DailyProgrammer Challenge #162i Solution (Novel Compression, pt. 2: Compressing the Data)

c162i.rb

def tokenize(word, dict)
  return word[0] if ['-', 'R*'].include? word
  token = (dict.find_index word.downcase.gsub(/[.,\?!;:]/, '')).to_s
  (dict << word.downcase.gsub(/[.,\?!;:]/, '');
    token = (dict.length - 1).to_s) unless token.length > 0
  case word
    when /^[a-z]*[.,\?!;:]?$/; # nothing
    when /^[A-Z][a-z]*[.,\?!;:]?$/; token += '^'
    when /^[A-Z]*[.,\?!;:]?$/; token += '!'
    else; puts "Error! Invalid case or punctuation in word #{word}."; abort
  end
  word.match(/[.,\?!;:]$/) {|m| word = word[0..(word.length - 1)]; token += ' ' + m[0]}
  (puts "Error! Invalid punctuation #{word[-1]}."; abort) unless word[-1].match /[a-zA-Z.,\?!;:]/
  return token
end

def compress(data)
  dict = []
  word_list = data.lines.map do |l|
     (l + ' R*')
      .chomp
      .gsub(/\-/, ' - ')
      .split(' ')
  end.flatten.map do |w|
    tokenize(w, dict)
  end
  word_data = word_list
    .each_slice(24)
    .to_a
    .map {|ws| ws.join ' '}
    .join "\n"
  return "#{dict.length}\n#{dict.join "\n"}\n#{word_data} E"
end

data = gets(nil)
puts compress data