Testing

Let's build an action on github that runs automatic tests whenever a push is made to the main branch.

First, we will need to build the testing infrastructure in a repository.

Create Repository

Create Folders

├── .gitignore
├── README.md
├── requirements.txt
├── src/
└── tests/

Let's create a lambda function similar to those created in previous classes. It will be a simple word processing function.

def word_count_handler(event, context):
    """
    A simple function that counts
    the number of words in a string
    """
    msg = event["body"]

    n_words = len(msg.split())

    return {
        "len": len(msg),
        "words": n_words,
    }

├── .gitignore
├── README.md
├── requirements.txt
├── src
│   └── word_count.py
└── tests/

Add Testing

Let's add unit tests to our project. Initially we will run the tests locally. Then, we will automate this task to be executed in a github action.

[pytest]
pythonpath = src

import pytest
import sys
import word_count as wc


def test_simple_text_count():
    """Tests the word len and word count of a simple text"""
    event = {"body": "Hello World"}
    expected = {
        "len": 11,
        "words": 2,
    }

    # Test if return is as expected
    assert wc.word_count_handler(event, None) == expected

├── .gitignore
├── pytest.ini
├── README.md
├── requirements.txt
├── src
│   └── word_count.py
└── tests
    └── test_word_count.py

$ pytest
== test session starts ===
platform linux -- Python 3.10.12, pytest-7.4.0, pluggy-1.2.0
rootdir: /path/to/directory
configfile: pytest.ini
plugins: anyio-3.7.1
collected 1 item                                                                                                     

tests/test_word_count.py .                                                                                     [100%]

=== 1 passed in 0.00s ====

Automating

In an environment where several Data Scientists work on the data team's various projects, it is desirable that there are automatic verification routines whenever modifications are sent to the repository.

Let's see how to automate the execution of tests using github actions.

Create Github Action

Let's create a directed acyclic graph (DAG) or pipeline with the sequence of steps to verify the tests in our repository.

The actions to be performed are stored in the .github/workflows folder in the repository root and are represented in YAML format.

name: An example of an automatic testing action
on:
  push:
    branches:
      - main
jobs:
  build-and-test:
    runs-on: ubuntu-latest
    steps:

      - name: Checkout code
        uses: actions/checkout@v4

      - name: Set up Python
        uses: actions/setup-python@v4
        with:
          python-version: '3.10'

      - name: Install dependencies
        run: pip install -r requirements.txt

      - name: Run tests
        run: pytest

├── .github
│   └── workflows
│       └── test_workflow.yaml
├── .gitignore
├── pytest.ini
├── README.md
├── requirements.txt
├── src
│   └── word_count.py
└── tests
    └── test_word_count.py

Explaining workflow YAML:

• First, we provide a description for the workflow with:

  name: An example of an automatic testing action
  

• We set the action to be run whenever there is a push on the main branch when doing:

  on:
  push:
      branches:
      - main
  

• To define a job, which is a group of steps that are executed together as part of a workflow run, we do:

  jobs:
    build-and-test:
      runs-on: ubuntu-latest
      steps:
  

  Note that we defined that the task must be executed in an ubuntu container.

• Then we define a sequence of steps for our job:
  • To bring code from the repository into the container:

    - name: Checkout code
      uses: actions/checkout@v4
    

  • To set up Python:

    - name: Set up Python
      uses: actions/setup-python@v4
      with:
        python-version: '3.10'
    

  • To install dependencies

    - name: Install dependencies
      run: pip install -r requirements.txt
    

  • To run tests:

    - name: Run tests
      run: pytest
    

Check if it worked

Let's check if the action is executed correctly! Before everything:

$ git add .
$ git commit -m "Try testing with gh actions"
$ git push

Tasks

def test_no_body():
    event = {}
    expected = {"error": "no body"}
    assert wc.word_count_handler(event, None) == expected

$ pytest