Shulou(Shulou.com)06/02 Report--

This article mainly introduces "how to understand PyPy in Python can make the code run faster", in daily operation, I believe many people in Python how to understand PyPy can make the code run faster on the question there are doubts, Xiaobian consulted all kinds of information, sorted out simple and easy to use operation methods, hope to answer "Python how to understand PyPy can make the code run faster" doubts help! Next, please follow the small series to learn together!

different methods

Python is an interpreted language, CPython reads and executes source code line by line. Interpreted languages (including JavaScript) have many advantages:

easy to write

Strong metaprogramming capabilities

Compile does not fail

Of course, it also has some disadvantages:

When parsing the source code, the performance cost is huge

Compile does not fail

As you can see, the pros and cons include "compilation will not fail." This is sometimes based on different behaviors (like prototyping or production), but I still tend to see it as a drawback. PyPy is slightly different in that it is not a pure interpreter, but implements trace just-in-time (JIT) compilation.

JIT

Just-in-time compilation is somewhere between interpretation and traditional compilation ahead of time. A just-in-time compiler does not execute the source code itself, but instead generates a set of low-level instructions (usually assembly) that can be executed immediately.

This illustration helps us understand the difference. In compiled languages (C, C++, Rust), the compilation phase is divided strictly according to the development environment. It generates a runnable binary file and sends it to production.

In interpreted languages, the opposite is true: the source code (after * conversion, hello JS) is pushed entirely into production and executed by the interpreter. Just-in-time compiled languages also carry source code (or bytecode, such as Java or C#), but it is compiled and run as a regular compiled language rather than interpreted line-by-line.

Not that one approach is better than the other, each use case will specify the right choice based on its unique requirements. However, if performance is critical and you feel comfortable using Python interpreters, you can opt for PyPy.

Track just-in-time compilation

Just like compilation or interpretation, there are different ways to implement just-in-time compilation. The traditional approach is method/function scoping. When code calls a function, the just-in-time compiler takes its source code, compiles it, and provides an executable binary file. PyPy takes a slightly different approach, depending on Python's features and use cases.

PyPy's compiler does not call by method, but rather computes loops. Since Python is widely used in data science, machine learning, advanced algorithms, and data structures, this makes the most sense. In short, PyPy is an optimization layer on top of Python.

PyPy doesn't handle loops as rigorously as it is understood. In addition to the regular for and while constructs, PyPy optimizes arbitrary blocks of code if it detects that compilation work is worthwhile.

disadvantages

Of course, PyPy has its drawbacks. Even if it can greatly improve performance, you need to know the following:

Not all Python is supported. It supports most of the code, but it doesn't work if you deal with the underlying CPython implementation details or have Cython bindings.

Go back to the future. PyPy is currently at version 3.4 and Python is currently stable at 3.8, but backtracking is a technique Python developers excel at.

Optimization is a good thing, but not an excuse to write bad code. If the code can't be read, how can PyPy understand it?

The global interpreter lock is still in place. If performing heavy multithreading, choose another implementation.

As with any tool, we should consider all the details before adopting it. However, the next time you log in to Codeforces for a challenge, try PyPy. It is possible that O(n^3) error codes will pass, whereas in pure Python only O(n log n) will pass.

other implementations

In addition to CPython and PyPy, there are other notable Python implementations:

StacklessPython。It's the same as Python, but without the global interpreter lock, which is used on the back end of Eve.

IronPython is a Python language implemented on NET that provides very simple interoperability between Python and c#code.

JPython is the same, but there is Java.

At this point, the study of "how to understand PyPy in Python can make code run faster" is over, hoping to solve everyone's doubts. Theory and practice can better match to help you learn, go and try it! If you want to continue learning more relevant knowledge, please continue to pay attention to the website, Xiaobian will continue to strive to bring more practical articles for everyone!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.