API tools faq
paste
Guest User

Untitled

a guest
Jun 20th, 2013
114
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
PHP 10.64 KB | None | 0 0
raw download clone embed print report
  1. <?php
  2. /**
  3. * Website Layout Checker
  4. *
  5. * @package  
  6. * @author Marais Rossouw ([email protected])
  7. * @copyright Vuly
  8. * @version 2013
  9. * @access public
  10. */
  11.  
  12. require_once '../setup.php';
  13. ini_set('max_execution_time', 6000);
  14.  
  15. class layout {
  16.  
  17.     private $_LAYOUT, $_URL, $_DOC, $_LAYOUT_ARRAY, $_SAVE_TO_JSON, $_SAVE_TO_HTML, $_HTML_BODY;
  18.  
  19.     private $_CONSOLE = array();
  20.  
  21.     public function __construct($url) {
  22.  
  23.         // Get's the contents of the page specified.
  24.         try {
  25.             $client = new Zend_Http_Client;
  26.             $client->setUri($url);
  27.             $client->setConfig(array('strictredirects' => true, 'maxredirects' => 10, 'timeout' => 8));
  28.             $response = $client->request();
  29.  
  30.             $this->_LAYOUT = $response->getBody();
  31.             $this->_URL = $url;
  32.         } catch (Exception $e) {
  33.             $this->consoleLog($e);
  34.         }
  35.  
  36.         // Creates a DOMDocument
  37.         try {
  38.             $this->_INIT();
  39.         } catch (Exception $e) {
  40.             $this->consoleLog($e);
  41.         }
  42.  
  43.         // Save the files
  44.         try {
  45.             file_put_contents($this->_SAVE_TO_JSON, json_encode($this->_LAYOUT_ARRAY));
  46.             file_put_contents($this->_SAVE_TO_HTML, $this->_LAYOUT);
  47.  
  48.             $this->consoleLog("The JSON file was saved to: " . $this->_SAVE_TO_JSON);
  49.             $this->consoleLog("The HTML file was saved to: " . $this->_SAVE_TO_HTML);
  50.         } catch (Exception $e) {
  51.             $this->consoleLog($e);
  52.         }
  53.     }
  54.  
  55.     private function _INIT() {
  56.         $doc = new DOMDocument();
  57.  
  58.         libxml_use_internal_errors(true);
  59.  
  60.         $doc->loadHTML($this->_LAYOUT);
  61.         $this->_DOC = new DOMXpath($doc);
  62.  
  63.         $this->consoleLog("DOMDocument created");
  64.  
  65.         $parser = new htmlParser($this->_LAYOUT);
  66.         $this->_LAYOUT_ARRAY = $parser->toArray();
  67.  
  68.  
  69.         //var_dump($this->_LAYOUT_ARRAY[0]['childNodes']);exit;
  70.  
  71.         $this->consoleLog("Dom array created");
  72.  
  73.         $this->consoleLog("There are " . count($this->_LAYOUT_ARRAY, COUNT_RECURSIVE) . " elements in the dom array");
  74.  
  75.         $this->_FILE_NAME = "VULY_LAYOUT_CHECKER-" . sha1(htmlspecialchars(trim($this->_URL)) . date("Ymd") . rand(99, 9999));
  76.         $this->_FILE_PATH = "layout_checker\\"/*sys_get_temp_dir() . "\\"*/;
  77.  
  78.         $this->_SAVE_TO_JSON = $this->_FILE_PATH . $this->_FILE_NAME . ".txt";
  79.         $this->_SAVE_TO_HTML = $this->_FILE_PATH . $this->_FILE_NAME . ".html";
  80.  
  81.  
  82.         libxml_use_internal_errors(false);
  83.     }
  84.  
  85.     public function toString() {
  86.         return $this->_LAYOUT;
  87.     }
  88.  
  89.     public function getBody() {
  90.         $this->recurse($this->_LAYOUT_ARRAY);
  91.         return $this->_HTML_BODY;
  92.     }
  93.  
  94.     private function recurse($file) {
  95.         if ($this->_HTML_BODY != false) { return; }
  96.  
  97.         for ($i = 0; $i < count($file); $i++) {
  98.             if ($file[$i]['childNodes']) {
  99.                 if ($file[$i]['tag'] == "body") {
  100.                     $this->_HTML_BODY = $file[$i]['innerHTML'];
  101.                     return;
  102.                 } else {
  103.                     $this->recurse($file[$i]['childNodes']);
  104.                 }
  105.             }
  106.         }
  107.     }
  108.  
  109.     public function getJSON_FILE() {
  110.         return file_get_contents($this->_SAVE_TO_JSON);
  111.     }
  112.  
  113.     public function get_SAVE_TO_HTML() {
  114.         return $this->_SAVE_TO_HTML;
  115.     }
  116.  
  117.  
  118.     public function consoleLog($string) {
  119.         $this->_CONSOLE[] = $string;
  120.     }
  121.  
  122.     public function renderConsole() {
  123.         $return = "";
  124.  
  125.         $_PAD_SIZE = strlen(count($this->_CONSOLE)) + 2;
  126.  
  127.         foreach ($this->_CONSOLE as $key => $value) {
  128.             $return .= str_pad($key . ":", $_PAD_SIZE) . $value . "\n";
  129.         }
  130.         return $return;
  131.     }
  132.  
  133. }
  134.  
  135. class htmlParser {
  136.    
  137.     //your very own separator
  138.     //do not enter characters such as < or >
  139.     private $separator = '~';
  140.     //the tags that don't have any innerHTML in them
  141.     //feel free to add some if I missed any
  142.     private $singleTags = 'meta|img|hr|br|link|!--|!DOCTYPE|input';
  143.    
  144.     //-- Don't edit below this --
  145.    
  146.     private $html,$level;
  147.     public $levelArray;
  148.    
  149.     function __construct($html='') {
  150.         $this->html=$this->removeWhiteSpace($html);
  151.         $this->level=-1;
  152.         $this->levelArray=array();
  153.     }
  154.     function __destruct() {
  155.         //nothing yet;
  156.     }
  157.     private function getElement($value) {
  158.         $ar = explode($this->separator,$value);
  159.         $ar = explode('-',$ar[1]);
  160.         return $this->levelArray[$ar[0]][$ar[1]];
  161.     }
  162.     private function parseToHTML($str,$level) {
  163.         $ar=$this->getArrayOfReplacements($str);
  164.         foreach ($ar as $item) {
  165.             $elem = $this->getElement($item);
  166.             $str=str_replace($item,($level==0?$elem['htmlText']:'<'.$elem['tag'].$elem['attr'].'>'.$elem['htmlText'].'</'.$elem['tag'].'>'),$str);
  167.         }
  168.         return $str;
  169.     }
  170.     private function replaceSingleTags() {
  171.         //tags like img, input etc
  172.         $result=preg_match_all('/<('.$this->singleTags.')(.[^><]*)?>/is', $this->html, $m);
  173.         if ($result>0) {
  174.             foreach ($m[0] as $id => $value) {
  175.                 $this->html = str_replace($value,'',$this->html);
  176.             }
  177.         }
  178.     }
  179.     private function replaceSimpleTags() {
  180.         //tags that only have text in them (no other content)
  181.         $result=preg_match_all('/<(.[^\s]*)(.[^><]*)?>(.[^<]*)?<\/\1>/is', $this->html, $m);
  182.         if ($result>0) {
  183.             $this->level++;
  184.             $oneLevel=array();
  185.             foreach ($m[0] as $id => $value) {
  186.                 if ($this->level==0) $htmlText=$value;
  187.                 else $htmlText=$this->parseToHTML($m[3][$id],$this->level-1);
  188.                
  189.                 $oneLevel []= array('str' => $value, 'rep' => $this->separator.$this->level.'-'.$id.$this->separator, 'tag' => $m[1][$id], 'level' => $this->level, 'text' => $m[3][$id], 'attr' => $m[2][$id] , 'htmlText' => $htmlText);
  190.                
  191.                 $this->html = str_replace($value,$this->separator.$this->level.'-'.$id.$this->separator,$this->html);
  192.             }
  193.             $this->levelArray [$this->level] = $oneLevel;
  194.         }
  195.     }
  196.     private function replaceRemainingTags() {
  197.         //tags that remain after everything
  198.         $result=preg_match_all('/<(.[^\s]*)(.[^><]*)?>(.*)?<\/\1>/is', $this->html, $m);
  199.         if ($result>0) {
  200.             $this->level++;
  201.             $oneLevel=array();
  202.             foreach ($m[0] as $id => $value) {
  203.                 if ($this->level==0) $htmlText=$m[3][$id];
  204.                 else $htmlText=$this->parseToHTML($m[3][$id],$this->level-1);
  205.                
  206.                 $oneLevel []= array('str' => $value, 'rep' => $this->separator.$this->level.'-'.$id.$this->separator, 'tag' => $m[1][$id], 'level' => $this->level, 'text' => $m[3][$id], 'attr' => $m[2][$id] , 'htmlText' => $htmlText);
  207.                
  208.                 $this->html = str_replace($value,$this->separator.$this->level.'-'.$id.$this->separator,$this->html);
  209.             }
  210.             $this->levelArray [$this->level] = $oneLevel;
  211.         }
  212.     }
  213.     private function existSimpleTags() {
  214.         $result=preg_match('/<(.[^\s]*)(.[^><]*)?>(.[^<]*)?<\/\1>/is', $this->html);
  215.         return $result>0;
  216.     }
  217.     private function existSingleTags() {
  218.         $result=preg_match('/<('.$this->singleTags.')(.[^><]*)?>/is', $this->html);
  219.         return $result>0;
  220.     }
  221.     private function removeWhiteSpace ($string) {
  222.         $string = str_replace(array("\n","\r",'&nbsp;',"\t"),'',$string);
  223.         return preg_replace('|  +|', ' ', $string);
  224.     }
  225.     public function toArray($html='') {
  226.        
  227.         //first part: coding
  228.         if ($html!='') {
  229.             $this->html = $this->removeWhiteSpace($html);
  230.         }
  231.         while ($this->existSimpleTags() || $this->existSingleTags()) {
  232.             $this->replaceSingleTags();
  233.             $this->replaceSimpleTags();
  234.         }
  235.         $this->replaceRemainingTags();
  236.        
  237.         //now decoding
  238.         $ar=$this->getArray($this->html);
  239.        
  240.         return $ar;
  241.     }
  242.     private function getArrayOfReplacements($str) {
  243.         $final=array();
  244.         $ar=explode($this->separator,$str);
  245.         for ($i=0;$i<(count($ar)-1)/2;$i++) {
  246.             $final []= $this->separator.$ar[$i*2+1].$this->separator;
  247.         }
  248.         return $final;
  249.     }
  250.     private function startsWithText($str) {
  251.         $first=substr(trim(str_replace(array("\n","\r"),'',$str)),0,1);
  252.         if ($first=='<' || $first=='>') return false;
  253.         return true;
  254.     }
  255.     private function strInArray($array,$str) {
  256.         foreach ($array as $item) {
  257.             if (strpos($str,$item)!==false)
  258.                 return true;
  259.         }
  260.         return false;
  261.     }
  262.     private function getArray($html, $father='') {
  263.         $final=array();
  264.         if (strpos($html,$this->separator)!==false) {
  265.             $r=$this->getArrayOfReplacements($html);
  266.             foreach ($r as $i) {
  267.                
  268.                 $ar = explode($this->separator,$i);
  269.                 $ar = explode('-',$ar[1]);
  270.                 $elem = $this->levelArray[$ar[0]][$ar[1]];
  271.                 $this->levelArray[$ar[0]][$ar[1]]['father'] = $father;
  272.                
  273.                 $final []= array( 'tag' => $elem['tag'], 'innerHTML' => $elem['htmlText'], 'repl' => $elem['rep'],'stratr' => $elem['attr'], 'level' => $elem['level'], 'father' => $father, 'childNodes' => $this->getArray($elem['text'],$i));
  274.             }
  275.         }
  276.         return $final;
  277.     }
  278.     public function loadNode($rep) {
  279.         $elem = $this->getElement($rep);
  280.         return array( 'tag' => $elem['tag'], 'innerHTML' => $elem['htmlText'], 'repl' => $elem['rep'],'stratr' => $elem['attr'], 'level' => $elem['level'], 'father' => $elem['father']);
  281.     }
  282. }
  283.  
  284. if (isset($_REQUEST['layout'])) {
  285.     $layout = new layout($_REQUEST['layout']);
  286.     $console = $layout->renderConsole();
  287.     $json_file = $layout->getJSON_FILE();
  288.     $toString = $layout->toString();
  289.     $getBody = "http://" . $_SERVER['SERVER_NAME']."/etramp/scripts/" . $layout->get_SAVE_TO_HTML();
  290. } else {
  291.     $console = "";
  292.     $json_file = "";
  293.     $toString = "";
  294.     $getBody = "";
  295. }
  296.  
  297. ?>
  298.  
  299. <html>
  300. <head>
  301.     <title>Vuly Layout Checker</title>
  302.     <style type="text/css">
  303.     html {
  304.         height: 100%;
  305.         margin:0;padding:0;
  306.     }
  307.     body {
  308.         background: #728eaa;
  309.         background: -moz-linear-gradient(top, #25303C 0%, #728EAA 100%);
  310.         background: -webkit-gradient(linear, left top, left bottom, color-stop(0%, #25303C), color-stop(100%, #728EAA));
  311.         font-family: sans-serif;
  312.     }
  313.     input, select {
  314.         padding:10px;
  315.     }
  316.     select, input[type='submit'] {
  317.         cursor:pointer;
  318.     }
  319.     label {
  320.         color: #fff;
  321.         padding-right: 10px;
  322.     }
  323.     form {
  324.         margin: 50px auto 0 auto;
  325.         width: 684px;
  326.     }
  327.     .text1 {
  328.         width:49%; height:220px; resize: none; position:fixed; top:150px;
  329.     }
  330.     .text2 {
  331.         width:49%; resize: none; position:fixed; top: 380px; bottom:10px; height: 58%;
  332.     }
  333.     </style>
  334.     <script src="//ajax.googleapis.com/ajax/libs/jquery/1.10.1/jquery.min.js"></script>
  335.     <script type="text/javascript">
  336.     var file;
  337.     $( document ).ready(function() {
  338.     file = <?php echo $json_file; ?>;
  339.         recurse(file);
  340.     });
  341.     function recurse(file) {
  342.         console.log(file[i].tag);
  343.         for (var i = 0; i < file.length; i++) {
  344.             if (file[i].childNodes) {
  345.                 if (file[i].tag == "body") {
  346.                     console.log($(file[i].innerHeml, $('#NEW_LAYOUT').contents()));
  347.                     alert(file[i].tag);
  348.                 } else {
  349.                     recurse(file[i].childNodes);
  350.                 }      
  351.             }
  352.         }
  353.     }
  354.     </script>
  355. </head>
  356. <body>
  357.  
  358.     <textarea class="text1" style="left:10px;"><?php echo $console; ?></textarea>
  359.     <textarea class="text1" style="right:10px;"><?php echo $json_file; ?></textarea>
  360.  
  361.     <form>
  362.         <label for="layout">Website URL:</label>
  363.         <input type="text" name="layout" id="layout" style="width: 500px" value="<?php echo (isset($_REQUEST['layout'])) ? $_REQUEST['layout'] : "http://"; ?>">
  364.         <input type="submit">
  365.     </form>
  366.  
  367.     <textarea class="text2" style="left:10px;"><?php echo $toString; ?></textarea>
  368.     <iframe id="NEW_LAYOUT" class="text2" style="right:10px;" src="<?php echo $getBody; ?>"></iframe>
  369.  
  370. </body>
  371. </html>
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!