Cleaning Up Python Environment

The one challenge that I have had for many years is working with Python and the local system. So much so that I have had good discussions with many of the pure Python developers at NTC. It often involved my statements along the lines of putting everything inside of a Docker container or something to that effect. I still distinctly remember my first attempt at installing pyenv completely borked my Ansible environment. A very technical term I know.

Now today I was suggested to take a look at using Python UV for managing the Python tools and environments. After taking a few steps on the journey I am convinced that this is the method that I am going to be going on.

Making the Switch #

So the previously used environment consisted of pyenv for managing the versions of Python and Pipx for managing the tools that were installed via Python. Poetry is still the NTC tool of choice for managing Python packages and libraries. So this was a must have to integrate fine with Poetry for the NTC projects. To make the switch happen there is first the uninstall of the current tools and then the instalation of the new system.

Uninstalling Pyenv #

The first step is to remove pyenv from being loaded by any shells that may be using, such as by bash (.bashrc) or ZSH (.zshrc). I found this to be a critical step the hard way. So the first thing on my machines since I’m using ZSH is to start by commenting out the references to pyenv in that file.

vi ~/.zshrc

After commenting out the lines with # I reloaded the shell.

source ~/.zshrc

In my particular configuration I had two sections to comment out. A PATH statement near the top and near the bottom of the file a pyenv specific section.

Once the environment is set to no longer load pyenv, next up is to remove the directory that maintains the pyenv data and removing pyenv that was previously installed via Homebrew.

rm -rf ~/.pyenv

brew uninstall pyenv

❯ brew uninstall pyenv
Uninstalling /opt/homebrew/Cellar/pyenv/2.6.16... (1,407 files, 4.6MB)
==> Autoremoving 3 unneeded formulae:
autoconf
m4
pkgconf
Uninstalling /opt/homebrew/Cellar/pkgconf/2.5.1... (28 files, 532.2KB)
Uninstalling /opt/homebrew/Cellar/autoconf/2.72... (72 files, 3.8MB)
Uninstalling /opt/homebrew/Cellar/m4/1.4.20... (14 files, 802.1KB)

Once the

Removing Pipx #

Next up is to remove Pipx. This was primarily installed for me to install Python Poetry. This will still be installed, just done via Python UV instead. I had done the installation of pipx via HomeBrew:

brew uninstall pipx

Brew Cleanup #

At the end of removing these applications that were installed for me on Homebrew, I’m now running a brew cleanup command to remove the old pieces.

brew cleanup

UV For Tools #

I’m going to skip right past the UV installation steps since the UV documentation handles this nicely. First tool that I’m going to need to install is Python Poetry. The docs suggest to use Pipx, but we just removed that in the line before. The installation is actually quite simple. Just a single command of uv tool install poetry.

uv tool install poetry

❯ uv tool install poetry
Resolved 43 packages in 216ms
Prepared 29 packages in 261ms
Installed 43 packages in 29ms
 + anyio==4.12.0
 + build==1.3.0
 + cachecontrol==0.14.4
 + certifi==2025.11.12
 + cffi==2.0.0
 + charset-normalizer==3.4.4
 + cleo==2.1.0
 + crashtest==0.4.1
 + distlib==0.4.0
 + dulwich==0.24.10
 + fastjsonschema==2.21.2
 + filelock==3.20.1
 + findpython==0.7.1
 + h11==0.16.0
 + httpcore==1.0.9
 + httpx==0.28.1
 + idna==3.11
 + installer==0.7.0
 + jaraco-classes==3.4.0
 + jaraco-context==6.0.1
 + jaraco-functools==4.3.0
 + keyring==25.7.0
 + more-itertools==10.8.0
 + msgpack==1.1.2
 + packaging==25.0
 + pbs-installer==2025.12.17
 + pkginfo==1.12.1.2
 + platformdirs==4.5.1
 + poetry==2.2.1
 + poetry-core==2.2.1
 + pycparser==2.23
 + pyproject-hooks==1.2.0
 + rapidfuzz==3.14.3
 + requests==2.32.5
 + requests-toolbelt==1.0.0
 + shellingham==1.5.4
 + tomlkit==0.13.3
 + trove-classifiers==2025.12.1.14
 + typing-extensions==4.15.0
 + urllib3==2.6.2
 + virtualenv==20.35.4
 + xattr==1.3.0
 + zstandard==0.25.0
Installed 1 executable: poetry

With that Poetry is now installed.

Cookiecutter Installation #

Also as part of the installation in order to get started easily with creating Nautobot Apps is the use of Python CookieCutter. The tool installs in a snap with UV, taking less than half a second to install everything:

uv tool install cookiecutter

Python Versions #

The installation of different Python versions is just as easy. It’s a uv python install x.y where x, y, and z are the version numbers. Such as 3.13. You can also specify the patch version as well.

uv python install 3.13

This took only a few seconds to download and install!

❯ uv python install 3.13
Installed Python 3.13.11 in 578ms
 + cpython-3.13.11-macos-aarch64-none (python3.13)

Summary #

At this point I’m not going back. I hold myself the right to change or if there are other issues that arise. But the use of Python UV to handle the tools and Python software versions is here to state for a bit for me. What do you think? Let me know in the comments.

-Josh