Intro

Sample FastAPI app to demostrate how to use FastAPI

About

This is a GitHub Score Application to rank repositories based on forks and stars.

More info (Issues):

• https://github.com/omidraha/score/issues

Stack

In this application we used:

• Python 3.9.7
• FastAPI 0.68.1
• Docker 20.10.8
• Docker-compose 1.29.2
• Docker-compose file version: 3.4
• Gunicorn 20.1.0
• uvicorn 0.15.0

Images

• tiangolo/uvicorn-gunicorn-fastapi:python3.9

More info:

• https://fastapi.tiangolo.com/deployment/docker/#official-docker-image-with-gunicorn-uvicorn

Backend topology

gunicorn -> gunicorn worker -> uvicorn -> main.py

More info:

• https://www.uvicorn.org/deployment/#gunicorn
• https://www.uvicorn.org/#running-with-gunicorn

Environments

Before run the app or run their tests, copy example.env as new .env file:

$ cp example.env .env

Set the value of GITHUB_ACCESS_TOKEN to your valid GITHUB access token.

Run application

Set Environments as previous section and then run the application:

$ docker-compose -p score build
$ docker-compose -p score up

For run in daemon, use -d if it's required:

$ docker-compose -p score up -d

Run tests

Set Environments as previous section and then:

$ docker-compose -p test run --rm web pytest

Health check

To see health check of service, use this command:

$ docker-compose -p score ps

Sample output:

score_web_1   /start.sh   Up (healthy)   0.0.0.0:80->80/tcp,:::80->80/tcp

To see logs of health check:

docker inspect --format='{{json .State.Health}}' score_web_1

Browse

http://127.0.0.1/docs/

http://127.0.0.1/score/

Curl

$ curl -X 'POST' 'http://127.0.0.1/score/' -H 'accept: application/json' -H 'Content-Type: application/json' -d '{"repo":"python/cpython"}'

TODO

• Cache external GitHub service API response for some short period of the time for example for 1 hour in the Redis, And maybe provide a new cache option for example Cache-Control: no-cache header to the score API

• Use async library for communicate with Github service API

Extra useful links

• https://fastapi.tiangolo.com/tutorial/metadata/
• https://levelup.gitconnected.com/deploying-an-asynchronous-fastapi-on-nginx-unit-b038288bec5
• https://dev.to/shuv1824/deploy-fastapi-application-on-ubuntu-with-nginx-gunicorn-and-uvicorn-3mbl
• https://medium.com/analytics-vidhya/how-to-deploy-a-python-api-with-fastapi-with-nginx-and-docker-1328cbf41bc
• https://stackoverflow.com/questions/67435296/how-to-access-fastapi-swaggerui-docs-behind-an-nginx-proxy
• https://fastapi.tiangolo.com/advanced/behind-a-proxy/
• https://www.reddit.com/r/flask/comments/j2289k/anyone_using_fastapi_in_production/
• https://stackoverflow.com/questions/62976648/architecture-flask-vs-fastapi/62977786#62977786
• https://github.com/tiangolo/uvicorn-gunicorn-fastapi-docker
• https://fastapi.tiangolo.com/deployment/manually/
• https://unit.nginx.orga
• https://www.uvicorn.org/#running-with-gunicorn
• https://github.com/encode/uvicorn
• https://towardsdatascience.com/how-to-deploy-a-machine-learning-model-with-fastapi-docker-and-github-actions-13374cbd638a
• https://fastapi.tiangolo.com/tutorial/first-steps/
• https://unit.nginx.org/
• https://fastapi.tiangolo.com/advanced/behind-a-proxy/
• https://fastapi.tiangolo.com/deployment/manually/
• https://www.uvicorn.org/#running-with-gunicorn
• https://github.com/tiangolo/uvicorn-gunicorn-fastapi-docker
• https://stackoverflow.com/a/63427961
• https://stackoverflow.com/a/66426744
• https://levelup.gitconnected.com/deploying-an-asynchronous-fastapi-on-nginx-unit-b038288bec5
• https://dev.to/shuv1824/deploy-fastapi-application-on-ubuntu-with-nginx-gunicorn-and-uvicorn-3mbl
• fastapi/fastapi#1034 (comment)
• https://fastapi.tiangolo.com/tutorial/body/
• https://fastapi.tiangolo.com/tutorial/handling-errors/
• https://gist.github.com/omidraha/72817ed0c6173f6c47613e3eebf03ad7
• https://fastapi.tiangolo.com/deployment/docker/
• https://github.com/tiangolo/uvicorn-gunicorn-fastapi-docker