WebSearch

by: Steven Chalker

January 29, 2002

Version: 1.0

Type: Sample Code (HOWTO)

Category: HTTP

License: GNU General Public License

Description: Searches AllTheWeb, Google, Yahoo, Excite and more search engines and turns it into an array. Manual included in script.

Manual:
$search = WebSearch(your text to search here)

$search[search #]["num_results"] - Returns Number of Results
$search[search #]["link"] - Return the URL
$search[search #]["title"] - Title
$search[search #]["description"] - Part of text (eg. <B>Chocolate</B> is cool)
-----------------
<?
function WebSearch($tosearch) {
$q = urlencode($tosearch);
$start = time();

 $eng[0]['url']  = "http://www.google.com/search?q=$q&hl=en&lr=&safe=off&start=0&sa=N";
 $eng[0]['name'] = "Google";
 $eng[0]['reg']  = "/<p><A HREF=(.*?)>(.*?)<\/A><font.*?<br>(.*?)<br>/";
 $eng[0]['strip'] = "\n";

 $eng[1]['url'] = "http://www.alltheweb.com/search?cat=web&lang=any&query=$q";
 $eng[1]['name'] = "AllTheWeb";
 $eng[1]['reg'] = '/<dt>.*?href=\"(.*?)\">(.*?)<\/a>.*?<dd>(.*?)<br>/i';
 $eng[1]['strip'] = "\n";

 $eng[2]['url'] = "http://altavista.com/sites/search/web?q=$q&enc=&kl=XX&search=Search";
 $eng[2]['name'] = "Altavista";
 $eng[2]['reg'] = "/<A .*?status='(.*?)'.*?\">(.*?)<\/A>.*?<BR>(.*?)<BR>/i";
 $eng[2]['strip'] = "\n";

 $eng[3]['url'] = "http://search.msn.com/results.asp?ba=(0.0)0&co=(0.10)4.1..2.1.4.1&FORM=MSNH&RS=CHECKED&q=$q&v=1";
 $eng[3]['name'] = "MSN";
 $eng[3]['reg'] = "/<A CLASS.*?href=\"(.*?)\">(.*?)<\/A><br.*?>(.*?)<br/i";
 $eng[3]['strip'] = "";

 $eng[4]['url'] = "http://search.excite.com/search.gw?c=web&search=$q&onload=";
 $eng[4]['name'] = "Excite";
 $eng[4]['reg'] = "/pos=.*?;(http.*?)onMouse.*?<b>(.*?)<br>.*?size8>(.*?)<\/span>/i";
 $eng[4]['strip'] = "\n";

 $eng[5]['url'] = "http://google.yahoo.com/bin/query?p=$q&hc=1&hs=4";
 $eng[5]['name'] = "Yahoo";
 $eng[5]['reg'] = "/href.*?\*(.*?)\">(.*?)<br>.*?<\/b>(.*?)<br>/i";
 $eng[5]['strip'] = "\n"; 

 $urls = array();
 $out = array();
 $y=0;

 foreach($eng as $c)
   {
     $text1 = join("",file($c['url']));

     if($c[strip])
       $text1 = ereg_replace($c[strip],"",$text1);
       
     preg_match_all($c['reg'],$text1,$matches);
     $num = count($matches[0]);
     
     for($x=0;$x<$num;$x++) {
     $url = strip_tags($matches[1][$x]);
     $title = strip_tags($matches[2][$x]);
     $description = strip_tags($matches[3][$x]);
     $engine = $c[name];
     
     if($out["$url"])
       {
         $out["$url"][engine] .= ", $engine";
       }
     else
       {
         $out["$url"][title] = $title;
         $out["$url"][description] = $description;
         $out["$url"][engine] = $engine;
       }  
      $y++;
     } 
       
   }#rof

 while(list($url,$rec) = each($out)) {
 extract($rec);
 $returned_array = array();
 for($i=0; $i < count($out); $i++) {
 $returned_array[$i]["secs"] = time() - $start;
 $returned_array[$i]["num_results"] = $y;
 $returned_array[$i]["link"] = $url;
 $returned_array[$i]["title"] = $title;
 $returned_array[$i]["description"] = $description;
 return $returned_array;
 }
 }
}
?>

Comment and Contribute

0 Comments (click to add your comment)

Your comment has been submitted and is pending approval.

Author:

Steven Chalker

Comment:

Name/nickname:
Email:

Comment:

(Maximum characters: 1200). You have characters left.