`dispenv` – Disposable Python Environments

⚠️
WIP

Need to make an environment to work on a GitHub issue? Want to try out a new package and not leave the clutter of a virtual environment behind? dispenv is here to save the day!

pip install git+https://github.com/pmbaumgartner/dispenv.git

To get started after install, type dispenv create in the parent folder of where you’d like your environment folder to go. Then, you’ll be walked through a series of questions about your environment.

Example:

$ dispenv create

? What python version would you like to use? 3.8
? What type of virtual environment are you creating? (Use arrow keys)
 » conda
   docker (VSCode devcontainer)
? What should the folder be named? ba186b77-46bf-4506-9e8d-f9ba71377471
? What should the environment be named? ba186b77-46bf-4506-9e8d-f9ba71377471
? Paste link to a requirements.txt in a gist.  Optional https://gist.github.com/pmbaumgartner/af700cc2e009da40b1816cae66881e60

When you’re done, dispenv cleanup will remove that folder and clean up any environment artifacts.

Features:

Supports conda and docker virtual environments
- docker environments are setup to be run with VSCode as a devcontainer
When asked, if you paste a link to a GitHub Gist that contains a requirements.txt, dispenv will install those packages inside your environment after creation.
- Requires the to be installed.
- Requires you to be authenticated through the GitHub CLI
dispenv cleanup will automatically delete the conda environment or shut down and remove the containers and images if it’s a docker environment.

Recommendations

Since this is intended to be a global or system-wide tool, I recommend installing in your base environment (if you use conda) or using a tool like pipx.