#!/usr/bin/perl

use String::LCSS_XS qw( lcss );
use utf8;
use warnings;


my $f1 = shift ; 
my $f2 = shift ; 

open (FILE1, "<:encoding(UTF-8)", "$f1") or die "can't open file '$f1' $!";
open (FILE2, "<:encoding(UTF-8)", "$f2") or die "can't open file '$f2' $!";

my @array1 = <FILE1>;
chomp(@array1);
close (FILE1);

my @array2 = <FILE2>;
chomp(@array2);
close (FILE2);

my $subseq;
my($bestsource, $besttarget, $bestalignment, $bestlength);

for my $i (0 .. $#array1) {
	my $best_subseq = "";
	my $best_subseq_words = 0;
	my $best_subseq_chars = 0;
	my $found = 0;

	for my $j (0 .. $#array2) {

		$subseq = lcssw ("$array1[$i]", "$array2[$j]");

		my $num_words = count_words ($subseq);
		my $num_chars = count_chars ($subseq);

		if ($num_words > $best_subseq_words && $num_chars > $best_subseq_chars) {
			$best_subseq = $subseq;
			$best_subseq_words = $num_words;
			$best_subseq_chars = $num_chars;
			$found = 1;
		}
	}
	
	if ($found == 1) {
		print "$best_subseq is the lcssw of $array1[$i]\n"
	}
}

sub lcssw {
   my ($s1, $s2) = @_;

   my $i;
   my %codes;
   my %words;

   for ($s1, $s2) {
      $_ = join '',
           map { $codes{$_} = chr(++$i) if !exists($codes{$_});
                 $codes{$_}
               }
           $_ =~ /\w+/g;
   }

   my $lcss = lcss($s1, $s2);
   $lcss = "" if (!defined $lcss);

   @words{values %codes} = keys %codes;
   return join ' ', @words{ $lcss =~ /./sg };
}

sub count_words {
	my $line = shift ;
	my @text_words = split(/\s+/, $line);
	return scalar(@text_words); 
}

sub count_chars {
	my $line = shift ;
	my @text_words = split(//, $line);
	return scalar(@text_words); 
}