juanirm / DistributedReplays

👨‍💻 Rocket League Replay Collection and Analysis Server that powers calculated.gg

Calculated.gg Backend

Setup

• Install Python 3.6/pip
• pip3 install -r requirements.txt
• Install and run Redis with default port + settings (Windows is included, Ubuntu)
• Install postgreSQL (Windows, Ubuntu, Mac)
• Make sure that postgreSQL is in the path
• If only for local development change the password in Ubuntu to postgres
• Ensure you have the latest LTS version of node and npm installed
• Run cd webapp, npm install

Running

Ubuntu

• Run RLBotServer.py.
• cd webapp, npm start

Windows

• Run win_run.bat
• You can optionally kill the python process and start it in an IDE
• You can log into psql command line with psql postgresql://postgres:postgres@localhost
• If the included redis does not work here is install directions
• DO NOT USE IDLE, either use command line or pycharm (pycharm is recommended for development of backend)

Mac

• Run mac_run.sh

You need to run in python3. Mac comes with python2 by default so do not run any commands without the 3

Structure

The structure of the server is split into different directories:

• blueprints - Backend Python code, split into subsections
• database - Database objects and queries
• helpers - Various scripts to help in maintaining the server
• static - Static website files (JS/CSS/images)
• tasks - Contains celery code user for processing the queue of replays
• templates - Dynamic website files (rendered HTML templates)

---

Setup and Running with Docker (Docs WIP)

Overview

Through the usage of Docker and Docker Compose, all application and infrastructure dependencies and configurations are taken care of automatically. The only requirement will be to have docker installed on the machine. Upon initializing Docker Compose the following services will be started:

• Postgres DB
• Redis
• Nginx Load Balancer
• Flask
• Celery
• Node.js development server

Basic Dependencies

• Docker Community Edition (Stable)

  Docker will run Postgres, Redis, Flask, Celery, and Node inside linux based "Containers" on most platforms. Download and install Docker and Docker Compose for your platform:

  • Mac/Windows: Docker Desktop
  • Ubuntu/Debian-ish: Docker CLI

Environment variables (Optional)

Included in the repository there's a env.example file that can be used to customize the environment variables used in Docker Compose. Usage of this file is not mandatory but if you want to leverage on it, you should rename it to .env and edit accordingly.

Run Everything

# Start containers. Should be in project root directory (where docker-compose.yml is)
docker-compose up

# See your containers running
docker ps

Note: It is possible the first time that Docker Compose executes that Flask will time-out until Postgres DB recreates the basic structure. If that happens run the following commands:

docker-compose down --remove-orphans
docker-compose up

Now go to localhost:8080 and the site should be running. Any changes/edits that you perform on the Webapp folder should be reflected immediately on the site.

Testing

We run with pytest and use some features from it. If you are using an ide (pycharm/intellij) you must change the default runner from unit tests to pytests You can make this change by going to settings and searching for "unittest" Python unit test configuration