Python描述符进阶：__get__/__set__协议与属性拦截底层深度解析

class ReadOnly:
    """非数据描述符：只实现 __get__，不实现 __set__"""
    def __init__(self, value):
        self._value = value

    def __get__(self, instance, owner):
        # instance 为 None 表示通过类访问，例如 MyClass.attr
        if instance is None:
            return self
        return self._value


class Box:
    tag = ReadOnly("fixed")   # 定义在类上的描述符


b = Box()
print(b.tag)        # -> fixed（走 __get__）
b.tag = "changed"   # 非数据描述符不管 __set__，于是写进了实例字典
print(b.tag)        # -> changed（实例字典盖过了非数据描述符）
print(Box.tag)      # -> <__main__.ReadOnly object>（类访问时 instance=None）

class Validated:
    def __init__(self, min_=0, max_=100):
        self.min_ = min_
        self.max_ = max_
        self._store = {}          # 用实例 id 存真实值，避免递归

    def __set_name__(self, owner, name):
        self.name = name          # Python 3.6+ 自动注入字段名

    def __get__(self, instance, owner):
        if instance is None:
            return self
        return self._store.get(id(instance))

    def __set__(self, instance, value):
        if not (self.min_ <= value <= self.max_):
            raise ValueError(f"{self.name} 必须落在 [{self.min_}, {self.max_}]")
        self._store[id(instance)] = value   # 真值存在描述符自身，不会被 __setattr__ 递归


class Player:
    hp = Validated(0, 100)


p = Player()
p.hp = 80
print(p.hp)        # -> 80
p.hp = 200         # ValueError: hp 必须落在 [0, 100]

class Celsius:
    def __init__(self, temp=0):
        self._temp = temp

    @property                 # 非数据？错——property 三个方法都实现，是数据描述符
    def temp(self):
        return self._temp

    @temp.setter
    def temp(self, value):
        if value < -273.15:
            raise ValueError("低于绝对零度")
        self._temp = value


# 等价的手工描述符写法：
def _get(self):    return self._temp
def _set(self, v):
    if v < -273.15: raise ValueError("低于绝对零度")
    self._temp = v
class Celsius2:
    temp = property(_get, _set)

class Field:
    """通用字段描述符：类型 + 范围 + 必填校验"""
    def __init__(self, typ, min_=None, max_=None, required=True):
        self.typ = typ
        self.min_, self.max_ = min_, max_
        self.required = required
        self._data = {}

    def __set_name__(self, owner, name):
        self.name = name

    def __get__(self, instance, owner):
        if instance is None:
            return self
        if self.name not in self._data:
            if self.required:
                raise AttributeError(f"{self.name} 尚未赋值")
            return None
        return self._data[self.name]

    def __set__(self, instance, value):
        if not isinstance(value, self.typ):
            raise TypeError(f"{self.name} 必须是 {self.typ.__name__}")
        if self.min_ is not None and value < self.min_:
            raise ValueError(f"{self.name} 不能小于 {self.min_}")
        if self.max_ is not None and value > self.max_:
            raise ValueError(f"{self.name} 不能大于 {self.max_}")
        self._data[self.name] = value


class User:
    uid = Field(int, min_=1)
    name = Field(str)
    age = Field(int, min_=0, max_=150)


u = User()
u.uid = 7
u.name = "老周"
u.age = 29
print(u.uid, u.name, u.age)     # -> 7 老周 29
u.age = 200                     # ValueError: age 不能大于 150