extract_5chars.rb

# extract_5chars.rb
# extract effective 5-character words for the initial word of Wordle.
#
# Written by: Shimpei Makimoto
# License: MIT License https://makimoto.mit-license.org/
#
# usage: ruby extract_5chars.rb <input_file>
# The input_file is a list of words that are separated by newline.
#   such as: https://github.com/tabatkins/wordle-list/blob/main/words
#            https://dumps.wikimedia.org/enwiktionary/

WORD_LENGTH = 5
FREQ_CHARACTERS_THRESHOLD = 5
LISTED_CANDIDATES_THRESHOLD = 50

words = []
ARGF.each do |l|
  l.chomp!
  if l =~ /\A[a-z]{#{WORD_LENGTH}}\z/
    words << l
  end
end

puts "words: #{words.size}"

char_counts = Array.new(WORD_LENGTH) { Hash.new(0) }

words.each do |w|
  w.each_char.with_index do |c, i|
    char_counts[i][c] += 1
  end
end

chars_order_by_freq = Array.new(WORD_LENGTH) { [] }
char_counts.each_with_index do |char_count, i|
  puts "\nchar_count[#{i}]:"
  char_count.sort {|a, b| b[1] <=> a[1] }.first(FREQ_CHARACTERS_THRESHOLD).each do |c|
    chars_order_by_freq[i] << c[0]
    puts "#{c[0]}: #{c[1]}"
  end
end

found_words = []

chars_order_by_freq[0].each_with_index do |c0, i0|
  chars_order_by_freq[1].each_with_index do |c1, i1|
    chars_order_by_freq[2].each_with_index do |c2, i2|
      chars_order_by_freq[3].each_with_index do |c3, i3|
        chars_order_by_freq[4].each_with_index do |c4, i4|
          word = [c0, c1, c2, c3, c4]
          index = [i0, i1, i2, i3, i4]
          if words.include?(word.join) && word.uniq.size == WORD_LENGTH
            found_words << [index, word]
          end
        end
      end
    end
  end
end

puts "\nfound_words:"
found_words.sort_by {|w| w[0].inject(:+) }.first(LISTED_CANDIDATES_THRESHOLD).each do |w|
  puts "#{w[0].join}: #{w[1].join}"
end