Four Essential Tips for Writing Modern Python

To leverage the latest and most powerful features of Python, focus on these four key areas.

Python type hinting

Python's newly introduced type hinting syntax enables linters and third-party code quality tools to analyze your code before it runs, identifying potential errors early on. The more you write Python code for others to use, the more beneficial type hints become for everyone involved.

Each new version of Python introduces more advanced and powerful type annotations. By familiarizing yourself with type annotations now, you'll be better prepared to utilize new type hinting features as they are released.

It's crucial to remember that type hints are optional. Not every project requires them. Use type hints to clarify larger projects, but you can skip them for small, quick scripts. While type hints aren't enforced at runtime, you can use Pydantic to achieve runtime enforcement. Many popular Python projects, like FastAPI, extensively use Pydantic.

Python virtual environments and dependency management

For simple projects or basic development tasks, Python's built-in venv tool is often sufficient to isolate projects and their dependencies. However, recent advancements in Python tooling offer more sophisticated options:

• Pyenv: This tool is invaluable for managing multiple Python versions to meet various project requirements. It allows you to switch between Python versions globally or on a per-project basis, making it ideal for working directly at the command line with different Python editions. Although there's no official Windows support, an unofficial port is available.

• Pipenv: Marketed as "Python dev workflow for humans," Pipenv handles both virtual environment management and project dependencies. It ensures deterministic dependencies, meaning you get the exact versions needed, working in the specified combination. However, Pipenv does not manage packaging, making it less suitable for projects destined for PyPI or sharing with others.

• Poetry: Building on Pipenv's functionality, Poetry manages project dependencies and simplifies deployment to PyPI. It also handles virtual environments separately from your project directories, streamlining the development process.

• PDM (Python Development Master): A cutting-edge tool similar to Poetry and Pipenv, PDM offers a unified interface for setting up projects, managing dependencies, and creating distribution artifacts. PDM utilizes the PEP 582 standard for local package storage, eliminating the need for per-project virtual environments. Given its novelty, ensure PDM fits your needs before using it in production.

• Hatch: This tool covers project setup, management, and packaging for PyPI redistribution. Hatch also supports testing and other essential functions, making it a comprehensive solution for Python development.

• UV: An experimental project from the creators of the ruff Python linting tool, uv aims to replace pip, venv, and other command-line tools with a single, Rust-based solution for speed. Its commands are familiar to users of the tools it replaces, easing the learning curve.

For new projects intended for team collaboration or distribution (e.g., via PyPI), adopt the modern pyproject.toml format for specifying requirements and project configuration. While you can still use requirements.txt alongside pyproject.toml, the latter supports a broader range of use cases and ensures your projects are forward-compatible.

New Python Syntax

Python's continuous evolution has introduced several new syntactic features, making the language more powerful and concise. While these additions are not mandatory, they are increasingly utilized by third-party modules, so familiarizing yourself with them is beneficial.

Here are three significant recent syntax additions:

Pattern Matching

A major enhancement in Python 3.10 is structural pattern matching, which goes beyond the traditional "switch/case" statement seen in other languages. This feature allows you to make control-flow decisions based on the structure and content of objects. Essentially, it enables matching based on types or the shapes of types (for example, a list containing an integer and a string) rather than just values.

The Walrus Operator

Introduced in Python 3.8, the walrus operator (`:=`) allows for assignment expressions, enabling you to assign a value to a variable and simultaneously use that variable in an expression. This feature reduces verbosity in common scenarios, such as evaluating a function’s return value while keeping the result for further use.

Positional-Only Parameters

A more subtle but useful addition, positional-only parameters allow you to specify which function parameters must be given positionally, not as keyword arguments. This helps enhance code clarity and facilitates easier future development—objectives shared by many of Python's recent features.

Python Testing

Writing tests for your codebase is essential, akin to daily flossing: universally acknowledged as beneficial, yet often neglected or improperly done. Modern Python projects should include test suites, and the latest testing tools make this easier than ever.

Python’s built-in testing framework, unittest, serves as a decent default, but its design is somewhat outdated. The Pytest framework has become a popular alternative due to its flexibility (you can declare tests anywhere in your code) and minimal boilerplate. Additionally, Pytest offers numerous plugins to enhance its capabilities, including support for asynchronous code testing.

Code coverage is another crucial aspect of testing, indicating how much of your code is exercised by tests. The Coverage module handles this well, and Pytest includes a plugin to integrate seamlessly with it.