时间:2023-01-20 09:33:57 | 栏目:Python代码 | 点击:次
前言:
枚举(enumeration
)在许多编程语言中常被表示为一种基础的数据结构使用,枚举帮助组织一系列密切相关的成员到同一个群组机制下,一般各种离散的属性都可以用枚举的数据结构定义,比如颜色、季节、国家、时间单位等
在Python中没有内置的枚举方法,起初模仿实现枚举属性的方式是
class Directions: NORTH = 1 EAST = 2 SOUTH = 3 WEST = 4
使用成员:
Direction.EAST
Direction.SOUTH
检查成员:
>>> print("North的类型:", type(Direction.NORTH)) >>> print(isinstance(Direction.EAST, Direction)) North的类型: <class 'int'> False
成员NORTH的类型是int,而不是Direction
,这个做法只是简单地将属性定义到类中
Python
标准库enum实现了枚举属性的功能,接下来介绍enum的在实际工作生产中的用法
enum
规定了一个有限集合的属性,限定只能使用集合内的值,明确地声明了哪些值是合法值,,如果输入不合法的值会引发错误,只要是想要从一个限定集合取值使用的方式就可以使用enum
来组织值。
from enum import Enum class Directions(Enum): NORTH = 1 EAST = 2 SOUTH = 3 WEST = 4
使用和类型检查:
>>> Directions.EAST <Directions.EAST: 2> >>> Directions.SOUTH <Directions.SOUTH: 3> >>> Directions.EAST.name 'EAST' >>> Directions.EAST.value 2 >>> print("South的类型:", type(Directions.SOUTH)) South的类型: <enum 'Directions'> >>> print(isinstance(Directions.EAST, Directions)) True >>>
检查示例South
的的类型,结果如期望的是Directions
。name
和value
是两个有用的附加属性。
实际工作中可能会这样使用:
fetched_value = 2 # 获取值 if Directions(fetched_value) is Directions.NORTH: ... elif Directions(fetched_value) is Directions.EAST: ... else: ...
输入未定义的值时:
>>> Directions(5) ValueError: 5 is not a valid Directions
>>> for name, value in Directions.__members__.items(): ... print(name, value) ... NORTH Directions.NORTH EAST Directions.EAST SOUTH Directions.SOUTH WEST Directions.WEST
可以用于将定义的值转换为获取需要的值
from enum import Enum class Directions(Enum): NORTH = 1 EAST = 2 SOUTH = 3 WEST = 4 def angle(self): right_angle = 90.0 return right_angle * (self.value - 1) @staticmethod def angle_interval(direction0, direction1): return abs(direction0.angle() - direction1.angle()) >>> east = Directions.EAST >>> print("SOUTH Angle:", east.angle()) SOUTH Angle: 90.0 >>> west = Directions.WEST >>> print("Angle Interval:", Directions.angle_interval(east, west)) Angle Interval: 180.0
from enum import Enum from functools import partial def plus_90(value): return Directions(value).angle + 90 class Directions(Enum): NORTH = 1 EAST = 2 SOUTH = 3 WEST = 4 PLUS_90 = partial(plus_90) def __call__(self, *args, **kwargs): return self.value(*args, **kwargs) @property def angle(self): right_angle = 90.0 return right_angle * (self.value - 1) print(Directions.NORTH.angle) print(Directions.EAST.angle) south = Directions(3) print("SOUTH angle:", south.angle) print("SOUTH angle plus 90: ", Directions.PLUS_90(south.value))
输出:
0.0
90.0
SOUTH angle: 180.0
SOUTH angle plus 90: 270.0
key: 1.将函数方法用partial包起来;2.定义__call__
方法。
忽略大小写:
class TimeUnit(Enum): MONTH = "MONTH" WEEK = "WEEK" DAY = "DAY" HOUR = "HOUR" MINUTE = "MINUTE" @classmethod def _missing_(cls, value: str): for member in cls: if member.value == value.upper(): return member print(TimeUnit("MONTH")) print(TimeUnit("Month"))
继承父类Enum
的_missing_
方法,在值的比较时将case改为一致即可
输出:
TimeUnit.MONTH
TimeUnit.MONTH
第一种,执行SomeEnum
(“abc”)时想要引发自定义错误,其中"abc"是未定义的属性值
class TimeUnit(Enum): MONTH = "MONTH" WEEK = "WEEK" DAY = "DAY" HOUR = "HOUR" MINUTE = "MINUTE" @classmethod def _missing_(cls, value: str): raise Exception("Customized exception") print(TimeUnit("MONTH")) TimeUnit("abc")
输出:
TimeUnit.MONTH
ValueError: 'abc' is not a valid TimeUnit
...
Exception: Customized exception
第二种:执行SomeEnum.__getattr__
(“ABC”)时,想要引发自定义错误,其中"ABC"是未定义的属性名称,需要重写一下EnumMeta中的__getattr__方法,然后指定实例Enum对象的的metaclass
from enum import Enum, EnumMeta from functools import partial class SomeEnumMeta(EnumMeta): def __getattr__(cls, name: str): value = cls.__members__.get(name.upper()) # (这里name是属性名称,可以自定义固定传入大写(或小写),对应下面的A1是大写) if not value: raise Exception("Customized exception") return value class SomeEnum1(Enum, metaclass=SomeEnumMeta): A1 = "123" class SomeEnum2(Enum, metaclass=SomeEnumMeta): A1 = partial(lambda x: x) def __call__(self, *args, **kwargs): return self.value(*args, **kwargs) print(SomeEnum1.__getattr__("A1")) print(SomeEnum2.__getattr__("a1")("123")) print(SomeEnum2.__getattr__("B")("123"))
输出:
SomeEnum1.A1
123
...
Exception: Customized exception
动态创建和修改Enum对象,可以在不修改原定义好的Enum类的情况下,追加修改,这里借用一个说明示例,具体的场景使用案例可以看下面的场景举例
>>> # Create an Enum class using the functional API ... DirectionFunctional = Enum("DirectionFunctional", "NORTH EAST SOUTH WEST", module=__name__) ... # Check what the Enum class is ... print(DirectionFunctional) ... # Check the items ... print(list(DirectionFunctional)) ... print(DirectionFunctional.__members__.items()) ... <enum 'DirectionFunctional'> [<DirectionFunctional.NORTH: 1>, <DirectionFunctional.EAST: 2>, <DirectionFunctional.SOUTH: 3>, <DirectionFunctional.WEST: 4>] dict_items([('NORTH', <DirectionFunctional.NORTH: 1>), ('EAST', <DirectionFunctional.EAST: 2>), ('SOUTH', <DirectionFunctional.SOUTH: 3>), ('WEST', <DirectionFunctional.WEST: 4>)]) >>> # Create a function and patch it to the DirectionFunctional class ... def angle(DirectionFunctional): ... right_angle = 90.0 ... return right_angle * (DirectionFunctional.value - 1) ... ... ... DirectionFunctional.angle = angle ... ... # Create a member and access its angle ... south = DirectionFunctional.SOUTH ... print("South Angle:", south.angle()) ... South Angle: 180.0
注:这里没有使用类直接声明的方式来执行枚举(定义时如果不指定值默认是从1开始的数字,也就相当于NORTH = auto(),auto是enum中的方法),仍然可以在后面为这个动态创建的DirectionFunctional
创建方法,这种在运行的过程中修改对象的方法也就是python
的monkey patching
。
Functional APIs的用处和使用场景举例:
在不修改某定义好的Enum类的代码块的情况下,下面示例中是Arithmethic
类,可以认为是某源码库我们不想修改它,然后增加这个Enum类的属性,有两种方法:
1.enum.Enum对象的属性不可以直接被修改,但我们可以动态创建一个新的Enum类,以拓展原来的Enum对象
例如要为下面的Enum对象Arithmetic增加一个取模成员MOD="%",但是又不能修改Arithmetic类中的代码块:
# enum_test.py from enum import Enum class Arithmetic(Enum): ADD = "+" SUB = "-" MUL = "*" DIV = "/"
就可以使用enum的Functional APIs方法:
# functional_api_test.py from enum import Enum DynamicEnum = Enum("Arithmetic", {"MOD": "%"}, module="enum_test", qualname="enum_test.Arithmetic") print(DynamicEnum.MOD) print(eval(f"5 {DynamicEnum.MOD.value} 3"))
输出:
Arithmetic.MOD
2
注意:动态创建Enum对象时,要指定原Enum类所在的module名称: "Yourmodule",否则执行时可能会因为找不到源无法解析,qualname要指定类的位置:"Yourmodule.YourEnum",值用字符串类型
2.使用aenum.extend_enum可以动态修改enum.Enum对象
为enum.Enum
类Arithmetic
增加一个指数成员EXP="**",且不修改原来的Arithmetic类的代码块:
# functional_api_test.py from aenum import extend_enum from enum_test import Arithmetic extend_enum(Arithmetic, "EXP", "**") print(Arithmetic, list(Arithmetic)) print(eval(f"2 {Arithmetic.EXP.value} 3"))
输出:
<enum 'Arithmetic'> [<Arithmetic.ADD: '+'>, <Arithmetic.SUB: '-'>, <Arithmetic.MUL: '*'>, <Arithmetic.DIV: '/'>, <Arithmetic.EXP: '**'>]
8