Python best practices

Best Practices Guide Tutorial

Introduction

Python is a vast language and even though it’s easy to learn, it’s quite difficult to master. On your journey, you would soon realize that having the right tools to aid you is essential to the completion of your journey. What do I mean by a tool? Anything which lets you write code faster, more efficiently and easily or helps put down my thoughts and ideas into something less abstract.

Editors/IDEs

Even though I put editors and IDEs under the same category, they are actually quite different. For some languages like C++ and Rust, IDEs perform much better than editors. However for Python, this isn’t really an issue.

VSCode is the highest rated code editor and it stays at the top for a number of reasons.I won’t go into them but let me tell you, once you are comfortable with VSCode and its vast ecosystem of extensions, you would have to think twice before switching to any other code editor. We will be using VSCode for this post.

Another popular IDE is Pycharm. I don’t use it since it doesn’t handle non python files very well and most of my projects use more than just python. On the other hand, VSCode does this quite well.

Version Control System (VCS)

VCS gives you the ability to go back in time and pick up versions of code you need. This might not sound very useful but you’d know how useful it is once you have messed up a project so much that CTRL+Z can’t take you back far enough in the code history to fix the issue. Moreover, with services like Github, Gitlab etc., you can share your code with ease. These days most companies would look at you Github history before hiring you for a programming job, because it is often an accurate measure of how actively you engage in productive programming.

Auto-completions

Auto-completions are another fundamental part of any project. Imagine having a huge variable in your project (maybe like DJANGO_RESTFRAMEWORK_PERMISSIONS) and having to retype it every time you have to use it. You’ll surely end up with some bruises on your finger tips. Auto-completions solve exactly this problem (along with a few more).

Among the many auto-completion tools that I have used, Kite has been the most effective. It uses machine learning to analyze your code and then provides suggestions to speed up the development by quite a bit. I also suggest you buy their PRO version which is so good at reading your mind that it sometimes feels like it is your reading mind. You can get it for free if you’re a student from here.

Some python based tools

I will be describing a bunch of problems here and how these tools solve them. Most of them can be installed in virtual environments (except the first one) and then be used without any configuration .

python-poetry: This has been a game changer for me ever since I started to use it. Poetry takes care of virtual environment management for you, and also provides an easy to use interface to add and remove project dependencies. It features advanced dependency resolution (something at which pip isn’t very good at) and can also publish packages for you to PyPI or any other repository.
flake8 and autopep8: These are linters and formatters you can use to make sure your project stays pep8 compliant and has good industry-accepted standards. It also makes it easier to coordinate with team members when all of you follow the same code style.Moreover, VSCode has excellent support for these tools and can be configure to use them easily.
commitizen: Ever found yourself writing a commit message saying fixed stuff and then committing it, only to come back to it in the future and having no idea what that commit did? I think we all can.

Commitizen solves this problem by forcing you to write better commit messages. It follows the conventional-commit standards and uses it to define a set of configurable rules.
pre-commit: Git hook scripts are useful for identifying simple issues before submission to code review. We run our hooks on every commit to automatically point out issues in code such as missing semicolons, trailing whitespace, and debug statements. By pointing these issues out before code review, this allows a code reviewer to focus on the architecture of a change while not wasting time with trivial style nitpicks.

Some VSCode extensions

Kite completion: This is the VSCode extension that can be used to auto complete python code. It also provides HTML, Javascript and Go support.
Settings Sync: This will help you synchronize your VSCode settings between different development devices you use.
Waka Time: This measures the amount of time you spend coding, breaks it down by the project, language, editor used and even time spent per commit (pro feature).
Sourcery: Sourcery via IDE works as an AI-powered coding assistant which helps you write better Python code faster. It works by providing refactoring suggestions on the fly that you can instantly integrate into your code.
Code spell checker: Checks spelling and provides suggestions for incorrect ones. Quite useful especially if you write documentation in VSCode.
Error lens: Though VSCode comes with an inbuilt problem explorer, it’s not really intuitive because it is difficult to open up the problems tab and find the line where your error is and fix it. This extension displays the error right on the line where it is encountered.
VSCode Test Explorer UI: Every good developer writes tests, so this is a really useful tool to have in your tool-belt. It displays a breakdown of all your tests and provides buttons to execute them on the fly.

Finishing up

So there’s the list of tools I use for my python development. I hope they help you as much as they help me. Thank you for taking the time to read this and hope to see you soon!