Documentation

A Directory object is essentially a URL. But it was introduced for two reasons. 1) Several Policy certificates contain urls to search for certificates all types of certificates. And since these urls are used in the signable of these certificates, we must maintain the exact string representation of the url. Example: "http://hostname:8080/filename "http://hostname:80/filename/ "http://hostname/filename/ The URL class can parse this but the URL.toString() will not return the exact string. Should it? 2) A URLHandler may not be installed. And not being able to parse a particular url should not trigger an exception. Of course we can cacth the exception. But that option is a little awkward. Moreover a URLHandler may be installed at a later time. Hope u are convinced. :) A Directory object provides us a cleaner and a more flexible solution for the future.