Python 🐍

Keep it simple. New project? Go to the project directory and run these.

python -m venv .venv
source .venv/bin/activate
python -m pip install -U pip
python -m pip install -r requirements.txt

Ez.

In general, use libraries that have frequent updates, are actively and well-maintained, and typically these have a large or growing community. Good libraries of recent have used Rust to implement the core logic in a fast and secure way, and then expose a Python API to the user, so that all the good stuff works, and the annoying stuff is under the hood. [Tokenizers](https://github.com/huggingface/tokenizers) is a good example of this pattern.

These are good for most Python projects. I keep them handy for just about every requirements.txt I work start.

• [rich](https://github.com/willmcgugan/rich) for pretty printing.
• [click](https://github.com/pallets/click) for building command line interfaces.
• [tqdm](https://github.com/tqdm/tqdm) for progress bars.
• [ruff](https://github.com/astral-sh/ruff) for linting, formatting, import sorting,and type checking. I no longer use [black](https://github.com/psf/black) for formatting, or isort for import sorting, because it's good enough to just use ruff for that.
• [pytest](https://github.com/pytest-dev/pytest) for testing.
• [pydantic](https://github.com/pydantic/pydantic) for data validation.
• [polars](https://github.com/pola-rs/polars) for data manipulation. Like pandas on crack.
• Stop using requests and start using aiohttp+asyncio to get async fast HTTP requests. You may be shocked at how efficient it is.
  • [aiohttp](https://github.com/aio-libs/aiohttp) for HTTP.
  • [asyncio](https://docs.python.org/3/library/asyncio.html) for async programming.