#!/usr/bin/perl -w

use strict;

######################################################################
# Variables                                                          #
######################################################################

# File with the list of domains to process.  The format of the file should be:
# .xx Country name
# .xy Some other country
# .xz Yet another country
my $tlds              = 'tlds.txt';

# Dictionary file.  Words are assumed to be sorted alphabetically, one per line
my $dictionary        = '/usr/share/dict/linux.words';

# Minimum length of word to consider (3 letters, 2 of which are from the domain)
my $min_length        = 3;

# Words that start with a match to this regexp will be ignored
my $ignore_start_with = qr/-/;

# Words that match this regexp will be ignored
my $ignore_match      = qr/_/;

######################################################################
# Code                                                               #
# ####################################################################

# Load TLDs
my %domains = ();
open (TLDS, "<$tlds") or die "Can't load domains from $tlds : $!\n";
while (my $line = <TLDS>) {
	chomp($line);
	my ($domain,$country) = split(/\s+/, $line, 2);
	$domain =~ s/^\.//;
	$domains{$domain} = $country;
}
close(TLDS);

# Search dictionary for words ending in each domain
foreach my $domain (sort keys %domains) {
	print "Domains for $domain ($domains{$domain}):\n";
	open (DICT, "<$dictionary") or die "Can't load dictionary from $dictionary : $!\n";
	my $cached_word = '';
	while (my $line = <DICT>) {
		chomp($line);
		next if (length $line < $min_length);
		next if ($line =~ m/^${ignore_match}/);
		next if ($line =~ m/${ignore_start_with}/);
		$line = lc($line);
		# Print matches only once (assuming dictionary is sorted)
		if (($line =~ m/$domain$/) && ($line ne $cached_word)) {
			print " $line\n";
			$cached_word = $line;
		}
	}
	close(DICT);
	print "\n";
}