social_network_fastAPI

Social Network fastAPI

       

This is a full API for simaple social network develop with FastApi Postgresql and MongoDB in microservices architecture, deployed as containerized application with Docker You can view the demo run on Ubuntu here You can get the project images here

---

Table of Contents

• Social Network fastAPI
• Overview
• Getting Started
  • Docker-compose
  • Kubernetes
  • systemd
• Features
• Authentication & Security
• Tests
• CI CD
• Database & Alembic
• env

---

Overview

This is a full API for simaple social network which was developed in microservices architecture, The architecture looks something like this:

Example of docker-compose run on Ubuntu machine. I am not a professional developer, I have background in the IT field and recently I moved to devops The architecture could have been better and so the development itself

Getting Started

You can view the Getting Started presentation Here

Docker-compose

Kubernetes

systemd

Features

Currently the project contains the following topics: Users There is a registration and login, the registration is verified by email verification, and the login is based on JWT. Most features are conditional on a authenticated user and verified user. There is email verification for new users Posts The user can create a post visible to everyone or in a specific group. Can update or delete, and filter posts of your choice. Votes Each user has the option to like the post, and can cancel the like Comments Each user has the option to add a comment to the post. The user can update or delete the comment. Groups Groups is the most multi-function feature, Any verified user can create and manage a group, Users can send a join request, the onwer of the group can choose whether to confirm or not. The administrator of the group can block a user from the group, can exclude users, or can replace himself with another member of the group

Authentication & Security

I decided to gain experience with Keycloak, so user authentication is performed with Keycloak and postgresql, I'm not sure if it fits the requirement for this kind of software, but for the purpose of studying I used it. I used the fastapi-keycloak package, and Keycloak 16.0.1, it's a bit outdated, but it's what fits the package right now. Because I didn't want user communication with Keycloak (That's why I said it might not be suitable for the project), For some of the requirements I did not find a way to work with the keycloak API, action-token for example, So I added some features with JWT. CORS (Cross Origin Resource Sharing) is implemented Using environment variables, you can see more in the .env section.

Tests

There are currently 335 tests, implemented with pytest, Each test is isolated and independent of another, using @pytest.fixture There is a separate database for tests to maintain a proper test, which is initialized after each test For data validation I used Field and validator imported from Pydantic And Query Path and Body imported from fastapi You can see more in validators.py and schemas.py Files

CI CD

The project has a simple CI/CD based on GitHub Actions. For each push or pull request, a build is performed for a test environment with all the dependencies, If the build is completed successfully, two security jobs start running, The first is git-guardian-scanning - uses gitguardian's feature for scanning The second is trufflehog-credential-verification - Credential Verification Scanning If these two jobs are completed successfully, three jobs for deploy start running, The first is deploy-to-heroku the second is deploy-to-ubunto-server and the third is update-docker-image The last job that runs is notification, This is a custom email that gives status update on the other jobs All jobs use environmental variables, which are explained in the .env section

Database & Alembic

Postgres

Mongodb In the email_verification_server I used mongodb and there are 1 DB and 1 collection, Use for track users who need to verify email

env

A file containing the environment variables: For These environment variables, it is required to set up a database first: DATABASE_HOSTNAME= DATABASE_PORT= DATABASE_PASSWORD= DATABASE_NAME= DATABASE_USERNAME= These environment variables required for create token, password hash and expiration for token: SECRET_KEY= ALGORITHM= ACCESS_TOKEN_EXPIRE_MINUTES= Note that there is a separate test database that also needs to be set up, At the moment his name is {settings.database_name}_test (like the main only with _test at the end) You can change its definition here: /tests/conftest.py To use Workflow for GitHub Actions, you required to include additional environment variables in GitHub Secrets