Typically I code using the easiest-to-read-and-understand method but I was just playing with an idea earlier and was curious on the "better" ways you can come up with to solve this problem.

You have a dictionary file of, say, 1 million words. You want to write a script to take RANDOMLY 100 words of each letter (a .. z). You want 100 words beginning with A, 100 words beginning with B, etc.

This is what I did. It's untested and doesn't need to work since I'm looking for how you guys would solve this.

open(DICT, "dictionary.txt" or die "Error: $!";
my @words = <DICT>;
close(DICT) or die "Error: $!";

@words = shuffle(@words); # assume shuffle is defined

my @new_words;

my @letters = (a .. z);
my @cnt = 0;

foreach my $letter (@letters)
{
  $cnt = 0;

  foreach my $word (@words)
  {
     if ($cnt > 100)
     {
        last;  # break loop, our 100 word is found
     }

     if ($word =~ m/^$letter/i)
     {
        $cnt++;
        push (@new_words, $word);
     }
  }
        
}
[download]