This starter kit is designed to allow you to create very fast your Flask API.

The primary goal of this project is to remain as unopinionated as possible. Its purpose is not to dictate your project structure or to demonstrate a complete sample application, but to provide a set of tools intended to make back-end development robust, easy, and, most importantly, fun.

This starter kit comes with a tutorial. Check it out if you want a quick tutorial on how to use Flask with this architecure.

1. Dependencies
2. Getting Started
3. Commands
4. Database
5. Application Structure
6. Development
7. Testing
8. Lint
9. Swagger

You will need docker and docker-compose.

First, clone the project:

$ git clone https://github.com/antkahn/flask-api-starter-kit.git <my-project-name>
$ cd <my-project-name>

Then install dependencies and check that it works

$ make install      # Install the pip dependencies on the docker container
$ make start        # Run the container containing your local python server

If everything works, you should see the available routes here.

The API runs locally on docker containers. You can easily change the python version you are willing to use here, by fetching a docker image of the python version you want.

While developing, you will probably rely mostly on make start; however, there are additional scripts at your disposal:

The database is in PostgreSql.

Locally, you can connect to your database using :

$ make db/connect

However, you will need before using this command to change the docker database container's name here.

This kit contains a built in database versioning using alembic. Once you've changed your models, which should reflect your database's state, you can generate the migration, then upgrade or downgrade your database as you want. See Commands for more information.

The migration will be generated by the container, it may possible that you can only edit it via sudo or by running chown on the generated file.

The application structure presented in this boilerplate is grouped primarily by file type. Please note, however, that this structure is only meant to serve as a guide, it is by no means prescriptive.

.
├── devops                   # Project devops configuration settings
│   └── deploy               # Environment-specific configuration files for shipit
├── migrations               # Database's migrations settings
│   └── versions             # Database's migrations versions generated by alembic
├── src                      # Application source code
│   ├── models               # Python classes modeling the database
│   │   ├── abc.py           # Abstract base class model
│   │   └── user.py          # Definition of the user model
│   ├── repositories         # Python classes allowing you to interact with your models
│   │   └── user.py          # Methods to easily handle user models
│   ├── resources            # Python classes containing the HTTP verbs of your routes
│   │   └── user.py          # Rest verbs related to the user routes
│   ├── routes               # Routes definitions and links to their associated resources
│   │   ├── __init__.py      # Contains every blueprint of your API
│   │   └── user.py          # The blueprint related to the user
│   ├── swagger              # Resources documentation
│   │   └── user             # Documentation of the user resource
│   │       └── GET.yml      # Documentation of the GET method on the user resource
│   ├── util                 # Some helpfull, non-business Python functions for your project
│   │   └── parse_params.py  # Wrapper for the resources to easily handle parameters
│   ├── config.py            # Project configuration settings
│   ├── manage.py            # Project commands
│   └── server.py            # Server configuration
└── test                     # Unit tests source code

To develop locally, here are your two options:

$ make start           # Create the containers containing your python server in your terminal
$ make daemon          # Create the containers containing your python server as a daemon

The containers will reload by themselves as your source code is changed. You can check the logs in the ./server.log file.

To add a unit test, simply create a test_*.py file anywhere in ./test/, prefix your test classes with Test and your testing methods with test_. Unittest will run them automaticaly. You can add objects in your database that will only be used in your tests, see example. You can run your tests in their own container with the command:

$ make tests

To lint your code using flake8, just run in your terminal:

$ make lint

It will run the flake8 commands on your project in your server container, and display any lint error you may have in your code.

Your API needs a description of it's routes and how to interact with them. You can easily do that with the swagger package included in the starter kit. Simply add a docstring to the resources of your API like in the user example. The API description will be available here.