#!/usr/bin/ruby

# code generated by "translate.ruby.pss" a pep script
# http://bumble.sf.net/books/pars/tr/
# require 'something'

class Machine 
  # make a new machine 
  attr_accessor :work, :charsRead, :linesRead, :escape, :delimiter, 
    :counter, :stack, :tape, :cell, :marks, :eof, :peep
  
  def initialize() 
    @size = 300      # how many elements in stack/tape/marks
    @eof = false     # end of stream reached?
    @charsRead = 0   # how many chars already read
    @linesRead = 1   # how many lines already read
    @escape = "\\"
    @delimiter = "*" # push/pop delimiter (default "*")
    @counter = 0     # a counter for anything
    @work = ""       # the workspace
    @stack = []      # stack for parse tokens 
    @cell = 0                # current tape cell
    @tape = Array.new(@size) {String.new}    # a list of attribute for tokens 
    @marks = Array.new(@size) {String.new}   # marked tape cells 
    # or dont initialse peep until "parse()" calls "setInput()"
    @peep = ARGF.readchar
  end

  # multiline strings are ok in ruby
  def printSizeError()
    puts "
  Tape max size exceeded!
  tape maximum size = #{@size}
  tape cell (current) = #{@cell} 
    You can increase the @size value in the ruby script 
    but normally this error indicates an error in your parsing 
    script. The only exception would be massively nested structures
    in the source data."
  end

  def setInput(newInput) 
    puts "to be implemented"
  end

  # read one character from the input stream and 
  #    update the machine.
  def read 
    # require 'io/console'
    if @eof then exit end
    @charsRead += 1;
    # increment lines
    if @peep == "\n" then @linesRead += 1 end
    @work += @peep
    @peep = ARGF.readchar
    if @peep.nil? then @eof = true; end
    # @peep = STDIN.getch
    # and check for control d
    # STDIN.getch.tap { |char| exit(1) if char == "u0003" }
    #if !@peep then @eof = true end
  end

  # test if all chars in workspace are in unicode category
  def isInCategory(cat) 
    #for ch in @work
    #  if not category(ch).start_with?(cat) then return false end
    #return True
  end

  # this needs to actually walk the string
  # eg "abcab\cabc"
  # not trivial
  def unescapeChar(c)
    @work.gsub!("#{@escape}#{c}", c)
  end

  # add escape character : trivial?
  def escapeChar(c)
    @work.gsub!(c, @escape+c)
  end

  # a helper for the multiescape until bug
  def countEscaped(suffix) 
    count = 0
    s = @work 
    s.sub!(/#{suffix}$/, "")
    while s.end_with?(@escape) 
      count++
      s.sub!(/#{@escape}$/, "")
    end
  end

  # reads the input stream until the workspace end with text 
  def until(suffix) 
    # read at least one character
    if @eof then return end
    self.read()
    while true do
      if @eof then return end
      # no, we need to count the @escape chars preceding suffix
      # if odd, keep reading, if even, stop
      if @work.end_with?(suffix) then 
        if (self.countEscaped(suffix) % 2 == 0) then return end
      end
      self.read()
    end
  end  

  # pop the first token from the stack into the workspace */
  def pop() 
    if @stack.length == 0 then return false end
    @work = @stack.pop() + @work
    if @cell > 0 then @cell -= 1 end
    return true
  end

  # push the first token from the workspace to the stack 
  def push() 
    # dont increment the tape pointer on an empty push
    if @work == "" then return false end
    # need to get this from the delimiter.
    iFirst = @work.index(@delimiter)
    if iFirst.nil?
      @stack.push(@work); @work = ""; return true
      # also @stack << @work
    end
    # s[i..j] means all chars from i to j
    # s[i,n] means n chars from i
    @stack.push(@work[0..iFirst])
    @work = @work[iFirst+1..-1]
    if @cell < @size then 
      @cell += 1
    else
      self.printSizeError(); exit 
    end
    return true
  end

  def printState() 
    puts "Stack[#{@stack.join(', ')}] Work[#{@work}] Peep[#{@peep}]"
    puts "Acc:#{@counter} Esc:#{@escape} Delim:#{@delimiter} Chars:#{@charsRead}" +
         " Lines:#{@linesRead} Cell:#{@cell}"
  end

  # this is where the actual parsing/compiling code should go
  # so that it can be used by other ruby classes/objects. Also
  # should have a stream argument.
  def parse(s) 
    # a reset or "setinput()" method would be useful to parse a 
    # different string/file/stream, without creating a new
    # machine object.
    # could use code like this to check if input is string or file
    #if isinstance(s, file)
      print("")
      # @reset(s)
      # @reader = s
    #elseif isinstance(s, string)
      #f = StringIO.StringIO("test")
      #for line in f  print(line)
    #else
    #  f = STDIN 
    #end
    #puts "not implemented"
  end

end

# end of Machine class definition

# will become:
# mm.parse(sys.stdin)  or 
# mm.parse("abcdef") or
# open f; mm.parse(f)

# the restart flag, which allows .restart to work before the 
# parse label, in languages (like ruby) that dont have 
# labelled loops
restart = false
mm = Machine.new 
mm.work += ""
mm.work += "\n    An attempt at basic natural language parsing. "
mm.work += "\n    Use the following words in simple sentences: "
mm.work += "\n"
mm.work += "\n     articles: the, this, her, his, a, one, some, "
mm.work += "\n     preposition: up, in, at, on, with, under, to"
mm.work += "\n     adjectives: simple, big, small, blue, beautiful, small,"
mm.work += "\n     nouns: flower, tree, dog, house, horse, girl, fish, meat,"
mm.work += "\n     verbs: runs, eats, sleeps, is, grows, digs, sings"
mm.work += "\n"
mm.work += "\n    End the sentence with a full stop \".\""
mm.work += "\n      eg: the small dog eats fish."
mm.work += "\n      eg: the simple horse runs on the house ."
mm.work += "\n   .\n"
print mm.work         # print
mm.work = ''          # clear
while !mm.eof do 
  
  # lex block 
  while true 
    mm.read()        # read
    if (mm.work.match?(/^[[:alpha:]]+$/)) then
      # while  
      while /^[[:alpha:]]+$/.match?(mm.peep)
        if mm.eof then break end
        mm.read()
      end
      mm.tape[mm.cell] = mm.work  # put 
      if (mm.work == "the" || mm.work == "this" || mm.work == "her" || mm.work == "his" || mm.work == "a" || mm.work == "one" || mm.work == "some") then
        mm.work = ''          # clear
        mm.work += "article*"
        mm.push();
        break
      end
      if (mm.work == "up" || mm.work == "in" || mm.work == "at" || mm.work == "on" || mm.work == "with" || mm.work == "under" || mm.work == "to") then
        mm.work = ''          # clear
        mm.work += "preposition*"
        mm.push();
        break
      end
      if (mm.work == "simple" || mm.work == "big" || mm.work == "small" || mm.work == "blue" || mm.work == "beautiful" || mm.work == "small") then
        mm.work = ''          # clear
        mm.work += "adjective*"
        mm.push();
        break
      end
      if (mm.work == "flower" || mm.work == "tree" || mm.work == "dog" || mm.work == "house" || mm.work == "horse" || mm.work == "girl" || mm.work == "fish" || mm.work == "meat") then
        mm.work = ''          # clear
        mm.work += "noun*"
        mm.push();
        break
      end
      if (mm.work == "runs" || mm.work == "eats" || mm.work == "sleeps" || mm.work == "is" || mm.work == "grows" || mm.work == "digs" || mm.work == "sings") then
        mm.work = ''          # clear
        mm.work += "verb*"
        mm.push();
        break
      end
      mm.tape[mm.cell] = mm.work  # put 
      mm.work = ''          # clear
      mm.work += "<"
      mm.work += mm.tape[mm.cell] # get
      mm.work += ">"
      mm.work += " Sorry, don't understand that word! \n"
      print mm.work         # print
      mm.work = ''          # clear
      exit
    end
    # use a full-stop to complete sentence
    if (mm.work == ".") then
      mm.tape[mm.cell] = mm.work  # put 
      mm.work = ''          # clear
      mm.work += "dot*"
      mm.push();
    end
    # ignore every thing else
    mm.work = ''          # clear
    break 
  end
  if restart then restart = false; next; end
  
  # parse block 
  while true 
    # 2 tokens
    mm.pop();
    mm.pop();
    if (mm.work == "article*noun*") then
      mm.work = ''          # clear
      mm.work += mm.tape[mm.cell] # get
      mm.work += " "
      mm.cell += 1                         # ++
      mm.work += mm.tape[mm.cell] # get
      if mm.cell > 0 then mm.cell -= 1; end # --
      mm.tape[mm.cell] = mm.work  # put 
      mm.work = ''          # clear
      mm.work += "nounphrase*"
      mm.push();
      next
    end
    if (mm.work == "verb*preposition*") then
      mm.work = ''          # clear
      mm.work += mm.tape[mm.cell] # get
      mm.work += " "
      mm.cell += 1                         # ++
      mm.work += mm.tape[mm.cell] # get
      if mm.cell > 0 then mm.cell -= 1; end # --
      mm.tape[mm.cell] = mm.work  # put 
      mm.work = ''          # clear
      mm.work += "verbphrase*"
      mm.push();
      next
    end
    # 3 tokens
    mm.pop();
    if (mm.work == "noun*verb*dot*" || mm.work == "nounphrase*verb*dot*" || mm.work == "noun*verbphrase*dot*" || mm.work == "nounphrase*verbphrase*dot*") then
      mm.work = ''          # clear
      mm.work += mm.tape[mm.cell] # get
      mm.work += " "
      mm.cell += 1                         # ++
      mm.work += mm.tape[mm.cell] # get
      if mm.cell > 0 then mm.cell -= 1; end # --
      mm.tape[mm.cell] = mm.work  # put 
      mm.work = ''          # clear
      mm.work += "sentence*"
      mm.push();
      next
    end
    if (mm.work == "article*adjective*noun*") then
      mm.work = ''          # clear
      mm.work += mm.tape[mm.cell] # get
      mm.work += " "
      mm.cell += 1                         # ++
      mm.work += mm.tape[mm.cell] # get
      mm.work += " "
      mm.cell += 1                         # ++
      mm.work += mm.tape[mm.cell] # get
      if mm.cell > 0 then mm.cell -= 1; end # --
      if mm.cell > 0 then mm.cell -= 1; end # --
      mm.tape[mm.cell] = mm.work  # put 
      mm.work = ''          # clear
      mm.work += "nounphrase*"
      mm.push();
      next
    end
    # 4 tokens
    mm.pop();
    if (mm.work == "nounphrase*verb*noun*dot*" || mm.work == "noun*verb*noun*dot*" || mm.work == "nounphrase*verb*nounphrase*dot*" || mm.work == "noun*verb*nounphrase*dot*" || mm.work == "nounphrase*verbphrase*nounphrase*dot*" || mm.work == "noun*verbphrase*nounphrase*dot*" || mm.work == "nounphrase*verbphrase*noun*dot*" || mm.work == "noun*verbphrase*noun*dot*") then
      mm.work = ''          # clear
      mm.work += mm.tape[mm.cell] # get
      mm.work += " "
      mm.cell += 1                         # ++
      mm.work += mm.tape[mm.cell] # get
      mm.work += " "
      mm.cell += 1                         # ++
      mm.work += mm.tape[mm.cell] # get
      if mm.cell > 0 then mm.cell -= 1; end # --
      if mm.cell > 0 then mm.cell -= 1; end # --
      mm.tape[mm.cell] = mm.work  # put 
      mm.work = ''          # clear
      mm.work += "sentence*"
      mm.push();
      next
    end
    mm.push();
    mm.push();
    mm.push();
    mm.push();
    if (mm.eof) then
      mm.pop();
      mm.pop();
      if (mm.work == "sentence*") then
        mm.work = ''          # clear
        mm.work += "It's an english sentence! \n("
        mm.work += mm.tape[mm.cell] # get
        mm.work += ") \n"
        mm.work += "But it may not make sense! \n"
        print mm.work         # print
        mm.work = ''          # clear
        exit
      end
      if (mm.work == "nounphrase*") then
        mm.work = ''          # clear
        mm.work += "its a noun-phrase! ("
        mm.work += mm.tape[mm.cell] # get
        mm.work += ") \n"
        print mm.work         # print
        mm.work = ''          # clear
        exit
      end
      if (mm.work == "verbphrase*") then
        mm.work = ''          # clear
        mm.work += "its a verb-phrase! ("
        mm.work += mm.tape[mm.cell] # get
        mm.work += ") \n"
        print mm.work         # print
        mm.work = ''          # clear
        exit
      end
      mm.push();
      mm.push();
      mm.work += "nope, not a sentence. \n"
      print mm.work         # print
      mm.work = ''          # clear
      mm.work += "The parse stack was: \n  "
      print mm.work         # print
      mm.work = ''          # clear
      while mm.pop() do next end   # unstack 
      mm.work += "\n"
      print mm.work         # print
      exit
    end
    break 
  end # parse
  
end


# end of generated code