Day 25 面向对象

面向对象基础

面向对象编程

面向过程编程:类似于工厂的流水线

• 优点:逻辑清晰
• 缺点:扩展性差

面向对象编程:核心是对象二字,对象属性和方法的集合体,面向对象编程就是一堆对象交互

• 优点:扩展性强
• 缺点:逻辑非常乱

类与对象

• 对象:属性和方法的集合体

• 类:一系列相同属性和方法的集合体

现实世界中先有对象后有类,python中先有类,再实例化出对象

对象的属性的查找顺序

先对象本身-->类-->父类-->父类的父类-->object-->自己定制的元类-->type

给对象定制独有属性

class People:
	pass

p1 = Peolple()
p1.name = 'nick'

p2 = People()
p2.name = 'tank'

对象的绑定方法

class People:
	def eat(self):
		print(self, 'eat....')

p1 = Peolple()
p1.eat()
p1.name = 'nick'

p2 = People()
p2.eat()
p2.name = 'tank'

类与数据类型

lis = [1,2,3]  # lis = list([1,2,3])

class foo:
	def __init__(self,name):
    	self.name = name

f = foo('name')

lis.append(4)  # 对象调对象绑定的方法,会自动传参
list.append(lis,4)  # 类调用对象绑定的方法,必须得传参

面向对象进阶

类的继承

继承父类,则会有父类的所有属性和方法

class ParentClass1():
	pass

class ParentClass2():
	pass

class SubClass(ParentClass1,ParentClass2):
	pass

类的派生

继承父类的同时自己有init,然后也需要父类的init

class ParentClass1():
	def __init__(self,name):
		pass

class SubClass(ParentClass):
	def __init__(self,age):
		# 1. ParentClass1.__init__(self,name)
		# 2. super(SubClass,self).__init__(name)
		self.age = age

类的组合

类对象可以引用/当做参数传入/当做返回值/当做容器元素,类似于函数对象

class ParentClass1():
	count = 0
	def __init__(self,name):
		pass

class SubClass(ParentClass):
	def __init__(self,age):
		self.age = age	

pc = ParentClass1()
sc = SubClass()

sc.parent_class = pc  # 组合
sc.parent_class.count  # 0

菱形继承问题

新式类:继承object的类,python3中全是新式类

经典类:没有继承object的类,只有python2中有

在菱形继承的时候,新式类是广度优先(老祖宗最后找);经典类深度优先(一路找到底,再找旁边的)

多态与多态性

一种事物的多种形态,动物-->人/猪/狗

# 多态
import abc

class Animal(metaclass=abc.ABCmeta):
	@abc.abstractmethod
	def eat():
		print('eat')

class People(Animal):
	def eat():
		pass

class Pig(Animal):
	def eat():
		pass
    def run():
        pass

class Dog(Animal):  # 报错
	def run():
		pass

# 多态性
peo = People()
peo.eat()
peo1 = People()
peo1.eat()
pig = Pig()
pig.eat()

def func(obj):
	obj.eat()

class Cat(Animal):
	def eat():
		pass
cat = Cat()

func(cat)

鸭子类型:只要长得像鸭子,叫的像鸭子,游泳像鸭子,就是鸭子.

类的封装

隐藏属性,只有类内部可以访问,类外部不可以访问

class Foo():
	__count = 0 

	def get_count(self):
		return self.__count

f = Foo()
f.__count  # 报错
f._Foo__count # 不能这样做

类的property特性

把方法变成属性引用

class People():
	def __init__(self,height,weight):
		self.height = height
		self.weight = weight

	@property
	def bmi(self):
		return weight/(height**2)

	@bmi.setter
	def bmi(self,value)
		print('setter')

    @bmi.deleter
    def bmi(self):
        print('delter')

peo = People
peo.bmi

类与对象的绑定方法和非绑定方法

没有任何装饰器装饰的方法就是对象的绑定方法, 类能调用, 但是必须得传参给self

被 @classmethod 装饰器装饰的方法是类的绑定方法,参数写成cls, cls是类本身, 对象也能调用, 参数cls还是类本身

被 @staticmethod 装饰器装饰的方法就是非绑定方法, 就是一个普通的函数

面向对象高级

isinstance,issubclass

isinstance判断是否为类的实例化对象,会检测父类,而type不会检测父类

issubclass,判断是否为其子类

反射

1. hasattr:通过字符串判断是否类属性存在
2. getattr:通过字符串获取类属性
3. setattr:通过字符串修改类属性
4. delattr:通过字符串删除类属性

call

class Foo:
    def __init__(self):
        print('Foo()会触发我')
    def __call__(self):
        print('Foo()()/f()会触发我')

f = Foo()
f()

new

class Foo:
    def __new__(self):
        print('new')
        obj = object.__new__(self)
        return obj

    def __init__(self):
        print('init')

f = Foo()

元类

元类用来造类的

元类()-->类-->init

元类()()-->对象--->call

类分为几部分:类名/类体名称空间/父类们

class Mymeta(type):
    def __init__(self,class_name,class_bases,class_dic):
        # 控制类的逻辑代码
        super().__init__(class_name,class_bases,class_dic)

    def __call__(self,*args,**kwargs):
        # 控制类实例化的参数

        obj = self.__new__(self)  # obj就是实例化的对象
        self.__init__(obj,*args,**kwargs)
        print(obj.__dict__)

        # 控制类实例化的逻辑

        return obj

class People(metaclass=Mymeta):
    def __init__(self,name,age):
        self.name = name
        self.age = age

单例模式

利用类的绑定方法的特性

NAME = 'nick'
AGE = 18

class People():

    __instance = None

    @classmethod
    def from_conf(cls):
        if cls.__instance:
            return cls.__instance

        cls.__instance = cls(NAME,AGE)
        return cls.__instance

People.from_conf()

People.from_conf()

利用装饰器

NAME = 'nick'
AGE = 18

def deco(cls):
    cls.__instance = cls(NAME,AGE)

    def wrapper(*args,**kwargs):
        if len(args) == 0 and len(kwargs) == 0:
            return cls.__instance

        res = cls(*args,**kwargs)
        return res

    return wrapper

@deco
class People():
    def __init__(self,name,age):
        self.name = name
        self.age = age

peo1 = People()

peo2 = People()

利用元类(正宗的)

NAME = 'nick'
AGE = 18

class Mymeta(type):
    def __init__(self,class_name,class_bases,class_dict):
        super().__init__(class_name,class_bases,class_dict)
        self.__instance = self(NAME,AGE)

    def __call__(self,*args,**kwargs):

        if len(args) == 0 and len(kwargs) == 0:
            return self.__instance

        obj = object.__new__(self)
        self.__init__(obj,*args,**kwargs)

        return obj

class People(metaclass=Mymeta):
    def __init__(self,name,age):
        self.name = name
        self.age = age

peo1 = People()
peo2 = People()

异常处理

捕捉异常

x = 10
y = 20
c = 30

try:
    1/0
except Exception as e:
    print(e)

raise

抛出异常

raise KeyboardInterrupt('中断捕捉')

assert

判断某一行代码是否有问题