Python duck types and polymorphism

Python duck types and polymorphism

1. Duck type

When you see a bird that walks like a duck, swims like a duck, and calls like a duck, then this bird can be called a duck

class Cat(object):
    def say(self):
        print("i am a cat")

class Dog(object):
    def say(self):
        print("i am a dog")
        
class Duck(object):
    def say(self):
        print("i am a duck")
        
animal = Cat()
animal.say() # i am a cat

The above three classes all contain saymethods.

A first instance of animalan object, then it may give the assignment to at run time Cat, Dogor Duckso is uncertain.

In javaimplementation, multi-state, this is generally done, define a parent class Animal, which comprises a saymethod, then the definition of an inherited Animalclass Catsubclasses and Catoverride subclass saymethod, so as to indicate Cata Animalsubclass.

Based on javaideas, use pythonlanguage to realize the concept of polymorphism

class Animal:
	def say(self):
        print("i am a animal")
class Cat(Animal):
	def say(self):
        print("i am a cat")      

Animal an = Cat() 
#在java中声明变量时需要定义变量类型,python中不能这样写
 an.say() #调用子类方法

The object instantiated in Python is a dynamic variable that can point to any type, which means that the method needs to be written in the class. The code is the same as below.

animal_list = [Cat,Dog,Duck]
for animal in animal_list:
	animal().say()
# i am a cat
# i am a dog
# i am a duck

When all types of methods have implemented a common method name, they can call the same method cyclically and achieve polymorphism. Compared with the method javathat needs to inherit a parent class and rewrite the class, it is much simpler.

In other words, when all classes or objects implement a common method and the method name is the same, these classes can be classified as one type.

Second, let's look at another example

a = ["bobby1", "bobby2"]

b = ["bobby2", "bobby"]
name_tuple = ["bobby3", "bobby4"]
name_set = set()
name_set.add("bobby5")
name_set.add("bobby6")
a.extend(b)
print(a) # ["bobby1", "bobby2","bobby2", "bobby"]

Think of a question, extendcan the function only pass through a list?

The extendfunction is defined like this in the source code

 def extend(self, iterable: Iterable[_T]) -> None: ...

So extendfunction only can be used as a parameter list, any object can be iterated are possible, and in the Pythonmiddle tuple, setare iterable object.

So you can extendpass the name_setsum to the function name_tuple.

a = ["bobby1", "bobby2"]

name_tuple = ["bobby3", "bobby4"]
name_set = set()
name_set.add("bobby5")
name_set.add("bobby6")
# a.extend(name_tuple)
# print(a) # ["bobby1", "bobby2","bobby3", "bobby4"]
a.extend(name_set)
print(a) # ["bobby1", "bobby2","bobby5", "bobby6"]

Three, pass an iterable class to the extend function

class Company(object):
    def __init__(self, employee_list):
        self.employee = employee_list

    def __getitem__(self, item):
        return self.employee[item]

    def __len__(self):
        return len(self.employee)

company = Company(["tom", "bob", "jane"])

a = ["bobby1", "bobby2"]

a.extend(company)
print(a) # ["bobby1", "bobby2","tom", "bob", "jane"]

extendThe function implicitly calls the iterator in the object, for example __iter__, __getitem__(it becomes an iterable object after the call).

Subsection

Magic functions in python make full use of the duck type. Many magic functions are written in objects, which can be recognized by the python interpreter itself. Many built-in python objects or built-in classes have very useful features, such as iterable Features, context manager, collection sequence related features, etc., the code itself is more flexible.