A simple understanding of python reference counting and weak references (with examples)

This article brings you a simple understanding of python reference counting and weak references (with examples). It has certain reference value. Friends in need can refer to it. I hope it will be helpful to you.

Written in front:

The previous socket series has come to an end, mainly to make a summary and record of what I have learned.

Next I plan to write an HTTP server that supports concurrency based on libevent. Because the master in the study group had already developed a server that supports both HTTP and WebSocket before, I felt very interested after reading it, so I bought this network programming book to read. I plan to refer to his to write a streamlined version,

only supports HTTP. After I finish writing it, I can share it with everyone. Although it cannot be used in a production environment, I believe that after learning the source code, it will be a great improvement for myself.

Next enter the topic:

First put the example code:

We create one for testing The x object includes a constructor and a destructor.

The x object is initialized for the first time and referenced by variable a, so the reference count here is 1. Then get the number of references through getrefcount, which here becomes 2, because calling the function getrefcount itself is also a reference to a.

Next, create a weak reference through weakref.ref(). After printing the reference count through getrefcount, it is found that the count has not changed.

w() is a This sentence means to determine whether the two instances are exactly the same, that is, whether they occupy the same memory address.

Then we use del a to dereference a to the x object. Note that del here does not directly recycle the x object, but dereferences it. We know that once the reference of a variable is 0, it will be Recycle. So the x object is recycled after calling del.

Through the previous code, we can see that the method of calling weak references is the method of class functions. However, you can use proxy to call using variable names instead.

At this point I believe that I have definitely mastered the use of weak references. But what exactly does a weak reference do?

Like many other high-level languages, Python also has a garbage collection mechanism. Each object has a reference count, and when this count reaches 0, the object can be safely destroyed.

Since only one object can be recycled at a time, reference counting cannot recycle objects with circular references. A set of objects that reference each other will survive forever if they are not directly referenced by other objects and are inaccessible.

If an application continues to generate such inaccessible object groups, a memory leak will occur.

So weak references are very suitable for handling this kind of circular reference scenario.

The above is the detailed content of A simple understanding of python reference counting and weak references (with examples). For more information, please follow other related articles on the PHP Chinese website!