解释器实例之算数计算（一）

实现一个简单的解释器，其中只包含整数和加减乘除符号，无括号，并且乘除优先级最高。

参考这里：https://github.com/rspivak/lsbasi

# Token types
#
# EOF (end-of-file) token is used to indicate that
# there is no more input left for lexical analysis
INTEGER, PLUS,MINUS,MULT, DIV, EOF = 'INTEGER','PLUS','MINUS', 'MULT','DIV', 'EOF'


class Token(object):
    def __init__(self, type, value):
        # token type: INTEGER, PLUS, or EOF
        self.type = type
        # token value: 0, 1, 2. 3, 4, 5, 6, 7, 8, 9, '+', or None
        self.value = value

    def __str__(self):
        """String representation of the class instance.
        Examples:
            Token(INTEGER, 3)
            Token(PLUS '+')
        """
        return 'Token({type}, {value})'.format(
            type=self.type,
            value=repr(self.value)
        )

    def __repr__(self):
        return self.__str__()

class Lexer(object):
    def __init__(self,text):
        self.text=text
        self.current_pos=0
        self.current_char=self.text[self.current_pos]

    def error(self):
        raise Exception('Invalid character!')
    def advance(self):
        self.current_pos+=1
        if self.current_pos > len(self.text)-1 :
            self.current_char=None
        else:
            self.current_char=self.text[self.current_pos]

    def skip_whitespace(self):
        while self.current_char is not None and self.current_char.isspace():
            self.advance()

    def integer(self):
        result=''
        while self.current_char is not None and self.current_char.isdigit():
            result+=self.current_char
            self.advance()
        return int(result)

    def get_next_token(self):
        while self.current_char is not None:
            #print('get_next_token---->'+self.current_char)
            if self.current_char.isspace():
                self.skip_whitespace()
                continue
            if self.current_char.isdigit():
                return Token(INTEGER,self.integer())
            if self.current_char=='+':
                self.advance()
                return Token(PLUS,'+')
            if self.current_char=='-':
                self.advance()
                return Token(MINUS,'-')
            if self.current_char=='*':
                self.advance()
                return Token(MULT,'*')
            if self.current_char=='/':
                self.advance()
                return Token(DIV,'/')
            self.error()

        return Token(EOF,None)

class Parser(object):
    def __init__(self,lexer):
        self.lexer=lexer
        self.current_token=self.lexer.get_next_token()

    def error(self):
        raise Exception('Invalid syntax')

    def eat(self,token_type):
        #print('eat---->'+self.current_token.type)
        if self.current_token.type==token_type:
            self.current_token=self.lexer.get_next_token()
        else:
            self.error()

    def factor(self):
        token=self.current_token
        self.eat(INTEGER)
        return token.value

    def term(self):
        r=self.factor()
        #print('term---->'+self.current_token.type)
        while self.current_token.type != EOF :
            if self.current_token.type==MULT:
                self.eat(MULT)
                r=r*self.factor()
            elif self.current_token.type==DIV:
                self.eat(DIV)
                r=r/self.factor()
            else:
                break
                #print("term error....."+self.current_token.type)
        return r

    def expr(self):
        t=self.term()
        #print('expr---->'+str(t))
        while self.current_token.type != EOF :
            if self.current_token.type==PLUS:
                self.eat(PLUS)
                t=t+self.term()
            elif self.current_token.type==MINUS:
                self.eat(MINUS)
                t=t-self.term()
            else:
                #????
                pass
        return t

    def parse(self):
        return self.expr()

def main():
    while True:
        try:
            # To run under Python3 replace 'raw_input' call
            # with 'input'
            text = raw_input('calc> ')
        except EOFError:
            break
        if not text:
            continue

        parser = Parser(Lexer(text))
        result=parser.parse()
        print(result)


if __name__ == '__main__':
    main()