cx86/lexer.rb

#--
# DO NOT MODIFY!!!!
# This file is automatically generated by rex 1.0.5
# from lexical definition file "lexer.rex".
#++

require 'racc/parser'
class CX86 < Racc::Parser
  require 'strscan'

  class ScanError < StandardError ; end

  attr_reader   :lineno
  attr_reader   :filename
  attr_accessor :state

  def scan_setup(str)
    @ss = StringScanner.new(str)
    @lineno =  1
    @state  = nil
  end

  def action
    yield
  end

  def scan_str(str)
    scan_setup(str)
    do_parse
  end
  alias :scan :scan_str

  def load_file( filename )
    @filename = filename
    open(filename, "r") do |f|
      scan_setup(f.read)
    end
  end

  def scan_file( filename )
    load_file(filename)
    do_parse
  end


  def next_token
    return if @ss.eos?
    
    # skips empty actions
    until token = _next_token or @ss.eos?; end
    token
  end

  def _next_token
    text = @ss.peek(1)
    @lineno  +=  1  if text == "\n"
    token = case @state
    when nil
      case
      when (text = @ss.scan(/int/))
         action {[:TYPE,:INT]}

      when (text = @ss.scan(/\(/))
         action {[:OPEN_PAREN,"("]}

      when (text = @ss.scan(/\)/))
         action {[:CLOSE_PAREN,")"]}

      when (text = @ss.scan(/\{/))
         action {[:OPEN_CURLY,"{"]}

      when (text = @ss.scan(/\}/))
         action {[:CLOSE_CURLY,"}"]}

      when (text = @ss.scan(/;/))
         action {[:SEMICOLON,";"]}

      when (text = @ss.scan(/-/))
         action {[:MINUS,"-"]}

      when (text = @ss.scan(/~/))
         action {[:COMP,"~"]}

      when (text = @ss.scan(/!/))
         action {[:LNEG,"!"]}

      when (text = @ss.scan(/\+=/))
         action {[:SETEQ,:plus]}

      when (text = @ss.scan(/\-\=/))
         action {[:SETEQ,:minus]}

      when (text = @ss.scan(/\/\=/))
         action {[:SETEQ,:mul]}

      when (text = @ss.scan(/\*\=/))
         action {[:SETEQ,:div]}

      when (text = @ss.scan(/\%\=/))
         action {[:SETEQ,:mod]}

      when (text = @ss.scan(/\<\<\=/))
         action {[:SETEQ,:shl]}

      when (text = @ss.scan(/\>\>\=/))
         action {[:SETEQ,:shr]}

      when (text = @ss.scan(/\&\=/))
         action {[:SETEQ,:and]}

      when (text = @ss.scan(/\|\=/))
         action {[:SETEQ,:or]}

      when (text = @ss.scan(/\^\=/))
         action {[:SETEQ,:xor]}

      when (text = @ss.scan(/\+/))
         action {[:PLUS,"+"]}

      when (text = @ss.scan(/\*/))
         action {[:MUL,"*"]}

      when (text = @ss.scan(/\//))
         action {[:DIV,"/"]}

      when (text = @ss.scan(/%/))
         action {[:MOD,"%"]}

      when (text = @ss.scan(/\&/))
         action {[:AND,"&"]}

      when (text = @ss.scan(/\|/))
         action {[:OR,"|"]}

      when (text = @ss.scan(/\^/))
         action {[:XOR,"^"]}

      when (text = @ss.scan(/<</))
         action {[:SHL,"<<"]}

      when (text = @ss.scan(/>>/))
         action {[:SHR,">>"]}

      when (text = @ss.scan(/\&\&/))
         action {[:LAND,"&&"]}

      when (text = @ss.scan(/\|\|/))
         action {[:LOR,"||"]}

      when (text = @ss.scan(/\=\=/))
         action {[:EQ,"=="]}

      when (text = @ss.scan(/\!\=/))
         action {[:NE,"!="]}

      when (text = @ss.scan(/\</))
         action {[:LT,"<"]}

      when (text = @ss.scan(/\<\=/))
         action {[:LE,"<="]}

      when (text = @ss.scan(/\>/))
         action {[:GT,">"]}

      when (text = @ss.scan(/\>\=/))
         action {[:GE,">="]}

      when (text = @ss.scan(/=/))
         action {[:EQUALS,"="]}

      when (text = @ss.scan(/,/))
         action {[:COMMA,","]}

      when (text = @ss.scan(/return/))
         action {[:RETURN,"return"]}

      when (text = @ss.scan(/\d+/))
         action {[:NUM,text.to_i]}

      when (text = @ss.scan(/\w+/))
         action {[:IDENT,text]}

      when (text = @ss.scan(/[ \t\n]+/))
         action {}

      else
        text = @ss.string[@ss.pos .. -1]
        raise  ScanError, "can not match: '" + text + "'"
      end  # if

    else
      raise  ScanError, "undefined state: '" + state.to_s + "'"
    end  # case state
    token
  end  # def _next_token

def tokenize(code)
  scan_setup(code)
  tokens = []
  while token = next_token
    tokens << token
  end
  tokens
end # class
end