This is a function I use to determine if a file contains Binary information. I use this for my search engine so that it doesn't try to index files like .zip or .mp3 or any other file that doesn't contain readable information. It makes use of the Character Type Extension if it's loaded, if it's not then it uses Regular Expressions.
function is_binary($link)
{
$tmpStr = '';
@$fp = fopen($link, 'rb');
@$tmpStr = fread($fp, 256);
@fclose($fp);
if($tmpStr != '')
{
$tmpStr = str_replace(chr(10), '', $tmpStr);
$tmpStr = str_replace(chr(13), '', $tmpStr);
$tmpInt = 0;
for($i =0; $i < strlen($tmpStr); $i++)
{
if( extension_loaded('ctype') )
{
if( !ctype_print($tmpStr[$i]) )
$tmpInt++;
}
else
{
if( !eregi("[[:print:]]+", $tmpStr[$i]) )
$tmpInt++;
}
}
if($tmpInt > 5)
return(0);
else
return(1);
}
else
return(0);
}