Python AST-检测表达式中的非法标记

我应该如何优化

compile（）

函数（解析器）以：

仅接受标记：

'numeric string（integer）

，

ADD

，

SUB

，

MUL

和

DIV

检测非法令牌/表达式并显示错误（立即停止）：

错误：#4处出现意外令牌

ADD

、

SUB

、

MUL

和

DIV

必须预期正好两个令牌：

[

'numeric string（integer）

或

ADD

或

SUB

或

MUL

或

DIV

]

及

[

'numeric string（integer）

或

ADD

或

SUB

或

MUL

或

DIV

]

函数应该从最内向的表达式开始计算表达式

---

代码：

def is_operator(token):
    return (token in ['ADD', 'SUB', 'MUL', 'DIV'] or token.isnumeric())


# Compiles a list of tokens
# into an abstract syntax tree (AST). 
# 
# A token represents either a number (e.g. "42") 
# or an operator ("ADD", "SUB", "MUL" or "DIV"). 
# 
# A node of the returned AST is either a string 
# that represents a number (e.g. "42") 
# or a tuple in the form of (operator, child1, child2) 
# where child1 and child2 are the subtrees of the AST. 
# 
# This function returns (ast, unprocessed_tokens) 
# where unprocessed_tokens are the tokens yet to compile. 
# 
def compile(tokens):
    if len(tokens) == 0:
        return ('NOOP',)
    token = tokens[0]
    if is_operator(token):
        child1, unprocessed = compile(tokens[1:])
        child2, unprocessed = compile(unprocessed)
        return (token, child1, child2), unprocessed
    else:
        return token, tokens[1:]


line = ['ADD', 'SUB', '2', '2', 'MUL', '6', 'DIV', '10', '2']
print(compile(line)[0])
# => ('ADD', ('SUB', '2', '2'), ('MUL', '6', ('DIV', '10', '2')))

---

样本#1:

def is_operator(token):
    return (token in ['ADD', 'SUB', 'MUL', 'DIV'] or token.isnumeric())


# Compiles a list of tokens
# into an abstract syntax tree (AST). 
# 
# A token represents either a number (e.g. "42") 
# or an operator ("ADD", "SUB", "MUL" or "DIV"). 
# 
# A node of the returned AST is either a string 
# that represents a number (e.g. "42") 
# or a tuple in the form of (operator, child1, child2) 
# where child1 and child2 are the subtrees of the AST. 
# 
# This function returns (ast, unprocessed_tokens) 
# where unprocessed_tokens are the tokens yet to compile. 
# 
def compile(tokens):
    if len(tokens) == 0:
        return ('NOOP',)
    token = tokens[0]
    if is_operator(token):
        child1, unprocessed = compile(tokens[1:])
        child2, unprocessed = compile(unprocessed)
        return (token, child1, child2), unprocessed
    else:
        return token, tokens[1:]


line = ['ADD', 'SUB', '2', '2', 'MUL', '6', 'DIV', '10', '2']
print(compile(line)[0])
# => ('ADD', ('SUB', '2', '2'), ('MUL', '6', ('DIV', '10', '2')))

输入：

line = ['ADD', 'SUB', '3', '2', 'MUL', '6', 'DIV', '10', '2']

line = ['ADD', '3', '2', '1']

预期产出：

AST: ('ADD', ('SUB', '3', '2'), ('MUL', '6', ('DIV', '10', '2')))

ANSWER: 31

ERROR: Unexpected token at # 4

---

样本#2:

def is_operator(token):
    return (token in ['ADD', 'SUB', 'MUL', 'DIV'] or token.isnumeric())


# Compiles a list of tokens
# into an abstract syntax tree (AST). 
# 
# A token represents either a number (e.g. "42") 
# or an operator ("ADD", "SUB", "MUL" or "DIV"). 
# 
# A node of the returned AST is either a string 
# that represents a number (e.g. "42") 
# or a tuple in the form of (operator, child1, child2) 
# where child1 and child2 are the subtrees of the AST. 
# 
# This function returns (ast, unprocessed_tokens) 
# where unprocessed_tokens are the tokens yet to compile. 
# 
def compile(tokens):
    if len(tokens) == 0:
        return ('NOOP',)
    token = tokens[0]
    if is_operator(token):
        child1, unprocessed = compile(tokens[1:])
        child2, unprocessed = compile(unprocessed)
        return (token, child1, child2), unprocessed
    else:
        return token, tokens[1:]


line = ['ADD', 'SUB', '2', '2', 'MUL', '6', 'DIV', '10', '2']
print(compile(line)[0])
# => ('ADD', ('SUB', '2', '2'), ('MUL', '6', ('DIV', '10', '2')))

输入：

line = ['ADD', 'SUB', '3', '2', 'MUL', '6', 'DIV', '10', '2']

line = ['ADD', '3', '2', '1']

预期产出：

AST: ('ADD', ('SUB', '3', '2'), ('MUL', '6', ('DIV', '10', '2')))

ANSWER: 31

ERROR: Unexpected token at # 4

---

PS:

def is_operator(token):
    return (token in ['ADD', 'SUB', 'MUL', 'DIV'] or token.isnumeric())


# Compiles a list of tokens
# into an abstract syntax tree (AST). 
# 
# A token represents either a number (e.g. "42") 
# or an operator ("ADD", "SUB", "MUL" or "DIV"). 
# 
# A node of the returned AST is either a string 
# that represents a number (e.g. "42") 
# or a tuple in the form of (operator, child1, child2) 
# where child1 and child2 are the subtrees of the AST. 
# 
# This function returns (ast, unprocessed_tokens) 
# where unprocessed_tokens are the tokens yet to compile. 
# 
def compile(tokens):
    if len(tokens) == 0:
        return ('NOOP',)
    token = tokens[0]
    if is_operator(token):
        child1, unprocessed = compile(tokens[1:])
        child2, unprocessed = compile(unprocessed)
        return (token, child1, child2), unprocessed
    else:
        return token, tokens[1:]


line = ['ADD', 'SUB', '2', '2', 'MUL', '6', 'DIV', '10', '2']
print(compile(line)[0])
# => ('ADD', ('SUB', '2', '2'), ('MUL', '6', ('DIV', '10', '2')))

我知道解析库，我不会朝这个方向走。我需要通过使用本地python代码来解决这个问题

---

我已经编写了

lexer（）

函数。

你的问题到底是什么？问题更新了。你的问题到底是什么？问题更新了。