tokenizer

in lib/twitter_cldr/tokenizers/unicode_regex/unicode_regex_tokenizer.rb [18:48]

• 23 lines of code
• 2 McCabe index (conditional complexity)

      def tokenizer
        @tokenizer ||= begin
          recognizers = [
            
            TokenRecognizer.new(:variable, /\$\w[\w\d]*/),
            TokenRecognizer.new(:character_set, /\[:[\w\s=]+:\]|\\p\{[\w\s=]+\}/),  
            TokenRecognizer.new(:negated_character_set, /\[:\^[\w\s=]+:\]|\\P\{[\w\s=]+\}/),  
            TokenRecognizer.new(:unicode_char, /\\u\{?[a-fA-F0-9]{1,6}\}?/),
            TokenRecognizer.new(:multichar_string, /\{\w+\}/u),

            TokenRecognizer.new(:escaped_character, /\\./),
            TokenRecognizer.new(:negate, /\^/),
            TokenRecognizer.new(:ampersand, /&/),
            TokenRecognizer.new(:pipe, /\|/),
            TokenRecognizer.new(:dash, /-/),

            
            TokenRecognizer.new(:special_char, /\{\d,?\d?\}|[$?:{}()*+\.,\/\\]/),

            TokenRecognizer.new(:open_bracket, /\[/),
            TokenRecognizer.new(:close_bracket, /\]/),

            TokenRecognizer.new(:string, //u) do |val|
              val == " " ? val : val.strip
            end
          ]

          Tokenizer.new(recognizers)
        end
      end