libs/URLReader.class.php

<?

require_once "config.php";

class URLReader{
   var $_handler;
   var $_url_info;   // Used when relative paths are given in the links
  
   function URLReader($url){
     $start = strpos($url,"//") + strlen("//");
     $end = strpos($url,"/",$start);
     $host = substr($url,$start,$end-$start);
     $path = substr($url,$end);
     // Use sockets otherwise cgi fail to open
     if ($this->_handler = @fsockopen($host, 80, $errno, $errstr, 5)) {
       fputs($this->_handler, "GET $path HTTP/1.0\r\n");
       fputs($this->_handler, "Host: $host\r\n");
       fputs($this->_handler, "User-Agent: {$_SERVER['HTTP_USER_AGENT']}\r\n");
       fputs($this->_handler, "\r\n");
     }else
       die("Unable to connect: $errno :: $errstr");
     $end = strrpos($url,"/") + 1;      // Find last forward slash
     $end = strlen($url) - $end;        // Get actual number of chars
     $end *= -1;                        // Search from the end
     $this->_url_info = substr($url,0,$end);
   }

   function getLinksAfterAName($aname){
     $str = "";

     while(!feof($this->_handler)){
       $buff = fgets($this->_handler,4096);
       if (stristr($buff,"<a name"))      // Found <a name=""> element
         if (stristr($buff,$aname)){      // Found the one we are looking for
           while(!feof($this->_handler)){
             $buff = fgets($this->_handler,4096);
             if (stristr($buff,"<a href")){     // return all links
               if (!stristr($buff,"</a>"))
                 $buff = str_replace("\n","",$buff);
               $str .= $buff;
               while(!stristr($buff,"</a>")){   // not always on one line
                 $buff = fgets($this->_handler,4096);
                 if (!stristr($buff,"</a>"))
                   $buff = str_replace("\n","",$buff);
                 $str .= $buff;
               } 
             }
             if (stristr($buff,"<a name")  || stristr($buff,"</ul>") ||
                 stristr($buff,"</table>") || stristr($buff,"</ol>")){
               $str = substr($str,0,-1);    // Remove last \n
               return $str;
             }
           }
         }
     }

     return $str;
   }

   function getReleases($aname,$ignore_RC_and_M = true){
     $releases = array();
     $str = $this->getLinksAfterAName($aname);
     if ($str == "")                          // "<a name>" not found
       return $releases;

     $arr = explode("\n",$str);
     for($i=0;$i<count($arr);$i++){
       if (!preg_match(DDD,$arr[$i]))          // No digit-dot-digit version
         continue;
       if ($ignore_RC_and_M && preg_match(RCM,$arr[$i]))  // RC & M patterns
         continue;
       $begin_pos = strpos($arr[$i],"<a href=") + strlen("<a href=");
       $end_pos = strpos($arr[$i],">",$begin_pos);
       $tmp_pos = $end_pos;                          // Save for next search
       $end_pos -= $begin_pos;
       $tmp_str = substr($arr[$i],$begin_pos,$end_pos);
       $link = str_replace("\"","",$tmp_str);        // Get actual link
       if (!strstr($link,"tp://")) // Check for relative positioning
         $link = $this->_url_info.$link; 
       $begin_pos = $tmp_pos + 1;
       $end_pos = strpos($arr[$i],"<",$begin_pos);
       $end_pos -= $begin_pos;
       $name = substr($arr[$i],$begin_pos,$end_pos); // Get name for link
       $name = str_replace(" ","",$name);
       $releases[$name] = $link;
     }

     return $releases;
   }
}

?>