How to create a webcatching program?

<?php

$open = fopen("http://www.yahoo.com/index.html", "r"); 

// the above statement opens up the yahoo! page using the socket libraries and reads it,
// fopen takes the pages as the 1st arguement and type of opening (r in this case) as the 2nd 

$read = fread($open, 15000); 

// fread scours to the 15000th byte and captures all the encountered bytes into $read 

fclose($open); 

// close the socket connection 

$search = eregi("</b></td><td width=100%>(.*)</td></tr><tr><td 
align=right colspan=2><br> <a href=r/xn><small>more...</small>", $read, $printing); 

// the magic line, eregi is a regex which does case insensitive search on a data, in this case
// is the $read variable. It searches for all data in between the 1st instance of "..."
// and the 1st instance of "</td></tr>..." I have generally seen that it should be valid html
// tags for it to result in a positive search, not logically correct though! If the search is
// true then store it in $printing 

$printing[1] = str_replace("href=\"/homer/?", "href=\"", $printing[1]);
$printing[1] = str_replace("</td></tr><tr><td valign=top><b> ", "", $printing[1]);
$printing[1] = str_replace(" </b></td><td>", "", $printing[1]);
$printing[1] = str_replace("<small>", "", $printing[1]);
$printing[1] = str_replace("</small>", "", $printing[1]); 

// the above code block is string manipulations, 1st replacing /homer/ then doing common
// denomination, i.e. in every news headline there is a common denominator the last 4 lines
// just do string manipulation on this basis, helping us to shift each headline into an array
// element, rather than just bringing in the yahoo! interface into it. 

$content = $printing[1];
$content = explode("·", $content); 

// well this is also a common cancellation or rather the place where the shifting headlines into
// elements actually takes place with the help of "?" and explode(); 

$headlines = sizeof($content); 

// counting the array elements 

for ($i = 0; $i < $headlines; $i++) { 

print "News HeadLine (No. ".($i+1).") : $content[$i]
"; 

} 

// this is looping through each headline retrieved

?>