Python modules and packages can be confusing or intimidating at first. This post aims to demystify them once and for all by having you create your own working Python package and modules in under twenty minutes.

Terminology

A Python module is just any old .py file containing definitions and statements. You’ll create two modules, one with a single definition and one with a single statement.

A package is just a way of organizing modules into folders, and importing them as dotted paths. We’ll see how this works as well.

Here are the official docs if you’d like to know more.

Step 1 - Your first Python module

Create a directory to hold our entire package, then navigate inside. Create a file called my_module.py and define a trivial function.

Step 2 - Import and run your code

From inside the same directory as your module, test out different ways of importing and running your function. (You can import the module and use a dotted path to access the function, or you can import the function directly. You can use python -c to execute a command, or use the python interpreter if you prefer.)

You can even write a script that imports your function.

Step 3 - Create a submodule

Now make a new directory inside your package directory. Create another module there, this one containing only an assignment statement.

Step 4 - Get an ImportError

Try importing your submodule using a dotted path. It doesn’t work!

Step 5 - Add init.py

The command line program touch creates empty files. Use it to make a file called __init__.py in your submodule directory. Now the import works!

That’s it

You just made a Python package containing two modules. They don’t do much, but the way they are laid out and imported is no different from any other package you’ll come across.

Final challenge

Navigate up one directory – to the folder containing your package. Try importing my_module from here. Why doesn’t it work? Can you fix it?