ahem/lib/ahem/lexer.rb

class Lexer
  def initialize(source)
    @source = source.split("\n").join(' ')
    @position = 0
  end

  def get_token
    return Token.new(TokenKinds::EOF) if at_end

    @position += 1 while !at_end && @source.slice(@position).match(/\s/)

    source = @source.slice(@position..-1)
    if source.match(/^null/)
      @position += 4
      Token.new(TokenKinds::NULL)
    elsif source.match(/^true/)
      @position += 4
      Token.new(TokenKinds::BOOLEAN, true)
    elsif source.match(/^false/)
      @position += 5
      Token.new(TokenKinds::BOOLEAN, false)
    elsif source.match(/^if/)
      @position += 2
      Token.new(TokenKinds::IF)
    elsif source.match(/^elseif/)
      @position += 6
      Token.new(TokenKinds::ELSEIF)
    elsif source.match(/^else/)
      @position += 4
      Token.new(TokenKinds::ELSE)
    elsif source.match(/^\d+(\.\d+)?/)
      number = source.match(/^\d+(\.\d+)?/)[0]
      @position += number.size
      Token.new(TokenKinds::NUMBER, number.to_f)
    elsif source.match(/^"(.*)"/)
      string = source.match(/^"([^"]*)"/)[1]
      @position += (string.size + 2)
      Token.new(TokenKinds::STRING, string)
    elsif source.match(/^\:([a-z][a-zA-Z0-9_]*)/)
      atom = source.match(/^\:([a-z][a-zA-Z0-9_]*)/)[1]
      @position += atom.size + 1
      Token.new(TokenKinds::ATOM, atom.to_sym)
    elsif source.match(/^\+/)
      @position += 1
      Token.new(TokenKinds::OPERATOR, :+)
    elsif source.match(/^\-/)
      @position += 1
      Token.new(TokenKinds::OPERATOR, :-)
    elsif source.match(/^\*/)
      @position += 1
      Token.new(TokenKinds::OPERATOR, :*)
    elsif source.match(%r{^\/})
      @position += 1
      Token.new(TokenKinds::OPERATOR, :/)
    elsif source.match(/^\{/)
      @position += 1
      Token.new(TokenKinds::LBRACE)
    elsif source.match(/^\}/)
      @position += 1
      Token.new(TokenKinds::RBRACE)
    elsif source.match(/^\(/)
      @position += 1
      Token.new(TokenKinds::LPAREN)
    elsif source.match(/^\)/)
      @position += 1
      Token.new(TokenKinds::RPAREN)
    elsif source.match(/^\[/)
      @position += 1
      Token.new(TokenKinds::LBRACKET)
    elsif source.match(/^\]/)
      @position += 1
      Token.new(TokenKinds::RBRACKET)
    elsif source.match(/^\;/)
      @position += 1
      Token.new(TokenKinds::SEMICOLON)
    elsif source.match(/^,/)
      @position += 1
      Token.new(TokenKinds::COMMA)
    elsif source.match(/^\./)
      @position += 1
      Token.new(TokenKinds::DOT)
    elsif source.match(/^=>/)
      @position += 2
      Token.new(TokenKinds::ROCKET)
    elsif source.match(/^!/)
      @position += 1
      Token.new(TokenKinds::OPERATOR, :!)
    elsif source.match(/^==/)
      @position += 2
      Token.new(TokenKinds::OPERATOR, :==)
    elsif source.match(/^\<=/)
      @position += 2
      Token.new(TokenKinds::OPERATOR, :<=)
    elsif source.match(/^\>\=/)
      @position += 2
      Token.new(TokenKinds::OPERATOR, :>=)
    elsif source.match(/^\</)
      @position += 1
      Token.new(TokenKinds::OPERATOR, :<)
    elsif source.match(/^\>/)
      @position += 1
      Token.new(TokenKinds::OPERATOR, :>)
    elsif source.match(/^and/)
      @position += 3
      Token.new(TokenKinds::OPERATOR, :and)
    elsif source.match(/^or/)
      @position += 2
      Token.new(TokenKinds::OPERATOR, :or)
    elsif source.match(/^not/)
      @position += 3
      Token.new(TokenKinds::OPERATOR, :not)
    elsif source.match(/^\=/)
      @position += 1
      Token.new(TokenKinds::EQUALS)
    elsif source.match(/^let/)
      @position += 3
      Token.new(TokenKinds::LET)
    elsif source.match(/^function/)
      @position += 8
      Token.new(TokenKinds::FUNCTION)
    elsif source.match(/^class/)
      @position += 5
      Token.new(TokenKinds::CLASS)
    elsif source.match(/^public/)
      @position += 6
      Token.new(TokenKinds::PUBLIC)
    elsif source.match(/^private/)
      @position += 7
      Token.new(TokenKinds::PRIVATE)
    elsif source.match(/^for/)
      @position += 3
      Token.new(TokenKinds::FOR)
    elsif source.match(/^in/)
      @position += 2
      Token.new(TokenKinds::IN)
    elsif source.match(/^[a-z][a-zA-Z0-9_]*/)
      identifier = source.match(/^[a-z][a-zA-Z0-9_]*/)[0]
      @position += identifier.size
      Token.new(TokenKinds::IDENTIFIER, identifier)
    elsif source.match(/^[A-Z][a-zA-Z0-9_]*/)
      class_name = source.match(/^[A-Z][a-zA-Z0-9_]*/)[0]
      @position += class_name.size
      Token.new(TokenKinds::CLASS_NAME, class_name)
    else
      raise "Unrecognized character #{source[0]}"
    end
  end

  def scan_all
    tokens = Array.new
    until at_end
      if @source.slice(@position).match(/\s/)
        @position += 1
      else
        tokens << get_token
      end
    end
    tokens << Token.new(TokenKinds::EOF)
    tokens
  end

  private

  def at_end
    @position == @source.size
  end
end