Spotifiuby Messages Server

Installing the project

The only dependency required to use this template is poetry. The recommended method to install it is through pip.

$ pip3 install poetry
$ poetry config virtualenvs.in-project true

Remember to commit to the repo the poetry.lock file generated by poetry install.

Dependencies

The virtual environment is automatically created and activated via poetry.

$ cd to-project-path
$ poetry install

To make sure everything is set up correctly, run the following command which must show the virtual environment path:

$ poetry show -v

If still having problems, you can run source command to enable the virtual environment from the shell:

$ poetry show -v
Using virtualenv: /home/<User>/<path>/taller2/fastapi-template/.venv
$ source .venv/bin/activate

Adding new dependencies

Check the full poetry docs, but here goes a quick reminder,

poetry add <dependency> [--dev]

Style guide

This template follows PEP8.

For this purpose, we use:

• black: an opinionated code formatting tool
• flake8: a tool to enforce style guide
• pylint: a source code, bug and quality checker

Linters

flake8 && pylint <module_name>

Formatter

black .

Running the server

• Development: uvicorn src.main:app --reload
• Production: uvicorn src.main:app

Docker

You need docker-compose and docker to run the following containers and commands.

Developing with container and database

$ sudo ./scripts/test-container.sh

Running tests within container

You have two options here, one is simple running:

$ sudo ./scripts/coverage-container.sh

which will execute all the tests and will exit the container afterwards.

Alternately, you can also run a container which will run the app and will provide the database. In one terminal run:

$ sudo ./scripts/test-container.sh

Then in other terminal run docker ps and copy the container ID from the docker_fastapi-server image:

$ docker ps
CONTAINER ID   IMAGE                   COMMAND                  CREATED              STATUS              PORTS                                       NAMES
b1e9c7c4e040   docker_fastapi-server   "./docker-entrypoint…"   About a minute ago   Up About a minute   0.0.0.0:8082->8082/tcp, :::8082->8082/tcp   docker_fastapi-server_1
620c0b75ce2a   mongo                   "docker-entrypoint.s…"   About a minute ago   Up About a minute   0.0.0.0:5438->5432/tcp, :::5438->5432/tcp   docker_postgres_1

Finally, enter the container with docker exec -it [container-id] bash and run the tests within the container using poetry run pytest:

$ docker exec -it [container-id] bash
$ root@b1e9c7c4e040:/code# ls
docker-entrypoint.sh  poetry.lock  pyproject.toml  src  tests  scripts
$ root@b1e9c7c4e040:/code# poetry run pytest

API Documentation

Documentation will be automatically generated at {app}/docs

Tests

We use the pytest framework to test. The easiest way to run tests is pytest. Remember to create functions with a name that starts with test_ (this is standard pytest conventions).

Github Actions

A few pipelines have been set to run on github actions to ensure code quality and deployment.

• Run Linter
• Run Tests
• Upload Test Coverage
• Deploy to Heroku using docker image

Upload Coverage to Codecov

The pipeline automatically generates a coverage report and uploads it to codecov

You'll need to set the following actions secrets:

• CODECOV_TOKEN: Repo Token. Can be obtained on codecov when setting up or on settings

Heroku

You'll need to set the following actions secrets:

• HEROKU_APP_NAME: App name
• HEROKU_EMAIL: Account email
• HEROKU_API_KEY: Account API key

MongoDB

This server uses MongoDB as its database. Set MONGO_URL as an environment variable and action secret to connect to the database.

Firebase

To access the database and storage, you'll need to generate a Firebase private key.

To do so, go to Project configuration > Service accounts > Generate new private key. [Link]

Save the file as google-credentials.json in the root directory of the repository.

You can also set TESTING=1 as an environment variable to use mocks of the database and storage for testing purposes.

In order to load the credentials in Heroku, set GOOGLE_CREDENTIALS as an environment variable in Heroku, and paste the content of the google-credentials.json file.

Datadog

The heroku Dockerfile includes the DataDog agent. Create a new DataDog API Key from here. Also, you need to set the following config vars in Heroku (you can use Heroku CLI if you want):

DD_API_KEY=<api_key_from_datadog>
DD_DYNO_HOST=false
HEROKU_APP_NAME=<app_name>
DD_TAGS=service:<meaningful_tag_for_datadog>