letterpress word finder

letter.rb

# letters are either the first argument or something piped in
#
# e.g.
#   ruby letter.rb KTVHROBDRBDLCYTPLEWAFZYMB
# or
#   echo "KTVHROBDRBDLCYTPLEWAFZYMB" | ruby letter.rb
# or
#   ruby -r ./board_parser -e "puts BoardParser.new('light.png').tiles.join" | ruby letter.rb

letters = (ARGV[0] || STDIN.read).downcase

words = File.read("/usr/share/dict/words").downcase.split("\n")

# via http://stackoverflow.com/questions/11349544/ruby-optimize-the-comparison-of-two-arrays-with-duplicates
def is_subset?(word, letters)
  !word.chars.find{|char| word.count(char) > letters.count(char)}
end

matching_words = words.select do |word|
  is_subset?(word, letters)
end

puts matching_words.sort_by(&:length)