readityourself.php

   1 <?php
   2
   3 define("VERSION", "0.0.3");
   4
   5 header('Content-type:text/html; charset=utf-8');
   6 // Set locale to French
   7 setlocale(LC_ALL, 'fr_FR');
   8
   9 // set timezone to Europe/Paris
  10 date_default_timezone_set('Europe/Paris');
  11
  12 // set charset to utf-8 important since all pages will be transform to utf-8
  13 header('Content-Type: text/html;charset=utf-8');
  14
  15 // get readability library
  16 require_once dirname(__FILE__).'/inc/Readability.php';
  17
  18 // get Encoding library.
  19 require_once dirname(__FILE__).'/inc/Encoding.php';
  20
  21 // appel de la libraire RainTPL.
  22 require_once dirname(__FILE__).'/inc/rain.tpl.class.php';
  23
  24 // FUNCTIONS BEGIN
  25
  26
  27
  28 function url(){
  29   $protocol = "http";
  30   if(isset($_SERVER['HTTPS']))
  31     if($_SERVER['HTTPS'] != "off")
  32       $protocol = "https";
  33
  34   return $protocol . "://" . $_SERVER['HTTP_HOST'] . $_SERVER['REQUEST_URI'];
  35 }
  36
  37 function generate_page($url,$title,$content) {
  38         raintpl::$tpl_dir = './tpl/'; // template directory
  39         raintpl::$cache_dir = "./cache/"; // cache directory
  40         raintpl::$base_url = url(); // base URL of blog
  41         raintpl::configure( 'path_replace', false );
  42         raintpl::configure('debug', false);
  43
  44         $tpl = new raintpl(); //include Rain TPL
  45
  46         $tpl->assign( "url", $url);
  47         $tpl->assign( "title", $title);
  48         $tpl->assign( "content", $content);
  49
  50         $tpl->assign( "version", VERSION);
  51
  52         $tpl->draw( "index"); // draw the template
  53 }
  54
  55 // function define to retrieve url content
  56 function get_external_file($url, $timeout) {
  57         // spoofing FireFox 18.0
  58         $useragent="Mozilla/5.0 (Windows NT 5.1; rv:18.0) Gecko/20100101 Firefox/18.0";
  59
  60         if  (in_array  ('curl', get_loaded_extensions())) {
  61                 // Fetch feed from URL
  62                 $curl = curl_init();
  63                 curl_setopt($curl, CURLOPT_URL, $url);
  64                 curl_setopt($curl, CURLOPT_TIMEOUT, $timeout);
  65                 curl_setopt($curl, CURLOPT_FOLLOWLOCATION, true);
  66                 curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
  67                 curl_setopt($curl, CURLOPT_HEADER, false);
  68
  69                 // FeedBurner requires a proper USER-AGENT...
  70                 curl_setopt($curl, CURL_HTTP_VERSION_1_1, true);
  71                 curl_setopt($curl, CURLOPT_ENCODING, "gzip, deflate");
  72                 curl_setopt($curl, CURLOPT_USERAGENT, $useragent);
  73
  74                 $data = curl_exec($curl);
  75
  76                 $httpcode = curl_getinfo($curl, CURLINFO_HTTP_CODE);
  77
  78                 $httpcodeOK = isset($httpcode) and ($httpcode == 200 or $httpcode == 301);
  79
  80                 curl_close($curl);
  81         } else {
  82
  83                 // create http context and add timeout and user-agent
  84                 $context = stream_context_create(array('http'=>array('timeout' => $timeout, // Timeout : time until we stop waiting for the response.
  85                                                                                                                                                                         'header'=> "User-Agent: ".$useragent, // spoot Mozilla Firefox
  86                                                                                                                                                                         'follow_location' => true
  87                                                                                                                 )));
  88
  89                 // only download page lesser than 4MB
  90                 $data = @file_get_contents($url, false, $context, -1, 4000000); // We download at most 4 MB from source.
  91                 //      echo "<pre>http_response_header : ".print_r($http_response_header);
  92
  93                 if(isset($http_response_header) and isset($http_response_header[0])) {
  94                         $httpcodeOK = isset($http_response_header) and isset($http_response_header[0]) and ((strpos($http_response_header[0], '200 OK') !== FALSE) or (strpos($http_response_header[0], '301 Moved Permanently') !== FALSE));
  95                 }
  96         }
  97
  98         // if response is not empty and response is OK
  99         if (isset($data) and isset($httpcodeOK) and $httpcodeOK ) {
 100
 101                 // take charset of page and get it
 102                 preg_match('#<meta .*charset=.*>#Usi', $data, $meta);
 103
 104                 // if meta tag is found
 105                 if (!empty($meta[0])) {
 106                         // retrieve encoding in $enc
 107                         preg_match('#charset="?(.*)"#si', $meta[0], $enc);
 108
 109                         // if charset is found set it otherwise, set it to utf-8
 110                         $html_charset = (!empty($enc[1])) ? strtolower($enc[1]) : 'utf-8';
 111
 112                 } else {
 113                         $html_charset = 'utf-8';
 114                         $enc[1] = '';
 115                 }
 116
 117                 // replace charset of url to charset of page
 118                 $data = str_replace('charset='.$enc[1], 'charset='.$html_charset, $data);
 119
 120                 return $data;
 121         }
 122         else {
 123                 return FALSE;
 124         }
 125 }
 126
 127 function rel2abs($rel, $base)
 128 {
 129     /* return if already absolute URL */
 130     if (parse_url($rel, PHP_URL_SCHEME) != '') return $rel;
 131
 132     /* queries and anchors */
 133     if ($rel[0]=='#' || $rel[0]=='?') return $base.$rel;
 134
 135     /* parse base URL and convert to local variables:
 136        $scheme, $host, $path */
 137     extract(parse_url($base));
 138
 139     /* remove non-directory element from path */
 140     $path = preg_replace('#/[^/]*$#', '', $path);
 141
 142     /* destroy path if relative url points to root */
 143     if ($rel[0] == '/') $path = '';
 144
 145     /* dirty absolute URL */
 146     $abs = "$host$path/$rel";
 147
 148     /* replace '//' or '/./' or '/foo/../' with '/' */
 149     $re = array('#(/\.?/)#', '#/(?!\.\.)[^/]+/\.\./#');
 150     for($n=1; $n>0; $abs=preg_replace($re, '/', $abs, -1, $n)) {}
 151
 152     /* absolute URL is ready! */
 153     return $scheme.'://'.$abs;
 154 }
 155
 156 // $str=preg_replace('#(href|src)="([^:"]*)("|(?:(?:%20|\s|\+)[^"]*"))#','$1="http://wintermute.com.au/$2$3',$str);
 157
 158 function absolutes_links($data, $base) {
 159         // cherche les balises 'a' qui contiennent un href
 160         $matches = array();
 161         preg_match_all('#(href|src)="([^:"]*)("|(?:(?:%20|\s|\+)[^"]*"))#Si', $data, $matches, PREG_SET_ORDER);
 162
 163         // ne conserve que les liens ne commençant pas par un protocole « protocole:// » ni par une ancre « # »
 164         foreach($matches as $i => $link) {
 165                 $link[1] = trim($link[1]);
 166
 167                 if (!preg_match('#^(([a-z]+://)|(\#))#', $link[1]) ) {
 168
 169                         $absolutePath=rel2abs($link[2],$base);
 170
 171                         $data = str_replace($matches[$i][2], $absolutePath, $data);
 172                 }
 173
 174         }
 175         return $data;
 176 }
 177
 178
 179 // FUNCTIONS END
 180
 181 // EXUCUTION CODE
 182
 183
 184 if(isset($_GET['url']) && $_GET['url'] != null && trim($_GET['url']) != "") {
 185         // get url link
 186         if(strlen(trim($_GET['url'])) > 2048) {
 187                 echo "Error URL is too large !!";
 188         } else {
 189                 $url = trim($_GET['url']);
 190
 191                 // decode it
 192                 $url = html_entity_decode($url);
 193
 194                 // if url use https protocol change it to http
 195                 if (!preg_match('!^https?://!i', $url)) $url = 'http://'.$url;
 196
 197                 // convert page to utf-8
 198                 $html = Encoding::toUTF8(get_external_file($url,15));
 199
 200                 if(isset($html) and strlen($html) > 0) {
 201
 202                         // send result to readability library
 203                         $r = new Readability($html, $url);
 204
 205                         if($r->init()) {
 206                                 generate_page($url,$r->articleTitle->innerHTML,$r->articleContent->innerHTML);
 207                         } else {
 208                                 // return data into an iframe
 209                                 echo "<iframe id='readabilityframe'>".$html."</iframe>";
 210                         }
 211                 } else {
 212                         echo "Error unable to get link : ".$url;
 213                 }
 214         }
 215 }
 216 ?>