one-file-projects/gerlang.py

class Analyzer:
	def __init__(self, itr):
		self.itr = itr

	def next(self,amount=1):
		if len(self.itr) == 0: return None
		if amount == 1: return self.itr[0]
		return self.itr[:amount]

	def take(self,amount=1):
		m, self.itr = self.next(amount), self.itr[amount:]
		return m

	def __len__(self):
		return len(self.itr)

class StringAnalyzer(Analyzer):
	def nextOrd(self):
		m = self.next()
		if m is None: return None
		return ord(m)

	def between(self,cmin,cmax):
		c = self.nextOrd()
		if c is None: return False
		return c >= ord(cmin) and c <= ord(cmax)

	def is_a(self,c):
		return self.next() == c

class TokenListAnalyzer(Analyzer):
	def takeUntilType(self,end):
		return self.takeUntil(lambda t: t[0] == end)

	def takeUntil(self,condition, dropEnd=True):
		t = []
		while self.next() is not None and not condition(self.next()):
			t.append( self.take() )
		if dropEnd and self.next() is not None:
			self.take()
		return t


class Lexer:
	keywords = ["setze","auf","durch","schreibe"]
	operators = ["plus","minus","mal","geteilt"]
	IDENT = 0
	KEYWORD = 1
	INT = 2
	FLOAT = 3
	OP = 4
	BRACE_OPEN = 5
	BRACE_CLOSE = 6
	NEWLINE = 7

	def lex(self, source):
		tokens = []
		sa = StringAnalyzer(source)
		braces = 0
		stack = 0
		while len(sa) != 0:
			if sa.between('a', 'z') or sa.between('A', 'Z'): #identifier or keyword
				ident = ""
				while sa.between('a', 'z') or sa.between('A', 'Z') or sa.between('0', '9') or sa.is_a('_'):
					ident += sa.take()
				if ident.lower() in self.keywords:
					tokens.append( (self.KEYWORD,ident.lower()) )
				elif ident.lower() in self.operators:
					tokens.append( (self.OP,ident.lower()) )
				elif ident.lower() == "wenn":
					tokens.append( ( self.KEYWORD, "wenn", stack ) )
					stack += 1
				elif ident.lower() == "ende":
					stack -= 1
					tokens.append( ( self.KEYWORD, "ende", stack ) )
				else:
					tokens.append( (self.IDENT,ident) )
			elif sa.between('0', '9'): #number
				num = ""
				t = (self.INT, int)
				while sa.between('0', '9'):
					num += sa.take()
				if sa.is_a(','):
					t = (self.FLOAT, float)
					sa.take()
					num += "."
					while sa.between('0', '9'):
						num += sa.take()

				tokens.append( (t[0],t[1](num)) )
			elif sa.is_a('('):
				tokens.append( (self.BRACE_OPEN,braces) )
				braces+=1
			elif sa.is_a(')'):
				braces-=1
				tokens.append( (self.BRACE_CLOSE,braces) )
			elif sa.is_a('\n'):
				tokens.append( (self.NEWLINE,) )
				sa.take()
			elif sa.is_a(' ') or sa.is_a('\t') or sa.is_a('\r'):
				sa.take()
			else:
				raise ParserException("WTF is %s" % sa.take() )

		return tokens

class ParserException(Exception):
	pass

class Parser:
	def parse(self,tokens):
		block = BlockTerm()
		ta = TokenListAnalyzer(tokens)
		while len(ta) > 0:
			if ta.next()[0] == Lexer.KEYWORD:
				if ta.next()[1] == "setze":
					ta.take()
					if ta.next()[0] != Lexer.IDENT:
						raise ParseException("missing identifier after setze")
					ident = ta.take()[1]
					if ta.next()[0] != Lexer.KEYWORD or ta.next()[1] != "auf":
						raise ParserException("missing auf after identifier")
					ta.take()
					term = self.__parseTerm(ta.takeUntil(lambda t: t[0] == Lexer.NEWLINE))
					block.append(AssignmentTerm(ident,term))
				elif ta.next()[1] == "schreibe":
					ta.take()
					term = self.__parseTerm(ta.takeUntilType(Lexer.NEWLINE))
					block.append(PrintTerm(term))
				elif ta.next()[1] == "wenn":
					stack = ta.next()[2]
					ta.take()
					condition = ta.takeUntil(lambda t: t[0] == Lexer.NEWLINE)
					b = ta.takeUntil(lambda t: t[0] == Lexer.KEYWORD and t[1] == "ende" and t[2] == stack)
					block.append( ConditionalTerm(self.__parseTerm(condition), self.parse(b) ) )
				else:
					raise Exception("what? %s" % str(ta.next()))
			elif ta.next()[0] == Lexer.NEWLINE:
				ta.take()
			else:
				raise Exception("huh? %s" % str(ta.next()))
 		return block

	def __parseTerm(self,tokens):
		t = tokens[0]
		if t[0] == Lexer.INT or t[0] == Lexer.FLOAT:
			return ValueTerm(t[1])
		elif t[0] == Lexer.IDENT:
			return IdentifierTerm(t[1])
		else:
			return ValueTerm(0)

class Context():
	pass

class SubContext(Context):
	pass

class Term(object):
	def run(self,context):
		raise Exception("get_value must be overwritten")

class Operator2(Term):
	token = None
	priority = 0
	def __init__(self, left, right):
		self.left = left
		self.right = right

	def run(self,context):
		return self.calc( self.left.run(context), self.right.run(context) )

	def calc(self, r, l):
		raise Exception("calc not implemented")

class IdentifierTerm:
	def __init__(self, ident):
		self.ident = ident

	def run(self,context):
		return context[self.ident]

class ValueTerm:
	def __init__(self, value):
		self.value = value

	def run(self,context):
		return self.value

class AssignmentTerm(Term):
	def __init__(self,ident,term):
		self.ident = ident
		self.term = term

	def run(self,context):
		context[self.ident] = self.term.run(context)
		return None

class PrintTerm(Term):
	def __init__(self,term):
		self.term = term

	def run(self,context):
		print self.term.run(context)
		return None

class ConditionalTerm(Term):
	def __init__(self,condition,block):
		self.condition = condition
		self.block = block

	def run(self,context):
		m = self.condition.run(context)
		if m != 0:
			self.block.run(context)


class BlockTerm(Term):
	def __init__(self):
		self.terms = []

	def append(self,item):
		self.terms.append(item)

	def run(self,context):
		result = None
		for term in self.terms:
			result = term.run(context)
		return result

def main():
	context = {}
	while True:
		code = raw_input(">>> ")
		tokens = Lexer().lex(code)
		term = Parser().parse(tokens)
		term.run(context)

if __name__ == '__main__':
	main()