In python, descriptor (descriptor) is like this.

In python, descriptor (descriptor) is like this.

Many tutorials often make descriptor very complex, long articles, and so on.

In fact, a single sentence is used to hook class operations to control behavior.

Most of the time, it is used to block access to instance attributes.
As long as there are _ get _ (), _ set _ (), and _ delete _ () methods in the class. it is a descriptor. let's think about it. If we can't escape the three methods for an operation on a class, we need to hook the methods for any operation we need.
The descriptor is not self host, but parasitic in other classes.
The implementation principles of property, classmethod, staticmethod, and super are exactly the descriptors.

To put it so much, we will use the code below to demonstrate it clearly.

# Coding = utf-8class Integer (object): # Integer is a descriptor, because the _ set _ () method is defined. def _ init _ (self, name): self. name = name def _ set _ (self, instance, value): # Because we only need to hook the behavior of "modifying attributes, therefore, we only need to define the _ set _ () method. We do not need _ get _ () or _ delete __(). if not isinstance (value, int): raise TypeError ('expected an int') instance. _ dict _ [self. name] = valueclass Point (object): x = Integer ('x') y = Integer ('y') def _ init _ (self, x, y ): self. x = x self. y = yp = Point (2, 3) p. x = 9p. x = 9.9 # This statement will throw a TypeError: Expected an int error. this is the role of the descriptor.