Up until now, you've been running your tests manually. You write a feature, run pytest, see the green dots, and push to your repository. That works fine for a solo project, but the moment you collaborate with anyone else—or even just move between a laptop and a desktop—you'll realize that "it worked on my machine" is a dangerous phrase. That's where Continuous Integration (CI) comes in. It's basically a robot that runs your tests every single time you push code, ensuring you didn't accidentally break something in the process.

Our Target: A Simple Currency Converter

To keep this concrete, let's imagine we're building a small library called fx_calc that handles currency conversions. I've already got a basic function that takes an amount and a rate, and a corresponding test suite. Here is the structure of the project I'm working with:

fx_calc/
├── fx_calc/
│   └── converter.py
├── tests/
│   └── test_converter.py
└── requirements.txt

The converter.py file is simple, but the test_converter.py file is where the magic happens. I'm using pytest because it's the industry standard for a reason—it's less boilerplate than unittest. Now, I want to make sure that whenever I push a change to GitHub, these tests run automatically.

Automating the Boring Stuff with GitHub Actions

I'm using GitHub Actions for this because it's integrated right into the repo. To get started, I need to create a specific directory structure: .github/workflows/. Inside that, I'll create a file called ci.yml. This YAML file is essentially a recipe that tells GitHub, "Here is the environment I need, and here are the commands I want you to run."

Here is my first attempt at the configuration:

name: Python CI

on: [push, pull_request]

jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - name: Set up Python
        uses: actions/setup-python@v4
        with:
          python-version: '3.10'
      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -r requirements.txt
      - name: Run tests
        run: pytest

Wait, Why Did the Build Fail?

I pushed the code, felt confident, and then... a red "X". The build failed. I checked the logs, and I saw a classic ImportError: No module named 'fx_calc'.

Here's what happened: I told the CI robot to install the dependencies listed in requirements.txt (which includes pytest), but I forgot that the fx_calc code itself isn't a "dependency" in that file—it's the project I'm currently building. When pytest runs in the GitHub environment, it doesn't automatically know that the root folder is a package it should be looking at.

I've made this mistake a dozen times. I was thinking about the environment in terms of "what libraries do I need?" rather than "how does Python find my local code?".

Getting the Green Checkmark

To fix this, I have two choices: I could mess around with PYTHONPATH, or I could do it the professional way and install my current project in "editable" mode. By adding pip install -e . to my installation step, I tell Python to treat the current directory as an installed package.

I updated the "Install dependencies" section of my ci.yml to look like this:

      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -r requirements.txt
          pip install -e .

I pushed the change, waited about 40 seconds, and there it was: the beautiful green checkmark. Now, if I accidentally change a formula in converter.py that breaks my tests, GitHub will scream at me before that code ever makes it into the main branch. That's the peace of mind CI provides.

📋 Practical Task

Implementing a CI Pipeline for a Markdown-to-HTML Parser

You have been handed a small project called md_parse. It contains a function that converts basic Markdown strings to HTML. The project structure is as follows:

md_parse/
├── md_parse/
│   └── parser.py
├── tests/
│   └── test_parser.py
└── requirements.txt

The requirements.txt file already contains pytest. Your task is to create a GitHub Actions workflow file (.github/workflows/ci.yml) that:

• Triggers on every push and pull_request.
• Uses ubuntu-latest as the runner.
• Sets up Python version 3.11.
• Installs the dependencies from requirements.txt.
• Installs the md_parse project itself so that the tests can import it.
• Executes the test suite using pytest.

Write the complete YAML configuration for this workflow.