├── LICENSE.md ├── README.md ├── simple_html_dom.php └── sitemap-generator.php /LICENSE.md: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2017 Hemn Chawroka 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # PHP XML Sitemap Generator 2 | 3 | This is a simple and small PHP script that I wrote quickly for myself to create a XML sitemap of my page for Google and other search engines. Maybe others can use the script too. 4 | 5 | Sitemap format: [http://www.sitemaps.org/protocol.html](http://www.sitemaps.org/protocol.html) 6 | 7 | ##Features 8 | - Actually crawls webpages like Google would 9 | - Generates seperate XML file which gets updated every time the script gets executed (Runnable via CRON) 10 | - Awesome for SEO 11 | - Crawls faster than online services 12 | - Adaptable 13 | 14 | ## Usage 15 | Usage is pretty strait forward: 16 | - Configure the crawler by modifying the `sitemap-generator.php` file 17 | - Select URL to crawl 18 | - Select the file to which the sitemap will be saved 19 | - Select accepted extensions ("/" is manditory for proper functionality) 20 | - Select change frequency (always, daily, weekly, monthly, never, etc...) 21 | - Choose priority (It is all relative so it may as well be 1) 22 | - Generate sitemap 23 | - Either send a GET request to this script or simply point your browser 24 | - A sitemap will be generated and displayed 25 | - Submit sitemap.xml to Google 26 | - Setup a CRON Job to send web requests to this script every so often, this will keep the sitemap.xml file up to date 27 | 28 | The script can be started as CLI script or as Website. CLI is the prefered way to start this script. 29 | 30 | CLI scripts are started from the command line, can be used with CRON and so on. You start it with the php program. 31 | 32 | CLI command to create the XML file: `php sitemap-generator.php` 33 | 34 | To start the program with your Webserver as Website change in the script the line 22 from 35 | ```php 36 | define ('CLI', true); 37 | ``` 38 | to 39 | ```php 40 | define ('CLI', false); 41 | ``` 42 | 43 | 44 | ## sitemap.xml 45 | Add the XML file to your `/robots.txt`. 46 | 47 | Example line for the robots.txt: 48 | 49 | ``` 50 | Sitemap: http://www.iprodev.com/sitemap.xml 51 | ``` 52 | 53 | 54 | ## Credits 55 | 56 | PHP XML Sitemap Generator was created by [Hemn Chawroka](http://iprodev.com) from [iProDev](http://iprodev.com). Released under the MIT license. 57 | 58 | Included scripts: 59 | 60 | - [PHP Simple HTML DOM Parser](http://simplehtmldom.sourceforge.net/) - A HTML DOM parser written in PHP5+ let you manipulate HTML in a very easy way!. 61 | -------------------------------------------------------------------------------- /simple_html_dom.php: -------------------------------------------------------------------------------- 1 | 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 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 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 .= '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('', '', $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 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("''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