$data = file_get_contents('page.html'); // получить содержимое страниц(ы)

preg_match_all("/a[\s]+[^>]*?href[\s]?=[\s\"\']+([http].*?)[\"\']+.*?>"."([^<]+|.*?)?<\/a>/is", $data, $matches);

print_r($matches[1]); // $matches[1] содержит найденные URL'ы

lynx -dump http://site/file | grep -A 1000 References | grep -v References | grep -v ^$

$html=file_get_contents ('http://www.popsu.net');
$url='popsu.net';
$vnut=array();
$vnech=array();
preg_match_all('~<a [^<>]*href=[\'"]([^\'"]+)[\'"][^<>]*>(((?!~si',$html, $matches);
foreach ($matches[1] as $val) {
if (!preg_match("~^[^=]+://~", $val) || preg_match("~^[^://]+://(www\.)?".$url."~i", $val)) { $vnut[]=$val; }
else $vnech[]=$val;
}
$vnut=array_unique ($vnut);
$vnech=array_unique ($vnech);
print_r ($vnut);
print_r ($vnech);

lynx -dump file | grep -A 1000 References | grep -v -E "References|Hidden|Visible" | grep -v ^$ | sed 's/ [0-9]*\. //g'

lynx -dump file | grep -A 1000 References | grep -v -E "References|Hidden|Visible" | grep -v ^$ | sed 's/ [0-9]*\. //g'