In the case where your XPath query will find only one node (e.g.: when referring to an attribute), instead of using findnodes, you could the findvalue method to get the text content of the matching node:

#!/usr/bin/perl

use strict;
use warnings;
use XML::LibXML;

my $filename = $ARGV[0];

my $parser = XML::LibXML->new();
my $doc    = $parser->parse_file($filename);
my $xc = XML::LibXML::XPathContext->new( $doc->documentElement()  );


foreach my $sections ($xc->findnodes('/HOSTNAME/PATROL/ACL/USERNAME'))
+ {
    my $username    = $sections->findvalue('./@name');
    my $permissions = $sections->findvalue('./PERMISSION');
    my $host        = $sections->findvalue('./HOST');
    print "$username\n";
    print "$permissions\n";
    print "$host\n";
}
[download]

Also, if your document doesn't actually use namespaces (which your example doesn't) then you might as well leave out XML::LibXML::XPathContext and call $doc->findnodes instead.

However if your document does use namespaces then instead of doing this:

$sections->findnodes('./PERMISSION');
[download]

You need to do this for the namespace prefixes to work:

$xc->findnodes('./PERMISSION', $sections);
[download]