This is an simple API with a few basic operations concerning people and its salaries.
Make sure to have the port 8000 free since these instructions consider this is the port where the project will run, or use another port, adjusting the suggested commands.
The project's requirements are present in the requirements.txt file and also Python3, which is not present in the requirements file. So, what you will need for this project is Python3, Django, Djangorestframework and psycopg.
Attention: this project was idealized to be ran through Docker, so it is recommended to make sure you have docker and docker-compose in your machine before proceding. With docker installed, just jump this section and go directly to Using Docker below.
Alternatively, using a package manager such as pip is enough to install the dependencies through the pip3 install -r requirements.txt command from the project's root directory. After installing the dependencies, make sure to add your local postgres configuration such as host, port, password and etc. in settings.py file. With that configured, you need to run the database migrations by running python3 manage.py migrate, and after that the project is ready to run. So, run the command
python3 manage.py runserver 0.0.0.0:8000
and it should be enough.
Docker makes it easier to run this project. From the project's root directory simply run
docker-compose up --build
and you should be ready to go in localhost:8000.
Acessing the project's root URL will display a Django 404 page, since there is no endpoint for that. The project has 2 main domain paths, them being the API itself and the Django's admin interface. They are present in /api/v1/ and /admin paths respectively.
Acessing /api/v1/ will display the necessary endpoints for this challenge. There is three, the requested /users and /salaries, and also a /statistics that display statistics about the salaries present in the database. This last one is a optional extra.
The /users endpoint has the full CRUD operations implemented, displaying a list and allowing the creation in its root path. Acessing an specific user through /users/:id makes it possible to retrieve the user with the specified ID and also update/delete it via DRF's native interface.
The retrieved users will display its native fields, such as id, name, cpf and date_of_birth and also the list of its registered salaries, along with some requested statistics about the salaries of this specific user, such as its average salary, average discounts, max and min salaries.
The /salaries endpoint is similar to the /users endpoint in the form of having all CRUD operations available. The difference is it shows no additional informations other than its own fields. The retrieved informations about salaries are id, date, value, discounts and the user it belongs to.
This endpoint is optional and it is a way to implement the same statistics shown in the /users endpoint, but in a global way, meaning it will calculate averages, max and min for every salary present in the database. This endpoint does not have all CRUD operations since it doesn't represent a database entity, and it is dynamically calculated. The only operations implemented for this endpoint is list and retrieve, where list returns a single information: the statistics about every salary present in the database.
Calling the retrieve action will require an ID, and it is the ID of an user. So, accessing /statistics/:user_id will retrieve the same statistics present in the /users endpoint, calculating averages, max and min for this specific user.
This project has a few model and viewsets tests implemented, just as an form to guarantee the minimal features are working. They are not very deep and were implemented as a form of understanding how Django API tests work.
To run the tests using docker, simply run the command docker-compose run web python manage.py test in the project's root directory.