=head1 NAME

domain - nnrpd domain resolver

=head1 SYNOPSIS

B<domain> B<domainname>

=head1 DESCRIPTION

This program can be used in F<readers.conf> to grant access based on the
subdomain part of the remote hostname.  In particular, it only returns
success if the remote hostname ends in B<domainname>.  (A leading dot on
B<domainname> is optional; even without it, the argument must match on
dot-separated boundaries).  The "username" returned is whatever initial
part of the remote hostname remains after B<domainname> is removed.  It
is an error if there is no initial part (that is, if the remote hostname
is I<exactly> the specified B<domainname>).

=head1 EXAMPLE

The following readers.conf(5) fragment grants access to hosts with
internal domain names:

    auth internal {
        res: "domain .internal"
	default-domain: "example.com"
    }

    access internal {
        users: "*@example.com"
        newsgroups: example.*
    }

Access is granted to the example.* groups for all connections from hosts
that resolve to hostnames ending in C<.internal>; a connection from
"foo.internal" would match access groups as "foo@example.com".

=head1 BUGS

It seems the code does not confirm that the matching part is actually at
the end of the remote hostname (e.g., "domain: example.com" would match
the remote host "foo.example.com.org" by ignoring the trailing ".org"
part).

Does this resolver actually provide any useful functionality not
available by using wildcards in the readers.conf(5) I<hosts> parameter?
If so, the example above should reflect this functionality.

=head1 HISTORY

This documentation was written by Jeffrey M. Vinocur <jeff@litech.org>.

$Id: domain.pod 5988 2002-12-12 23:02:14Z vinocur $

=head1 SEE ALSO

nnrpd(8), readers.conf(5)

=cut