This is probably too broad a question for the forum, but you'll want to look into cURL to pull the site content as a string. Something like the following will get you the content of a given URI:
$uri_to_get = 'http://www.wiktionary.org/some_page.html';
$ch = curl_init( $uri_to_get );
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_HEADER , 0);
$site_content = curl_exec($ch);