#!/usr/local/bin/perl

$\ = "\n";

# keep count of each word in a hash
%words = ();
# keep number of words per line in an array
@wordcount = ();

open(FH,"sherlock.txt");
while($line = <FH>) {
  chomp $line;
  # split line at whitespace (i.e. into words)
  @words = split(/\W/,$line);
  # iterate through all words in the line
  $wordcount = 0;
  for $word (@words) {
    if($word =~ /\w/) {
      # increment count for this word
      $words{$word}++;
      $wordcount++;
    }
  }
  # add number of passed words in this line to the array
  push @wordcount, $wordcount;
}

$wordcount_total = 0;
for $i (0..@wordcount-1) {
  $wordcount_total += $wordcount[$i];
  print qq(line $i had $wordcount[$i] words);
}

print qq(saw ),scalar(@wordcount),qq( lines in file);
print qq(saw $wordcount_total words in file);
print qq(average words per line ),$wordcount_total/@wordcount;

@words_common = sort { $words{$b} <=> $words{$a} } keys %words;
@words_length = sort { length($b) <=> length($a) } keys %words;

for $i (0..4) {
  print qq(common word $i $words_common[$i]);
}
for $i (0..4) {
  print qq(longest word $i $words_length[$i]);
}


close(FH);