#!/perl/bin/perl
# title:   Vigenere Pattern Grokker
# author:  Matt Rankin
# purpose: To detect recurring patterns in a Vigenere ciphertext.
# details: takes a file as an input, and some optional parameters to control
           what the program looks for and outputs.
# usage:   vpg 'filename' [size] [cutoff]

open(F,$ARGV[0]) or die("Epic I/O fail!\n");
# take command line params, otherwise use default values of 2 and 5
my $size=(defined $ARGV[1])?$ARGV[1]:2;
my $cutoff=(defined $ARGV[2])?$ARGV[2]:5;
my $cipher;
# pwn newlines
while(my $line=<F>){
   chomp($line);
   $cipher.=$line;
}
close(F);
my $l=length($cipher);
my %pattern;
for(my $i=0;$i<$l;$i++){
   my $seq=substr($cipher,$i,$size);
   $pattern{$seq}++;
}
# sort key-value pairs into descending order on value
# (i'd be lying if i claimed to know exactly how this works, such is Perl...)
my @key=sort{$pattern{$b} <=> $pattern{$a}}keys(%pattern);
my @val=sort{$b <=> $a}values(%pattern);
my @pairs;
for(my $i=0;$i<scalar(@key);$i++){
   last if($val[$i]<$cutoff);
   print("$key[$i] - $val[$i]\n");
   push(@pairs,$key[$i]);
}
my %index;
POS: for(my $i=0;$i<$l;$i++){
   my $seq=substr($cipher,$i,$size);
   foreach my $pair(@pairs){
      next unless($seq eq $pair);
      $index{$pair}.="$i ";
      next POS;
   }
}
foreach(keys %index){
   print("$_ - $index{$_} (");
   my @seq=split(/\s/,$index{$_});
   for(my $i=1;$i<scalar(@seq);$i++){
      print(",")if($i>1);
      print($seq[$i]-$seq[$i-1]);
   }
   print(")\n");
}