plm

Helps you manage python libraries and environments

Quickstart

pip install git+https://git.io/plm.git

Create and activate an environment:

plm
# equivalent to: plm create && plm activate

Then, you can install a package:

plm install requests numpy
# also works: plm i requests numpy

Now requests and numpy will be installed in your venv/ and they will also be added to requirements-dev.txt and requirements.txt.

Why would I need plm?

Traditionally, to create an environment and activate it:

python -m venv venv
source venv/bin/activate

Then, to install a library and track dependencies:

pip install requests numpy
echo -e "requests\nnumpy" >> requirements-dev.txt
pip freeze > requirements.txt

Instead, with plm, you just need to do this:

plm create
plm activate
plm i requests numpy

Furthermore, it's easier to manage a project dependencies for GitHub projects. If you used plm to manage a project and pushed it to a repo username/project, then someone can start using your project using the following commands:

git clone https://github.com/username/project.git
cd project/
plm
plm i

it is essentially equivalent to this:

git clone https://github.com/username/project.git
cd project/
python -m venv venv
/bin/bash --rcfile venv/bin/activate  # slighly different from source <script>
pip install -r requirements-dev.txt  # or requirements.txt

Managing virtual environments

You can customize the name of the venv:

plm create myvenv
plm activate myvenv

Installing libraries

plm install has the shorthand plm i. There's many ways to install packages:

plm install requests
plm i pandas==1.*
plm i numpy Pillow

You can also install from a file:

plm i -r requirements.txt

If you don't specify anything by running

plm i

plm will install the libraries specified in requirements-dev.txt, or fall back to requirements.txt if the former does not exist.

requirements-dev.txt vs requirements.txt

requirements-dev.txt loosely keeps track of your packages, which is good for readability and upgradability. It will be updated every time you pip install a package, e.g. if you run plm install <package 1> <package 2> then it will look like this:

<old package 1>
...
<old package N>
<package 1>
<package 2>

requirements.txt on the other hands keeps track of the exact requirements (e.g. for deploying to a server). It's the output of pip freeze > requirements.txt.

Note

plm's not meant as a replacement for pip or venv, it simply wraps usual/repetitive commands in a easy-to-remember CLI. The codebase is < 200 lines, which makes it ideally for forking and extending. If you are planning to do more serious environment managment, please use conda, which also has conda create and conda activate commands and has significantly more features. If you are planning to do serious dependencies management, use pipenv instead.

Acknowledgement

Commands like create, activate are inspired from conda. install (or i) is inspired from npm.