python-dotenv VS Gooey

I personally just use .env to keep api keys

similar to dotenv, I like this object oriented approach though.

Are you working on a Python project that uses pdm for dependency management and dotenv for local environment variable and secrets management? Do you find it frustrating when CLI tools like pgcli don't automatically pick up your .env file, forcing you to resort to npm install -g dotenv-cli? I've got a more convenient solution for you!

Similarly, I use python-dotenv which reads environment variables, or variables set in a .env file. Then in your settings.py file you can do things like:

Locally you may want to use a .env file that is also in .gitignore and python-dotenv to auto activate them. Also having an .env.template is also a good idea to help others working on the project know what they need to set in order for things to work.

But there's no web UI to set them for your local dev version. But there are various Python modules that will read environment variables from a file named .env. I like https://github.com/theskumar/python-dotenv myself. So try using that - create a .env file for your local dev site and use dotenv instead of os.getenv() to read the environment variables.

python-dotenv

The name conflicts with a similar python module that allows one to turn any Python console program into a GUI with one line.

https://github.com/chriskiehl/Gooey

Hey! Cool project! I have a question: why do you dump out sys.argv to a local file in the CWD? [0] tmp.txt is hardly a unique name… or am I missing something and this never triggers?

[0] https://github.com/chriskiehl/Gooey/blob/be4b11b8f27f500e732...

This might be of interest to you:

> Gooey - Turn (almost) any Python 3 Console Program into a GUI application with one line

https://github.com/chriskiehl/Gooey

I love gooey: https://github.com/chriskiehl/Gooey

It allows me to quickly slap a GUI on an existing script that accepts command-line-arguments. In the end, I get the best of both world: Discoverability from the GUI, automation through the script, and automatic feature parity between the two.

Downside: Control over the GUI layout is basic, and only "standard" GUI features work, but I never felt limited when using it.

similar for local/individual usage:

https://github.com/chriskiehl/Gooey - take a python-CLI, make a TK-windows

and then probably even simple dashboarding like streamlit.

Like Gooey?

As for how I'd handle it, I would probably combine both into one command-line program, with all filepaths using pathlib and letting argparse handle the options. That way you could optionally generate a GUI with gooey with little extra effort, although this would also add a dependency.