A Fast overview of uv — the brand new child on the block of Python bundle managers

For information scientists and Python programmers, pip wants no introduction. As a bundle supervisor, it’s both the go-to answer or the start line in our seek for the very best answer on the market.

Trivia alert 🚨: Whereas pip doesn’t require an introduction, I’ve solely simply discovered that it truly stands for “Pip Installs Packages” or “Most popular Installer Program”.

pip is just not the one bundle supervisor out there. Beneath, you could find the most well-liked instruments available on the market:

• pip — the usual bundle supervisor for Python, which additionally comes preinstalled with Python.
• conda — a bundle and setting administration system designed particularly for information science. Apart from Python, it may additionally set up packages written in different languages (for instance, R).
• poetry — it goals to simplify Python dependency administration by offering options reminiscent of environment friendly dependency decision and digital environments.
• pipenv — this device combines pip for bundle set up with virtualenv for creating remoted environments.

And as this XKCD comedian illustrates, there may be all the time room for a brand new joiner!

In the present day, we’ll look into uv, which boasts being over 100 instances quicker than pip. uv features as a Python bundle installer, digital setting creator and a resolver. To realize its blazing velocity, it was in-built Rust. Moreover, it was designed as a drop-in alternative for pip and pip-tools workflows.

That sounds promising! Earlier than placing it to the check, let’s briefly briefly contact upon the distinction between an installer and resolver:

• bundle installers (e.g., pip) are used to put in, replace, and take away Python packages from the environment.
• bundle resolvers (e.g., pip-tools) make sure the consistency and reproducibility of dependencies by producing exact lists and resolving model conflicts.

Earlier than we dive into testing, it’s price mentioning that I’m conducting the exams on an M1 Mac Mini. And a common disclaimer…