PHP & DevOps training and testing project

This is a PHP & DevOps training and testing project. The setup is intended to gauge the knowledge of the test taker into various skills needed for PHP API development & DevOps tasks.

Stack

The stack used is:

• Mezzio-Laminas for the PHP API
• Docker, minikube & Kubernetes for DevOps tasks
• GitHub Actions for CI/CD purposes

Task

For information about the task at hand, see the goal documentation.

Docker structure

Each type of container (nginx, php-fpm) has a separate directory. Each directory contains various layers:

• base: this is the base layer for each image and focuses mainly on dependencies and configuration that is applicable for all environments.
• dev: this is an image that extends base and contains additional instructions for development.
• production: this is an image that extends base and contains additional instructions for production environments.

Setup / getting started

Setup / getting started is easy.

1. Checkout the repository on your device
2. Run docker compose up -d to run the containers.
3. Run bin/composer install to install Composer dependencies.
4. Open localhost/api/ping

Running linting and testing:

Linting: bin/phpcs

Testing: bin/phpunit

API documentation

Getting the translation matrix

The translation matrix can be found on /api/translation-matrix. The expected response body is as follows:

{
  "matrix": [
    {
      "source": "human",
      "targets": [
        "labrador",
        "poodle",
        "parrot",
        "parakeet"
      ]
    },
    {
      "source": "labrador",
      "targets": [
        "poodle",
        "parrot"
      ]
    },
    {
      "source": "poodle",
      "targets": [
        "labrador",
        "parrot"
      ]
    },
    {
      "source": "parrot",
      "targets": []
    },
    {
      "source": "parakeet",
      "targets": [
        "parrot"
      ]
    }
  ]
}

Requesting translations for user input

To request translations for user input, make a POST call to /api/translate.

Required header: content-type: application/json

Required body:

{
  "sourceLanguage": "auto",
  "targetLanguage": "parrot",
  "input": "Hello there!"
}

The expected response body is as follows:

{
  "sourceLanguage": "auto",
  "targetLanguage": "parrot",
  "input": "Hello there!",
  "output": "Ik praat je na: Hello there!"
}

Semi-production environment

To run a semi-production environment (deploy to Minikube), run the following scripts:

1. /scripts/start-minikube.bash > This will start a local Minikube cluster.
2. Request an ImagePullSecret from the repository owner and create the supplied secret in the default namespace with the name github-registry.
3. /scripts/deploy-to-production.bash > This will perform a deployment to the local Minikube cluster.
4. /scripts/forward-for-production.bash > This will output the correct URL for your local device/local Minikube cluster, and show some developer information to help you perform requests against this environment.