | 
 369 |     // get dom node's outer text (with tag)
 370 |     function outertext()
 371 |     {
 372 |         global $debugObject;
 373 |         if (is_object($debugObject))
 374 |         {
 375 |             $text = '';
 376 |             if ($this->tag == 'text')
 377 |             {
 378 |                 if (!empty($this->text))
 379 |                 {
 380 |                     $text = " with text: " . $this->text;
 381 |                 }
 382 |             }
 383 |             $debugObject->debugLog(1, 'Innertext of tag: ' . $this->tag . $text);
 384 |         }
 385 | 
 386 |         if ($this->tag==='root') return $this->innertext();
 387 | 
 388 |         // trigger callback
 389 |         if ($this->dom && $this->dom->callback!==null)
 390 |         {
 391 |             call_user_func_array($this->dom->callback, array($this));
 392 |         }
 393 | 
 394 |         if (isset($this->_[HDOM_INFO_OUTER])) return $this->_[HDOM_INFO_OUTER];
 395 |         if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
 396 | 
 397 |         // render begin tag
 398 |         if ($this->dom && $this->dom->nodes[$this->_[HDOM_INFO_BEGIN]])
 399 |         {
 400 |             $ret = $this->dom->nodes[$this->_[HDOM_INFO_BEGIN]]->makeup();
 401 |         } else {
 402 |             $ret = "";
 403 |         }
 404 | 
 405 |         // render inner text
 406 |         if (isset($this->_[HDOM_INFO_INNER]))
 407 |         {
 408 |             // If it's a br tag...  don't return the HDOM_INNER_INFO that we may or may not have added.
 409 |             if ($this->tag != "br")
 410 |             {
 411 |                 $ret .= $this->_[HDOM_INFO_INNER];
 412 |             }
 413 |         } else {
 414 |             if ($this->nodes)
 415 |             {
 416 |                 foreach ($this->nodes as $n)
 417 |                 {
 418 |                     $ret .= $this->convert_text($n->outertext());
 419 |                 }
 420 |             }
 421 |         }
 422 | 
 423 |         // render end tag
 424 |         if (isset($this->_[HDOM_INFO_END]) && $this->_[HDOM_INFO_END]!=0)
 425 |             $ret .= '</'.$this->tag.'>';
 426 |         return $ret;
 427 |     }
 428 | 
 429 |     // get dom node's plain text
 430 |     function text()
 431 |     {
 432 |         if (isset($this->_[HDOM_INFO_INNER])) return $this->_[HDOM_INFO_INNER];
 433 |         switch ($this->nodetype)
 434 |         {
 435 |             case HDOM_TYPE_TEXT: return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
 436 |             case HDOM_TYPE_COMMENT: return '';
 437 |             case HDOM_TYPE_UNKNOWN: return '';
 438 |         }
 439 |         if (strcasecmp($this->tag, 'script')===0) return '';
 440 |         if (strcasecmp($this->tag, 'style')===0) return '';
 441 | 
 442 |         $ret = '';
 443 |         // In rare cases, (always node type 1 or HDOM_TYPE_ELEMENT - observed for some span tags, and some p tags) $this->nodes is set to NULL.
 444 |         // NOTE: This indicates that there is a problem where it's set to NULL without a clear happening.
 445 |         // WHY is this happening?
 446 |         if (!is_null($this->nodes))
 447 |         {
 448 |             foreach ($this->nodes as $n)
 449 |             {
 450 |                 $ret .= $this->convert_text($n->text());
 451 |             }
 452 | 
 453 |             // 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.
 454 |             if ($this->tag == "span")
 455 |             {
 456 |                 $ret .= $this->dom->default_span_text;
 457 |             }
 458 | 
 459 | 
 460 |         }
 461 |         return $ret;
 462 |     }
 463 | 
 464 |     function xmltext()
 465 |     {
 466 |         $ret = $this->innertext();
 467 |         $ret = str_ireplace('<![CDATA[', '', $ret);
 468 |         $ret = str_replace(']]>', '', $ret);
 469 |         return $ret;
 470 |     }
 471 | 
 472 |     // build node's text with tag
 473 |     function makeup()
 474 |     {
 475 |         // text, comment, unknown
 476 |         if (isset($this->_[HDOM_INFO_TEXT])) return $this->dom->restore_noise($this->_[HDOM_INFO_TEXT]);
 477 | 
 478 |         $ret = '<'.$this->tag;
 479 |         $i = -1;
 480 | 
 481 |         foreach ($this->attr as $key=>$val)
 482 |         {
 483 |             ++$i;
 484 | 
 485 |             // skip removed attribute
 486 |             if ($val===null || $val===false)
 487 |                 continue;
 488 | 
 489 |             $ret .= $this->_[HDOM_INFO_SPACE][$i][0];
 490 |             //no value attr: nowrap, checked selected...
 491 |             if ($val===true)
 492 |                 $ret .= $key;
 493 |             else {
 494 |                 switch ($this->_[HDOM_INFO_QUOTE][$i])
 495 |                 {
 496 |                     case HDOM_QUOTE_DOUBLE: $quote = '"'; break;
 497 |                     case HDOM_QUOTE_SINGLE: $quote = '\''; break;
 498 |                     default: $quote = '';
 499 |                 }
 500 |                 $ret .= $key.$this->_[HDOM_INFO_SPACE][$i][1].'='.$this->_[HDOM_INFO_SPACE][$i][2].$quote.$val.$quote;
 501 |             }
 502 |         }
 503 |         $ret = $this->dom->restore_noise($ret);
 504 |         return $ret . $this->_[HDOM_INFO_ENDSPACE] . '>';
 505 |     }
 506 | 
 507 |     /**
 508 |      * find elements by css selector
 509 |      * PaperG - added ability for find to lowercase the value of the selector.
 510 |      * @param string   $selector
 511 |      * @param int|null $idx
 512 |      * @param bool     $lowercase
 513 |      * @return simple_html_dom_node[]|simple_html_dom_node|null
 514 |      */
 515 |     function find($selector, $idx=null, $lowercase=false)
 516 |     {
 517 |         $selectors = $this->parse_selector($selector);
 518 |         if (($count=count($selectors))===0) return array();
 519 |         $found_keys = array();
 520 | 
 521 |         // find each selector
 522 |         for ($c=0; $c<$count; ++$c)
 523 |         {
 524 |             // The change on the below line was documented on the sourceforge code tracker id 2788009
 525 |             // used to be: if (($levle=count($selectors[0]))===0) return array();
 526 |             if (($levle=count($selectors[$c]))===0) return array();
 527 |             if (!isset($this->_[HDOM_INFO_BEGIN])) return array();
 528 | 
 529 |             $head = array($this->_[HDOM_INFO_BEGIN]=>1);
 530 | 
 531 |             // handle descendant selectors, no recursive!
 532 |             for ($l=0; $l<$levle; ++$l)
 533 |             {
 534 |                 $ret = array();
 535 |                 foreach ($head as $k=>$v)
 536 |                 {
 537 |                     $n = ($k===-1) ? $this->dom->root : $this->dom->nodes[$k];
 538 |                     //PaperG - Pass this optional parameter on to the seek function.
 539 |                     $n->seek($selectors[$c][$l], $ret, $lowercase);
 540 |                 }
 541 |                 $head = $ret;
 542 |             }
 543 | 
 544 |             foreach ($head as $k=>$v)
 545 |             {
 546 |                 if (!isset($found_keys[$k]))
 547 |                     $found_keys[$k] = 1;
 548 |             }
 549 |         }
 550 | 
 551 |         // sort keys
 552 |         ksort($found_keys);
 553 | 
 554 |         $found = array();
 555 |         foreach ($found_keys as $k=>$v)
 556 |             $found[] = $this->dom->nodes[$k];
 557 | 
 558 |         // return nth-element or array
 559 |         if (is_null($idx)) return $found;
 560 |         else if ($idx<0) $idx = count($found) + $idx;
 561 |         return (isset($found[$idx])) ? $found[$idx] : null;
 562 |     }
 563 | 
 564 |     // seek for given conditions
 565 |     // PaperG - added parameter to allow for case insensitive testing of the value of a selector.
 566 |     protected function seek($selector, &$ret, $lowercase=false)
 567 |     {
 568 |         global $debugObject;
 569 |         if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
 570 | 
 571 |         list($tag, $key, $val, $exp, $no_key) = $selector;
 572 | 
 573 |         // xpath index
 574 |         if ($tag && $key && is_numeric($key))
 575 |         {
 576 |             $count = 0;
 577 |             foreach ($this->children as $c)
 578 |             {
 579 |                 if ($tag==='*' || $tag===$c->tag) {
 580 |                     if (++$count==$key) {
 581 |                         $ret[$c->_[HDOM_INFO_BEGIN]] = 1;
 582 |                         return;
 583 |                     }
 584 |                 }
 585 |             }
 586 |             return;
 587 |         }
 588 | 
 589 |         $end = (!empty($this->_[HDOM_INFO_END])) ? $this->_[HDOM_INFO_END] : 0;
 590 |         if ($end==0) {
 591 |             $parent = $this->parent;
 592 |             while (!isset($parent->_[HDOM_INFO_END]) && $parent!==null) {
 593 |                 $end -= 1;
 594 |                 $parent = $parent->parent;
 595 |             }
 596 |             $end += $parent->_[HDOM_INFO_END];
 597 |         }
 598 | 
 599 |         for ($i=$this->_[HDOM_INFO_BEGIN]+1; $i<$end; ++$i) {
 600 |             $node = $this->dom->nodes[$i];
 601 | 
 602 |             $pass = true;
 603 | 
 604 |             if ($tag==='*' && !$key) {
 605 |                 if (in_array($node, $this->children, true))
 606 |                     $ret[$i] = 1;
 607 |                 continue;
 608 |             }
 609 | 
 610 |             // compare tag
 611 |             if ($tag && $tag!=$node->tag && $tag!=='*') {$pass=false;}
 612 |             // compare key
 613 |             if ($pass && $key) {
 614 |                 if ($no_key) {
 615 |                     if (isset($node->attr[$key])) $pass=false;
 616 |                 } else {
 617 |                     if (($key != "plaintext") && !isset($node->attr[$key])) $pass=false;
 618 |                 }
 619 |             }
 620 |             // compare value
 621 |             if ($pass && $key && $val  && $val!=='*') {
 622 |                 // If they have told us that this is a "plaintext" search then we want the plaintext of the node - right?
 623 |                 if ($key == "plaintext") {
 624 |                     // $node->plaintext actually returns $node->text();
 625 |                     $nodeKeyValue = $node->text();
 626 |                 } else {
 627 |                     // this is a normal search, we want the value of that attribute of the tag.
 628 |                     $nodeKeyValue = $node->attr[$key];
 629 |                 }
 630 |                 if (is_object($debugObject)) {$debugObject->debugLog(2, "testing node: " . $node->tag . " for attribute: " . $key . $exp . $val . " where nodes value is: " . $nodeKeyValue);}
 631 | 
 632 |                 //PaperG - If lowercase is set, do a case insensitive test of the value of the selector.
 633 |                 if ($lowercase) {
 634 |                     $check = $this->match($exp, strtolower($val), strtolower($nodeKeyValue));
 635 |                 } else {
 636 |                     $check = $this->match($exp, $val, $nodeKeyValue);
 637 |                 }
 638 |                 if (is_object($debugObject)) {$debugObject->debugLog(2, "after match: " . ($check ? "true" : "false"));}
 639 | 
 640 |                 // handle multiple class
 641 |                 if (!$check && strcasecmp($key, 'class')===0) {
 642 |                     foreach (explode(' ',$node->attr[$key]) as $k) {
 643 |                         // Without this, there were cases where leading, trailing, or double spaces lead to our comparing blanks - bad form.
 644 |                         if (!empty($k)) {
 645 |                             if ($lowercase) {
 646 |                                 $check = $this->match($exp, strtolower($val), strtolower($k));
 647 |                             } else {
 648 |                                 $check = $this->match($exp, $val, $k);
 649 |                             }
 650 |                             if ($check) break;
 651 |                         }
 652 |                     }
 653 |                 }
 654 |                 if (!$check) $pass = false;
 655 |             }
 656 |             if ($pass) $ret[$i] = 1;
 657 |             unset($node);
 658 |         }
 659 |         // It's passed by reference so this is actually what this function returns.
 660 |         if (is_object($debugObject)) {$debugObject->debugLog(1, "EXIT - ret: ", $ret);}
 661 |     }
 662 | 
 663 |     protected function match($exp, $pattern, $value) {
 664 |         global $debugObject;
 665 |         if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
 666 | 
 667 |         switch ($exp) {
 668 |             case '=':
 669 |                 return ($value===$pattern);
 670 |             case '!=':
 671 |                 return ($value!==$pattern);
 672 |             case '^=':
 673 |                 return preg_match("/^".preg_quote($pattern,'/')."/", $value);
 674 |             case '$=':
 675 |                 return preg_match("/".preg_quote($pattern,'/')."$/", $value);
 676 |             case '*=':
 677 |                 if ($pattern[0]=='/') {
 678 |                     return preg_match($pattern, $value);
 679 |                 }
 680 |                 return preg_match("/".$pattern."/i", $value);
 681 |         }
 682 |         return false;
 683 |     }
 684 | 
 685 |     protected function parse_selector($selector_string) {
 686 |         global $debugObject;
 687 |         if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
 688 | 
 689 |         // pattern of CSS selectors, modified from mootools
 690 |         // Paperg: Add the colon to the attrbute, so that it properly finds <tag attr:ibute="something" > like google does.
 691 |         // Note: if you try to look at this attribute, yo MUST use getAttribute since $dom->x:y will fail the php syntax check.
 692 | // Notice the \[ starting the attbute?  and the @? following?  This implies that an attribute can begin with an @ sign that is not captured.
 693 | // This implies that an html attribute specifier may start with an @ sign that is NOT captured by the expression.
 694 | // farther study is required to determine of this should be documented or removed.
 695 | //        $pattern = "/([\w-:\*]*)(?:\#([\w-]+)|\.([\w-]+))?(?:\[@?(!?[\w-]+)(?:([!*^$]?=)[\"']?(.*?)[\"']?)?\])?([\/, ]+)/is";
 696 |         $pattern = "/([\w-:\*]*)(?:\#([\w-]+)|\.([\w-]+))?(?:\[@?(!?[\w-:]+)(?:([!*^$]?=)[\"']?(.*?)[\"']?)?\])?([\/, ]+)/is";
 697 |         preg_match_all($pattern, trim($selector_string).' ', $matches, PREG_SET_ORDER);
 698 |         if (is_object($debugObject)) {$debugObject->debugLog(2, "Matches Array: ", $matches);}
 699 | 
 700 |         $selectors = array();
 701 |         $result = array();
 702 |         //print_r($matches);
 703 | 
 704 |         foreach ($matches as $m) {
 705 |             $m[0] = trim($m[0]);
 706 |             if ($m[0]==='' || $m[0]==='/' || $m[0]==='//') continue;
 707 |             // for browser generated xpath
 708 |             if ($m[1]==='tbody') continue;
 709 | 
 710 |             list($tag, $key, $val, $exp, $no_key) = array($m[1], null, null, '=', false);
 711 |             if (!empty($m[2])) {$key='id'; $val=$m[2];}
 712 |             if (!empty($m[3])) {$key='class'; $val=$m[3];}
 713 |             if (!empty($m[4])) {$key=$m[4];}
 714 |             if (!empty($m[5])) {$exp=$m[5];}
 715 |             if (!empty($m[6])) {$val=$m[6];}
 716 | 
 717 |             // convert to lowercase
 718 |             if ($this->dom->lowercase) {$tag=strtolower($tag); $key=strtolower($key);}
 719 |             //elements that do NOT have the specified attribute
 720 |             if (isset($key[0]) && $key[0]==='!') {$key=substr($key, 1); $no_key=true;}
 721 | 
 722 |             $result[] = array($tag, $key, $val, $exp, $no_key);
 723 |             if (trim($m[7])===',') {
 724 |                 $selectors[] = $result;
 725 |                 $result = array();
 726 |             }
 727 |         }
 728 |         if (count($result)>0)
 729 |             $selectors[] = $result;
 730 |         return $selectors;
 731 |     }
 732 | 
 733 |     function __get($name) {
 734 |         if (isset($this->attr[$name]))
 735 |         {
 736 |             return $this->convert_text($this->attr[$name]);
 737 |         }
 738 |         switch ($name) {
 739 |             case 'outertext': return $this->outertext();
 740 |             case 'innertext': return $this->innertext();
 741 |             case 'plaintext': return $this->text();
 742 |             case 'xmltext': return $this->xmltext();
 743 |             default: return array_key_exists($name, $this->attr);
 744 |         }
 745 |     }
 746 | 
 747 |     function __set($name, $value) {
 748 |         switch ($name) {
 749 |             case 'outertext': return $this->_[HDOM_INFO_OUTER] = $value;
 750 |             case 'innertext':
 751 |                 if (isset($this->_[HDOM_INFO_TEXT])) return $this->_[HDOM_INFO_TEXT] = $value;
 752 |                 return $this->_[HDOM_INFO_INNER] = $value;
 753 |         }
 754 |         if (!isset($this->attr[$name])) {
 755 |             $this->_[HDOM_INFO_SPACE][] = array(' ', '', '');
 756 |             $this->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_DOUBLE;
 757 |         }
 758 |         $this->attr[$name] = $value;
 759 |     }
 760 | 
 761 |     function __isset($name) {
 762 |         switch ($name) {
 763 |             case 'outertext': return true;
 764 |             case 'innertext': return true;
 765 |             case 'plaintext': return true;
 766 |         }
 767 |         //no value attr: nowrap, checked selected...
 768 |         return (array_key_exists($name, $this->attr)) ? true : isset($this->attr[$name]);
 769 |     }
 770 | 
 771 |     function __unset($name) {
 772 |         if (isset($this->attr[$name]))
 773 |             unset($this->attr[$name]);
 774 |     }
 775 | 
 776 |     // PaperG - Function to convert the text from one character set to another if the two sets are not the same.
 777 |     function convert_text($text)
 778 |     {
 779 |         global $debugObject;
 780 |         if (is_object($debugObject)) {$debugObject->debugLogEntry(1);}
 781 | 
 782 |         $converted_text = $text;
 783 | 
 784 |         $sourceCharset = "";
 785 |         $targetCharset = "";
 786 | 
 787 |         if ($this->dom)
 788 |         {
 789 |             $sourceCharset = strtoupper($this->dom->_charset);
 790 |             $targetCharset = strtoupper($this->dom->_target_charset);
 791 |         }
 792 |         if (is_object($debugObject)) {$debugObject->debugLog(3, "source charset: " . $sourceCharset . " target charaset: " . $targetCharset);}
 793 | 
 794 |         if (!empty($sourceCharset) && !empty($targetCharset) && (strcasecmp($sourceCharset, $targetCharset) != 0))
 795 |         {
 796 |             // Check if the reported encoding could have been incorrect and the text is actually already UTF-8
 797 |             if ((strcasecmp($targetCharset, 'UTF-8') == 0) && ($this->is_utf8($text)))
 798 |             {
 799 |                 $converted_text = $text;
 800 |             }
 801 |             else
 802 |             {
 803 |                 $converted_text = iconv($sourceCharset, $targetCharset, $text);
 804 |             }
 805 |         }
 806 | 
 807 |         // Lets make sure that we don't have that silly BOM issue with any of the utf-8 text we output.
 808 |         if ($targetCharset == 'UTF-8')
 809 |         {
 810 |             if (substr($converted_text, 0, 3) == "\xef\xbb\xbf")
 811 |             {
 812 |                 $converted_text = substr($converted_text, 3);
 813 |             }
 814 |             if (substr($converted_text, -3) == "\xef\xbb\xbf")
 815 |             {
 816 |                 $converted_text = substr($converted_text, 0, -3);
 817 |             }
 818 |         }
 819 | 
 820 |         return $converted_text;
 821 |     }
 822 | 
 823 |     /**
 824 |     * Returns true if $string is valid UTF-8 and false otherwise.
 825 |     *
 826 |     * @param mixed $str String to be tested
 827 |     * @return boolean
 828 |     */
 829 |     static function is_utf8($str)
 830 |     {
 831 |         $c=0; $b=0;
 832 |         $bits=0;
 833 |         $len=strlen($str);
 834 |         for($i=0; $i<$len; $i++)
 835 |         {
 836 |             $c=ord($str[$i]);
 837 |             if($c > 128)
 838 |             {
 839 |                 if(($c >= 254)) return false;
 840 |                 elseif($c >= 252) $bits=6;
 841 |                 elseif($c >= 248) $bits=5;
 842 |                 elseif($c >= 240) $bits=4;
 843 |                 elseif($c >= 224) $bits=3;
 844 |                 elseif($c >= 192) $bits=2;
 845 |                 else return false;
 846 |                 if(($i+$bits) > $len) return false;
 847 |                 while($bits > 1)
 848 |                 {
 849 |                     $i++;
 850 |                     $b=ord($str[$i]);
 851 |                     if($b < 128 || $b > 191) return false;
 852 |                     $bits--;
 853 |                 }
 854 |             }
 855 |         }
 856 |         return true;
 857 |     }
 858 |     /*
 859 |     function is_utf8($string)
 860 |     {
 861 |         //this is buggy
 862 |         return (utf8_encode(utf8_decode($string)) == $string);
 863 |     }
 864 |     */
 865 | 
 866 |     /**
 867 |      * Function to try a few tricks to determine the displayed size of an img on the page.
 868 |      * NOTE: This will ONLY work on an IMG tag. Returns FALSE on all other tag types.
 869 |      *
 870 |      * @author John Schlick
 871 |      * @version April 19 2012
 872 |      * @return array an array containing the 'height' and 'width' of the image on the page or -1 if we can't figure it out.
 873 |      */
 874 |     function get_display_size()
 875 |     {
 876 |         global $debugObject;
 877 | 
 878 |         $width = -1;
 879 |         $height = -1;
 880 | 
 881 |         if ($this->tag !== 'img')
 882 |         {
 883 |             return false;
 884 |         }
 885 | 
 886 |         // See if there is aheight or width attribute in the tag itself.
 887 |         if (isset($this->attr['width']))
 888 |         {
 889 |             $width = $this->attr['width'];
 890 |         }
 891 | 
 892 |         if (isset($this->attr['height']))
 893 |         {
 894 |             $height = $this->attr['height'];
 895 |         }
 896 | 
 897 |         // Now look for an inline style.
 898 |         if (isset($this->attr['style']))
 899 |         {
 900 |             // Thanks to user gnarf from stackoverflow for this regular expression.
 901 |             $attributes = array();
 902 |             preg_match_all("/([\w-]+)\s*:\s*([^;]+)\s*;?/", $this->attr['style'], $matches, PREG_SET_ORDER);
 903 |             foreach ($matches as $match) {
 904 |               $attributes[$match[1]] = $match[2];
 905 |             }
 906 | 
 907 |             // If there is a width in the style attributes:
 908 |             if (isset($attributes['width']) && $width == -1)
 909 |             {
 910 |                 // check that the last two characters are px (pixels)
 911 |                 if (strtolower(substr($attributes['width'], -2)) == 'px')
 912 |                 {
 913 |                     $proposed_width = substr($attributes['width'], 0, -2);
 914 |                     // Now make sure that it's an integer and not something stupid.
 915 |                     if (filter_var($proposed_width, FILTER_VALIDATE_INT))
 916 |                     {
 917 |                         $width = $proposed_width;
 918 |                     }
 919 |                 }
 920 |             }
 921 | 
 922 |             // If there is a width in the style attributes:
 923 |             if (isset($attributes['height']) && $height == -1)
 924 |             {
 925 |                 // check that the last two characters are px (pixels)
 926 |                 if (strtolower(substr($attributes['height'], -2)) == 'px')
 927 |                 {
 928 |                     $proposed_height = substr($attributes['height'], 0, -2);
 929 |                     // Now make sure that it's an integer and not something stupid.
 930 |                     if (filter_var($proposed_height, FILTER_VALIDATE_INT))
 931 |                     {
 932 |                         $height = $proposed_height;
 933 |                     }
 934 |                 }
 935 |             }
 936 | 
 937 |         }
 938 | 
 939 |         // Future enhancement:
 940 |         // Look in the tag to see if there is a class or id specified that has a height or width attribute to it.
 941 | 
 942 |         // Far future enhancement
 943 |         // 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
 944 |         // Note that in this case, the class or id will have the img subselector for it to apply to the image.
 945 | 
 946 |         // ridiculously far future development
 947 |         // 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.
 948 | 
 949 |         $result = array('height' => $height,
 950 |                         'width' => $width);
 951 |         return $result;
 952 |     }
 953 | 
 954 |     // camel naming conventions
 955 |     function getAllAttributes() {return array_map('html_entity_decode', $this->attr);}
 956 |     function getAttribute($name) {return html_entity_decode($this->__get($name));}
 957 |     function setAttribute($name, $value) {$this->__set($name, $value);}
 958 |     function hasAttribute($name) {return $this->__isset($name);}
 959 |     function removeAttribute($name) {$this->__set($name, null);}
 960 |     function getElementById($id) {return $this->find("#$id", 0);}
 961 |     function getElementsById($id, $idx=null) {return $this->find("#$id", $idx);}
 962 |     function getElementByTagName($name) {return $this->find($name, 0);}
 963 |     function getElementsByTagName($name, $idx=null) {return $this->find($name, $idx);}
 964 |     function parentNode() {return $this->parent();}
 965 |     function childNodes($idx=-1) {return $this->children($idx);}
 966 |     function firstChild() {return $this->first_child();}
 967 |     function lastChild() {return $this->last_child();}
 968 |     function nextSibling() {return $this->next_sibling();}
 969 |     function previousSibling() {return $this->prev_sibling();}
 970 |     function hasChildNodes() {return $this->has_child();}
 971 |     function nodeName() {return $this->tag;}
 972 |     function appendChild($node) {$node->parent($this); return $node;}
 973 | 
 974 | }
 975 | 
 976 | /**
 977 |  * simple html dom parser
 978 |  * Paperg - in the find routine: allow us to specify that we want case insensitive testing of the value of the selector.
 979 |  * Paperg - change $size from protected to public so we can easily access it
 980 |  * Paperg - added ForceTagsClosed in the constructor which tells us whether we trust the html or not.  Default is to NOT trust it.
 981 |  *
 982 |  * @package PlaceLocalInclude
 983 |  */
 984 | class simple_html_dom
 985 | {
 986 |     /** @var simple_html_dom_node $root */
 987 |     public $root = null;
 988 |     public $nodes = array();
 989 |     public $callback = null;
 990 |     public $lowercase = false;
 991 |     // Used to keep track of how large the text was when we started.
 992 |     public $original_size;
 993 |     public $size;
 994 |     protected $pos;
 995 |     protected $doc;
 996 |     protected $char;
 997 |     protected $cursor;
 998 |     protected $parent;
 999 |     protected $noise = array();
1000 |     protected $token_blank = " \t\r\n";
1001 |     protected $token_equal = ' =/>';
1002 |     protected $token_slash = " />\r\n\t";
1003 |     protected $token_attr = ' >';
1004 |     // Note that this is referenced by a child node, and so it needs to be public for that node to see this information.
1005 |     public $_charset = '';
1006 |     public $_target_charset = '';
1007 |     protected $default_br_text = "";
1008 |     public $default_span_text = "";
1009 | 
1010 |     // use isset instead of in_array, performance boost about 30%...
1011 |     protected $self_closing_tags = array('img'=>1, 'br'=>1, 'input'=>1, 'meta'=>1, 'link'=>1, 'hr'=>1, 'base'=>1, 'embed'=>1, 'spacer'=>1);
1012 |     protected $block_tags = array('root'=>1, 'body'=>1, 'form'=>1, 'div'=>1, 'span'=>1, 'table'=>1);
1013 |     // Known sourceforge issue #2977341
1014 |     // B tags that are not closed cause us to return everything to the end of the document.
1015 |     protected $optional_closing_tags = array(
1016 |         'tr'=>array('tr'=>1, 'td'=>1, 'th'=>1),
1017 |         'th'=>array('th'=>1),
1018 |         'td'=>array('td'=>1),
1019 |         'li'=>array('li'=>1),
1020 |         'dt'=>array('dt'=>1, 'dd'=>1),
1021 |         'dd'=>array('dd'=>1, 'dt'=>1),
1022 |         'dl'=>array('dd'=>1, 'dt'=>1),
1023 |         'p'=>array('p'=>1),
1024 |         'nobr'=>array('nobr'=>1),
1025 |         'b'=>array('b'=>1),
1026 | 		'option'=>array('option'=>1),
1027 |     );
1028 | 
1029 |     function __construct($str=null, $lowercase=true, $forceTagsClosed=true, $target_charset=DEFAULT_TARGET_CHARSET, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
1030 |     {
1031 |         if ($str)
1032 |         {
1033 |             if (preg_match("/^http:\/\//i",$str) || is_file($str))
1034 |             {
1035 |                 $this->load_file($str);
1036 |             }
1037 |             else
1038 |             {
1039 |                 $this->load($str, $lowercase, $stripRN, $defaultBRText, $defaultSpanText);
1040 |             }
1041 |         }
1042 |         // Forcing tags to be closed implies that we don't trust the html, but it can lead to parsing errors if we SHOULD trust the html.
1043 |         if (!$forceTagsClosed) {
1044 |             $this->optional_closing_array=array();
1045 |         }
1046 |         $this->_target_charset = $target_charset;
1047 |     }
1048 | 
1049 |     function __destruct()
1050 |     {
1051 |         $this->clear();
1052 |     }
1053 | 
1054 |     // load html from string
1055 |     function load($str, $lowercase=true, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
1056 |     {
1057 |         global $debugObject;
1058 | 
1059 |         // prepare
1060 |         $this->prepare($str, $lowercase, $stripRN, $defaultBRText, $defaultSpanText);
1061 |         // strip out comments
1062 |         $this->remove_noise("'<!--(.*?)-->'is");
1063 |         // strip out cdata
1064 |         $this->remove_noise("'<!\[CDATA\[(.*?)\]\]>'is", true);
1065 |         // Per sourceforge http://sourceforge.net/tracker/?func=detail&aid=2949097&group_id=218559&atid=1044037
1066 |         // Script tags removal now preceeds style tag removal.
1067 |         // strip out <script> tags
1068 |         $this->remove_noise("'<\s*script[^>]*[^/]>(.*?)<\s*/\s*script\s*>'is");
1069 |         $this->remove_noise("'<\s*script\s*>(.*?)<\s*/\s*script\s*>'is");
1070 |         // strip out <style> tags
1071 |         $this->remove_noise("'<\s*style[^>]*[^/]>(.*?)<\s*/\s*style\s*>'is");
1072 |         $this->remove_noise("'<\s*style\s*>(.*?)<\s*/\s*style\s*>'is");
1073 |         // strip out preformatted tags
1074 |         $this->remove_noise("'<\s*(?:code)[^>]*>(.*?)<\s*/\s*(?:code)\s*>'is");
1075 |         // strip out server side scripts
1076 |         $this->remove_noise("'(<\?)(.*?)(\?>)'s", true);
1077 |         // strip smarty scripts
1078 |         $this->remove_noise("'(\{\w)(.*?)(\})'s", true);
1079 | 
1080 |         // parsing
1081 |         while ($this->parse());
1082 |         // end
1083 |         $this->root->_[HDOM_INFO_END] = $this->cursor;
1084 |         $this->parse_charset();
1085 | 
1086 |         // make load function chainable
1087 |         return $this;
1088 | 
1089 |     }
1090 | 
1091 |     // load html from file
1092 |     function load_file()
1093 |     {
1094 |         $args = func_get_args();
1095 |         $this->load(call_user_func_array('file_get_contents', $args), true);
1096 |         // Throw an error if we can't properly load the dom.
1097 |         if (($error=error_get_last())!==null) {
1098 |             $this->clear();
1099 |             return false;
1100 |         }
1101 |     }
1102 | 
1103 |     // set callback function
1104 |     function set_callback($function_name)
1105 |     {
1106 |         $this->callback = $function_name;
1107 |     }
1108 | 
1109 |     // remove callback function
1110 |     function remove_callback()
1111 |     {
1112 |         $this->callback = null;
1113 |     }
1114 | 
1115 |     // save dom as string
1116 |     function save($filepath='')
1117 |     {
1118 |         $ret = $this->root->innertext();
1119 |         if ($filepath!=='') file_put_contents($filepath, $ret, LOCK_EX);
1120 |         return $ret;
1121 |     }
1122 | 
1123 |     // find dom node by css selector
1124 |     // Paperg - allow us to specify that we want case insensitive testing of the value of the selector.
1125 |     function find($selector, $idx=null, $lowercase=false)
1126 |     {
1127 |         return $this->root->find($selector, $idx, $lowercase);
1128 |     }
1129 | 
1130 |     // clean up memory due to php5 circular references memory leak...
1131 |     function clear()
1132 |     {
1133 |         foreach ($this->nodes as $n) {$n->clear(); $n = null;}
1134 |         // This add next line is documented in the sourceforge repository. 2977248 as a fix for ongoing memory leaks that occur even with the use of clear.
1135 |         if (isset($this->children)) foreach ($this->children as $n) {$n->clear(); $n = null;}
1136 |         if (isset($this->parent)) {$this->parent->clear(); unset($this->parent);}
1137 |         if (isset($this->root)) {$this->root->clear(); unset($this->root);}
1138 |         unset($this->doc);
1139 |         unset($this->noise);
1140 |     }
1141 | 
1142 |     function dump($show_attr=true)
1143 |     {
1144 |         $this->root->dump($show_attr);
1145 |     }
1146 | 
1147 |     // prepare HTML data and init everything
1148 |     protected function prepare($str, $lowercase=true, $stripRN=true, $defaultBRText=DEFAULT_BR_TEXT, $defaultSpanText=DEFAULT_SPAN_TEXT)
1149 |     {
1150 |         $this->clear();
1151 | 
1152 |         // set the length of content before we do anything to it.
1153 |         $this->size = strlen($str);
1154 |         // Save the original size of the html that we got in.  It might be useful to someone.
1155 |         $this->original_size = $this->size;
1156 | 
1157 |         //before we save the string as the doc...  strip out the \r \n's if we are told to.
1158 |         if ($stripRN) {
1159 |             $str = str_replace("\r", " ", $str);
1160 |             $str = str_replace("\n", " ", $str);
1161 | 
1162 |             // set the length of content since we have changed it.
1163 |             $this->size = strlen($str);
1164 |         }
1165 | 
1166 |         $this->doc = $str;
1167 |         $this->pos = 0;
1168 |         $this->cursor = 1;
1169 |         $this->noise = array();
1170 |         $this->nodes = array();
1171 |         $this->lowercase = $lowercase;
1172 |         $this->default_br_text = $defaultBRText;
1173 |         $this->default_span_text = $defaultSpanText;
1174 |         $this->root = new simple_html_dom_node($this);
1175 |         $this->root->tag = 'root';
1176 |         $this->root->_[HDOM_INFO_BEGIN] = -1;
1177 |         $this->root->nodetype = HDOM_TYPE_ROOT;
1178 |         $this->parent = $this->root;
1179 |         if ($this->size>0) $this->char = $this->doc[0];
1180 |     }
1181 | 
1182 |     // parse html content
1183 |     protected function parse()
1184 |     {
1185 |         if (($s = $this->copy_until_char('<'))==='')
1186 |         {
1187 |             return $this->read_tag();
1188 |         }
1189 | 
1190 |         // text
1191 |         $node = new simple_html_dom_node($this);
1192 |         ++$this->cursor;
1193 |         $node->_[HDOM_INFO_TEXT] = $s;
1194 |         $this->link_nodes($node, false);
1195 |         return true;
1196 |     }
1197 | 
1198 |     // 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.
1199 |     // 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
1200 |     // (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.
1201 |     protected function parse_charset()
1202 |     {
1203 |         global $debugObject;
1204 | 
1205 |         $charset = null;
1206 | 
1207 |         if (function_exists('get_last_retrieve_url_contents_content_type'))
1208 |         {
1209 |             $contentTypeHeader = get_last_retrieve_url_contents_content_type();
1210 |             $success = preg_match('/charset=(.+)/', $contentTypeHeader, $matches);
1211 |             if ($success)
1212 |             {
1213 |                 $charset = $matches[1];
1214 |                 if (is_object($debugObject)) {$debugObject->debugLog(2, 'header content-type found charset of: ' . $charset);}
1215 |             }
1216 | 
1217 |         }
1218 | 
1219 |         if (empty($charset))
1220 |         {
1221 |             $el = $this->root->find('meta[http-equiv=Content-Type]',0);
1222 |             if (!empty($el))
1223 |             {
1224 |                 $fullvalue = $el->content;
1225 |                 if (is_object($debugObject)) {$debugObject->debugLog(2, 'meta content-type tag found' . $fullvalue);}
1226 | 
1227 |                 if (!empty($fullvalue))
1228 |                 {
1229 |                     $success = preg_match('/charset=(.+)/', $fullvalue, $matches);
1230 |                     if ($success)
1231 |                     {
1232 |                         $charset = $matches[1];
1233 |                     }
1234 |                     else
1235 |                     {
1236 |                         // If there is a meta tag, and they don't specify the character set, research says that it's typically ISO-8859-1
1237 |                         if (is_object($debugObject)) {$debugObject->debugLog(2, 'meta content-type tag couldn\'t be parsed. using iso-8859 default.');}
1238 |                         $charset = 'ISO-8859-1';
1239 |                     }
1240 |                 }
1241 |             }
1242 |         }
1243 | 
1244 |         // If we couldn't find a charset above, then lets try to detect one based on the text we got...
1245 |         if (empty($charset))
1246 |         {
1247 |             // Have php try to detect the encoding from the text given to us.
1248 |             $charset = (function_exists('mb_detect_encoding')) ? mb_detect_encoding($this->root->plaintext . "ascii", $encoding_list = array( "UTF-8", "CP1252" ) ) : false;
1249 |             if (is_object($debugObject)) {$debugObject->debugLog(2, 'mb_detect found: ' . $charset);}
1250 | 
1251 |             // and if this doesn't work...  then we need to just wrongheadedly assume it's UTF-8 so that we can move on - cause this will usually give us most of what we need...
1252 |             if ($charset === false)
1253 |             {
1254 |                 if (is_object($debugObject)) {$debugObject->debugLog(2, 'since mb_detect failed - using default of utf-8');}
1255 |                 $charset = 'UTF-8';
1256 |             }
1257 |         }
1258 | 
1259 |         // Since CP1252 is a superset, if we get one of it's subsets, we want it instead.
1260 |         if ((strtolower($charset) == strtolower('ISO-8859-1')) || (strtolower($charset) == strtolower('Latin1')) || (strtolower($charset) == strtolower('Latin-1')))
1261 |         {
1262 |             if (is_object($debugObject)) {$debugObject->debugLog(2, 'replacing ' . $charset . ' with CP1252 as its a superset');}
1263 |             $charset = 'CP1252';
1264 |         }
1265 | 
1266 |         if (is_object($debugObject)) {$debugObject->debugLog(1, 'EXIT - ' . $charset);}
1267 | 
1268 |         return $this->_charset = $charset;
1269 |     }
1270 | 
1271 |     // read tag info
1272 |     protected function read_tag()
1273 |     {
1274 |         if ($this->char!=='<')
1275 |         {
1276 |             $this->root->_[HDOM_INFO_END] = $this->cursor;
1277 |             return false;
1278 |         }
1279 |         $begin_tag_pos = $this->pos;
1280 |         $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1281 | 
1282 |         // end tag
1283 |         if ($this->char==='/')
1284 |         {
1285 |             $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1286 |             // This represents the change in the simple_html_dom trunk from revision 180 to 181.
1287 |             // $this->skip($this->token_blank_t);
1288 |             $this->skip($this->token_blank);
1289 |             $tag = $this->copy_until_char('>');
1290 | 
1291 |             // skip attributes in end tag
1292 |             if (($pos = strpos($tag, ' '))!==false)
1293 |                 $tag = substr($tag, 0, $pos);
1294 | 
1295 |             $parent_lower = strtolower($this->parent->tag);
1296 |             $tag_lower = strtolower($tag);
1297 | 
1298 |             if ($parent_lower!==$tag_lower)
1299 |             {
1300 |                 if (isset($this->optional_closing_tags[$parent_lower]) && isset($this->block_tags[$tag_lower]))
1301 |                 {
1302 |                     $this->parent->_[HDOM_INFO_END] = 0;
1303 |                     $org_parent = $this->parent;
1304 | 
1305 |                     while (($this->parent->parent) && strtolower($this->parent->tag)!==$tag_lower)
1306 |                         $this->parent = $this->parent->parent;
1307 | 
1308 |                     if (strtolower($this->parent->tag)!==$tag_lower) {
1309 |                         $this->parent = $org_parent; // restore origonal parent
1310 |                         if ($this->parent->parent) $this->parent = $this->parent->parent;
1311 |                         $this->parent->_[HDOM_INFO_END] = $this->cursor;
1312 |                         return $this->as_text_node($tag);
1313 |                     }
1314 |                 }
1315 |                 else if (($this->parent->parent) && isset($this->block_tags[$tag_lower]))
1316 |                 {
1317 |                     $this->parent->_[HDOM_INFO_END] = 0;
1318 |                     $org_parent = $this->parent;
1319 | 
1320 |                     while (($this->parent->parent) && strtolower($this->parent->tag)!==$tag_lower)
1321 |                         $this->parent = $this->parent->parent;
1322 | 
1323 |                     if (strtolower($this->parent->tag)!==$tag_lower)
1324 |                     {
1325 |                         $this->parent = $org_parent; // restore origonal parent
1326 |                         $this->parent->_[HDOM_INFO_END] = $this->cursor;
1327 |                         return $this->as_text_node($tag);
1328 |                     }
1329 |                 }
1330 |                 else if (($this->parent->parent) && strtolower($this->parent->parent->tag)===$tag_lower)
1331 |                 {
1332 |                     $this->parent->_[HDOM_INFO_END] = 0;
1333 |                     $this->parent = $this->parent->parent;
1334 |                 }
1335 |                 else
1336 |                     return $this->as_text_node($tag);
1337 |             }
1338 | 
1339 |             $this->parent->_[HDOM_INFO_END] = $this->cursor;
1340 |             if ($this->parent->parent) $this->parent = $this->parent->parent;
1341 | 
1342 |             $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1343 |             return true;
1344 |         }
1345 | 
1346 |         $node = new simple_html_dom_node($this);
1347 |         $node->_[HDOM_INFO_BEGIN] = $this->cursor;
1348 |         ++$this->cursor;
1349 |         $tag = $this->copy_until($this->token_slash);
1350 |         $node->tag_start = $begin_tag_pos;
1351 | 
1352 |         // doctype, cdata & comments...
1353 |         if (isset($tag[0]) && $tag[0]==='!') {
1354 |             $node->_[HDOM_INFO_TEXT] = '<' . $tag . $this->copy_until_char('>');
1355 | 
1356 |             if (isset($tag[2]) && $tag[1]==='-' && $tag[2]==='-') {
1357 |                 $node->nodetype = HDOM_TYPE_COMMENT;
1358 |                 $node->tag = 'comment';
1359 |             } else {
1360 |                 $node->nodetype = HDOM_TYPE_UNKNOWN;
1361 |                 $node->tag = 'unknown';
1362 |             }
1363 |             if ($this->char==='>') $node->_[HDOM_INFO_TEXT].='>';
1364 |             $this->link_nodes($node, true);
1365 |             $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1366 |             return true;
1367 |         }
1368 | 
1369 |         // text
1370 |         if ($pos=strpos($tag, '<')!==false) {
1371 |             $tag = '<' . substr($tag, 0, -1);
1372 |             $node->_[HDOM_INFO_TEXT] = $tag;
1373 |             $this->link_nodes($node, false);
1374 |             $this->char = $this->doc[--$this->pos]; // prev
1375 |             return true;
1376 |         }
1377 | 
1378 |         if (!preg_match("/^[\w-:]+$/", $tag)) {
1379 |             $node->_[HDOM_INFO_TEXT] = '<' . $tag . $this->copy_until('<>');
1380 |             if ($this->char==='<') {
1381 |                 $this->link_nodes($node, false);
1382 |                 return true;
1383 |             }
1384 | 
1385 |             if ($this->char==='>') $node->_[HDOM_INFO_TEXT].='>';
1386 |             $this->link_nodes($node, false);
1387 |             $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1388 |             return true;
1389 |         }
1390 | 
1391 |         // begin tag
1392 |         $node->nodetype = HDOM_TYPE_ELEMENT;
1393 |         $tag_lower = strtolower($tag);
1394 |         $node->tag = ($this->lowercase) ? $tag_lower : $tag;
1395 | 
1396 |         // handle optional closing tags
1397 |         if (isset($this->optional_closing_tags[$tag_lower]) )
1398 |         {
1399 |             while (isset($this->optional_closing_tags[$tag_lower][strtolower($this->parent->tag)]))
1400 |             {
1401 |                 $this->parent->_[HDOM_INFO_END] = 0;
1402 |                 $this->parent = $this->parent->parent;
1403 |             }
1404 |             $node->parent = $this->parent;
1405 |         }
1406 | 
1407 |         $guard = 0; // prevent infinity loop
1408 |         $space = array($this->copy_skip($this->token_blank), '', '');
1409 | 
1410 |         // attributes
1411 |         do
1412 |         {
1413 |             if ($this->char!==null && $space[0]==='')
1414 |             {
1415 |                 break;
1416 |             }
1417 |             $name = $this->copy_until($this->token_equal);
1418 |             if ($guard===$this->pos)
1419 |             {
1420 |                 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1421 |                 continue;
1422 |             }
1423 |             $guard = $this->pos;
1424 | 
1425 |             // handle endless '<'
1426 |             if ($this->pos>=$this->size-1 && $this->char!=='>') {
1427 |                 $node->nodetype = HDOM_TYPE_TEXT;
1428 |                 $node->_[HDOM_INFO_END] = 0;
1429 |                 $node->_[HDOM_INFO_TEXT] = '<'.$tag . $space[0] . $name;
1430 |                 $node->tag = 'text';
1431 |                 $this->link_nodes($node, false);
1432 |                 return true;
1433 |             }
1434 | 
1435 |             // handle mismatch '<'
1436 |             if ($this->doc[$this->pos-1]=='<') {
1437 |                 $node->nodetype = HDOM_TYPE_TEXT;
1438 |                 $node->tag = 'text';
1439 |                 $node->attr = array();
1440 |                 $node->_[HDOM_INFO_END] = 0;
1441 |                 $node->_[HDOM_INFO_TEXT] = substr($this->doc, $begin_tag_pos, $this->pos-$begin_tag_pos-1);
1442 |                 $this->pos -= 2;
1443 |                 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1444 |                 $this->link_nodes($node, false);
1445 |                 return true;
1446 |             }
1447 | 
1448 |             if ($name!=='/' && $name!=='') {
1449 |                 $space[1] = $this->copy_skip($this->token_blank);
1450 |                 $name = $this->restore_noise($name);
1451 |                 if ($this->lowercase) $name = strtolower($name);
1452 |                 if ($this->char==='=') {
1453 |                     $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1454 |                     $this->parse_attr($node, $name, $space);
1455 |                 }
1456 |                 else {
1457 |                     //no value attr: nowrap, checked selected...
1458 |                     $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_NO;
1459 |                     $node->attr[$name] = true;
1460 |                     if ($this->char!='>') $this->char = $this->doc[--$this->pos]; // prev
1461 |                 }
1462 |                 $node->_[HDOM_INFO_SPACE][] = $space;
1463 |                 $space = array($this->copy_skip($this->token_blank), '', '');
1464 |             }
1465 |             else
1466 |                 break;
1467 |         } while ($this->char!=='>' && $this->char!=='/');
1468 | 
1469 |         $this->link_nodes($node, true);
1470 |         $node->_[HDOM_INFO_ENDSPACE] = $space[0];
1471 | 
1472 |         // check self closing
1473 |         if ($this->copy_until_char_escape('>')==='/')
1474 |         {
1475 |             $node->_[HDOM_INFO_ENDSPACE] .= '/';
1476 |             $node->_[HDOM_INFO_END] = 0;
1477 |         }
1478 |         else
1479 |         {
1480 |             // reset parent
1481 |             if (!isset($this->self_closing_tags[strtolower($node->tag)])) $this->parent = $node;
1482 |         }
1483 |         $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1484 | 
1485 |         // If it's a BR tag, we need to set it's text to the default text.
1486 |         // This way when we see it in plaintext, we can generate formatting that the user wants.
1487 |         // since a br tag never has sub nodes, this works well.
1488 |         if ($node->tag == "br")
1489 |         {
1490 |             $node->_[HDOM_INFO_INNER] = $this->default_br_text;
1491 |         }
1492 | 
1493 |         return true;
1494 |     }
1495 | 
1496 |     // parse attributes
1497 |     protected function parse_attr($node, $name, &$space)
1498 |     {
1499 |         // Per sourceforge: http://sourceforge.net/tracker/?func=detail&aid=3061408&group_id=218559&atid=1044037
1500 |         // If the attribute is already defined inside a tag, only pay atetntion to the first one as opposed to the last one.
1501 |         if (isset($node->attr[$name]))
1502 |         {
1503 |             return;
1504 |         }
1505 | 
1506 |         $space[2] = $this->copy_skip($this->token_blank);
1507 |         switch ($this->char) {
1508 |             case '"':
1509 |                 $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_DOUBLE;
1510 |                 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1511 |                 $node->attr[$name] = $this->restore_noise($this->copy_until_char_escape('"'));
1512 |                 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1513 |                 break;
1514 |             case '\'':
1515 |                 $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_SINGLE;
1516 |                 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1517 |                 $node->attr[$name] = $this->restore_noise($this->copy_until_char_escape('\''));
1518 |                 $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1519 |                 break;
1520 |             default:
1521 |                 $node->_[HDOM_INFO_QUOTE][] = HDOM_QUOTE_NO;
1522 |                 $node->attr[$name] = $this->restore_noise($this->copy_until($this->token_attr));
1523 |         }
1524 |         // PaperG: Attributes should not have \r or \n in them, that counts as html whitespace.
1525 |         $node->attr[$name] = str_replace("\r", "", $node->attr[$name]);
1526 |         $node->attr[$name] = str_replace("\n", "", $node->attr[$name]);
1527 |         // 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.
1528 |         if ($name == "class") {
1529 |             $node->attr[$name] = trim($node->attr[$name]);
1530 |         }
1531 |     }
1532 | 
1533 |     // link node's parent
1534 |     protected function link_nodes(&$node, $is_child)
1535 |     {
1536 |         $node->parent = $this->parent;
1537 |         $this->parent->nodes[] = $node;
1538 |         if ($is_child)
1539 |         {
1540 |             $this->parent->children[] = $node;
1541 |         }
1542 |     }
1543 | 
1544 |     // as a text node
1545 |     protected function as_text_node($tag)
1546 |     {
1547 |         $node = new simple_html_dom_node($this);
1548 |         ++$this->cursor;
1549 |         $node->_[HDOM_INFO_TEXT] = '</' . $tag . '>';
1550 |         $this->link_nodes($node, false);
1551 |         $this->char = (++$this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1552 |         return true;
1553 |     }
1554 | 
1555 |     protected function skip($chars)
1556 |     {
1557 |         $this->pos += strspn($this->doc, $chars, $this->pos);
1558 |         $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1559 |     }
1560 | 
1561 |     protected function copy_skip($chars)
1562 |     {
1563 |         $pos = $this->pos;
1564 |         $len = strspn($this->doc, $chars, $pos);
1565 |         $this->pos += $len;
1566 |         $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1567 |         if ($len===0) return '';
1568 |         return substr($this->doc, $pos, $len);
1569 |     }
1570 | 
1571 |     protected function copy_until($chars)
1572 |     {
1573 |         $pos = $this->pos;
1574 |         $len = strcspn($this->doc, $chars, $pos);
1575 |         $this->pos += $len;
1576 |         $this->char = ($this->pos<$this->size) ? $this->doc[$this->pos] : null; // next
1577 |         return substr($this->doc, $pos, $len);
1578 |     }
1579 | 
1580 |     protected function copy_until_char($char)
1581 |     {
1582 |         if ($this->char===null) return '';
1583 | 
1584 |         if (($pos = strpos($this->doc, $char, $this->pos))===false) {
1585 |             $ret = substr($this->doc, $this->pos, $this->size-$this->pos);
1586 |             $this->char = null;
1587 |             $this->pos = $this->size;
1588 |             return $ret;
1589 |         }
1590 | 
1591 |         if ($pos===$this->pos) return '';
1592 |         $pos_old = $this->pos;
1593 |         $this->char = $this->doc[$pos];
1594 |         $this->pos = $pos;
1595 |         return substr($this->doc, $pos_old, $pos-$pos_old);
1596 |     }
1597 | 
1598 |     protected function copy_until_char_escape($char)
1599 |     {
1600 |         if ($this->char===null) return '';
1601 | 
1602 |         $start = $this->pos;
1603 |         while (1)
1604 |         {
1605 |             if (($pos = strpos($this->doc, $char, $start))===false)
1606 |             {
1607 |                 $ret = substr($this->doc, $this->pos, $this->size-$this->pos);
1608 |                 $this->char = null;
1609 |                 $this->pos = $this->size;
1610 |                 return $ret;
1611 |             }
1612 | 
1613 |             if ($pos===$this->pos) return '';
1614 | 
1615 |             if ($this->doc[$pos-1]==='\\') {
1616 |                 $start = $pos+1;
1617 |                 continue;
1618 |             }
1619 | 
1620 |             $pos_old = $this->pos;
1621 |             $this->char = $this->doc[$pos];
1622 |             $this->pos = $pos;
1623 |             return substr($this->doc, $pos_old, $pos-$pos_old);
1624 |         }
1625 |     }
1626 | 
1627 |     // remove noise from html content
1628 |     // save the noise in the $this->noise array.
1629 |     protected function remove_noise($pattern, $remove_tag=false)
1630 |     {
1631 |         global $debugObject;
1632 |         if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
1633 | 
1634 |         $count = preg_match_all($pattern, $this->doc, $matches, PREG_SET_ORDER|PREG_OFFSET_CAPTURE);
1635 | 
1636 |         for ($i=$count-1; $i>-1; --$i)
1637 |         {
1638 |             $key = '___noise___'.sprintf('% 5d', count($this->noise)+1000);
1639 |             if (is_object($debugObject)) { $debugObject->debugLog(2, 'key is: ' . $key); }
1640 |             $idx = ($remove_tag) ? 0 : 1;
1641 |             $this->noise[$key] = $matches[$i][$idx][0];
1642 |             $this->doc = substr_replace($this->doc, $key, $matches[$i][$idx][1], strlen($matches[$i][$idx][0]));
1643 |         }
1644 | 
1645 |         // reset the length of content
1646 |         $this->size = strlen($this->doc);
1647 |         if ($this->size>0)
1648 |         {
1649 |             $this->char = $this->doc[0];
1650 |         }
1651 |     }
1652 | 
1653 |     // restore noise to html content
1654 |     function restore_noise($text)
1655 |     {
1656 |         global $debugObject;
1657 |         if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
1658 | 
1659 |         while (($pos=strpos($text, '___noise___'))!==false)
1660 |         {
1661 |             // Sometimes there is a broken piece of markup, and we don't GET the pos+11 etc... token which indicates a problem outside of us...
1662 |             if (strlen($text) > $pos+15)
1663 |             {
1664 |                 $key = '___noise___'.$text[$pos+11].$text[$pos+12].$text[$pos+13].$text[$pos+14].$text[$pos+15];
1665 |                 if (is_object($debugObject)) { $debugObject->debugLog(2, 'located key of: ' . $key); }
1666 | 
1667 |                 if (isset($this->noise[$key]))
1668 |                 {
1669 |                     $text = substr($text, 0, $pos).$this->noise[$key].substr($text, $pos+16);
1670 |                 }
1671 |                 else
1672 |                 {
1673 |                     // do this to prevent an infinite loop.
1674 |                     $text = substr($text, 0, $pos).'UNDEFINED NOISE FOR KEY: '.$key . substr($text, $pos+16);
1675 |                 }
1676 |             }
1677 |             else
1678 |             {
1679 |                 // There is no valid key being given back to us... We must get rid of the ___noise___ or we will have a problem.
1680 |                 $text = substr($text, 0, $pos).'NO NUMERIC NOISE KEY' . substr($text, $pos+11);
1681 |             }
1682 |         }
1683 |         return $text;
1684 |     }
1685 | 
1686 |     // Sometimes we NEED one of the noise elements.
1687 |     function search_noise($text)
1688 |     {
1689 |         global $debugObject;
1690 |         if (is_object($debugObject)) { $debugObject->debugLogEntry(1); }
1691 | 
1692 |         foreach($this->noise as $noiseElement)
1693 |         {
1694 |             if (strpos($noiseElement, $text)!==false)
1695 |             {
1696 |                 return $noiseElement;
1697 |             }
1698 |         }
1699 |     }
1700 |     function __toString()
1701 |     {
1702 |         return $this->root->innertext();
1703 |     }
1704 | 
1705 |     function __get($name)
1706 |     {
1707 |         switch ($name)
1708 |         {
1709 |             case 'outertext':
1710 |                 return $this->root->innertext();
1711 |             case 'innertext':
1712 |                 return $this->root->innertext();
1713 |             case 'plaintext':
1714 |                 return $this->root->text();
1715 |             case 'charset':
1716 |                 return $this->_charset;
1717 |             case 'target_charset':
1718 |                 return $this->_target_charset;
1719 |         }
1720 |     }
1721 | 
1722 |     // camel naming conventions
1723 |     function childNodes($idx=-1) {return $this->root->childNodes($idx);}
1724 |     function firstChild() {return $this->root->first_child();}
1725 |     function lastChild() {return $this->root->last_child();}
1726 |     function createElement($name, $value=null) {return @str_get_html("<$name>$value</$name>")->first_child();}
1727 |     function createTextNode($value) {return @end(str_get_html($value)->nodes);}
1728 |     function getElementById($id) {return $this->find("#$id", 0);}
1729 |     function getElementsById($id, $idx=null) {return $this->find("#$id", $idx);}
1730 |     function getElementByTagName($name) {return $this->find($name, 0);}
1731 |     function getElementsByTagName($name, $idx=-1) {return $this->find($name, $idx);}
1732 |     function loadFile() {$args = func_get_args();$this->load_file($args);}
1733 | }
1734 | 


--------------------------------------------------------------------------------
/composer.json:
--------------------------------------------------------------------------------
 1 | {
 2 |     "name": "sunra/php-simple-html-dom-parser",
 3 |     "type": "library",
 4 |     "description": "Composer adaptation of: A HTML DOM parser written in PHP5+ let you manipulate HTML in a very easy way! Require PHP 5+. Supports invalid HTML. Find tags on an HTML page with selectors just like jQuery. Extract contents from HTML in a single line.",    
 5 |     "keywords": ["html", "dom", "parser"],
 6 |     "homepage": "https://github.com/sunra/php-simple-html-dom-parser",
 7 |     "license": "MIT",
 8 |     "authors": [
 9 |         {   
10 |             "name": "S.C. Chen",
11 |             "homepage": "http://sourceforge.net/projects/simplehtmldom/"
12 |         },
13 |         {
14 |             "name": "Sunra",
15 |             "email": "sunra@yandex.ru",
16 |             "homepage": "https://github.com/sunra"
17 |         }   
18 |     ],
19 |     "require": {
20 |         "php": ">=5.3.2",
21 |         "ext-mbstring": "*"
22 |     },
23 |     "autoload": {
24 |         "psr-0": { "Sunra\\PhpSimple\\HtmlDomParser": "Src/" }
25 |     }    
26 | }
27 | 


--------------------------------------------------------------------------------