HitFactor.Info

A Better Classification System for Action Shooting Sports

Project Goals

1. Pick Data-Driven Recommended HHF for All Classifiers/Divisions.
2. Implement Closer-To-Major-Match-Performance Classification Algorithm
3. Provide Better Classification System or at least Partial Improvements to All Interested Action Shooting Sport Organizations

License

This repository is licensed under the MIT License, except for the contents of the following directories:

• data/

See the README files in those directories for more details.

Running

Caution

This README is possibly (most likely) OUTDATED. If you're a developer willing to contribute, join our Discord

Locally with Docker-Compose

For faster turn around when developing API, use: This starts up the app with a Mongo instance running in Docker.

npm i
npm run local

Populate the local database with production data by doing the following:

1. Download a production mongodump archive (ask an existing contributor to provide you with the archive).
2. Extract zeta folder to be at the root level of this repo.
3. Run the following mongorestore command:

mongorestore --host localhost:27017 zeta -d test

NOTE: Before running this command, it is highly recommended to increase memory availability in Docker Desktop to at least 5 or 6 GB.

Otherwise, the restore process may crash due to Docker's OOMKiller.

Locally against MONGO_URL

Note: You must supply MONGO_URL={URL_OF_SANDBOX_DATABASE} to test in this manner.

npm i
npm start

Connect to the mongo instance via:

mongosh "mongodb://localhost:27017"

Run the Uploads Worker locally

The Uploads Worker fetches matches and populates scores.

To run it locally, run the following:

ALGOLIA_URL='from_env' PS_S3_ACCESS_KEY_ID=from_env PS_S3_SECRET_ACCESS_KEY=from_env MONGO_URL=mongodb://localhost:27017 NODE_OPTIONS='--max-old-space-size=512' node scripts/uploadsWorker.js

You may need to tweak scheduling of the runner to run at more frequent intervals for testing purposes.

In Production

Currently deployed on Koyeb using Dockerfiles. To run api/web in prod mode, use:

npm i
npm run prod

Note: npm i is required, because it uses vite build as a post-install step and serves frontend files from the node itself, instead of running two processes concurrently.

Technical Stack

• Main language: JavaScript (ES13), TypeScript when needed
• Monorepo, Node/Fastify Backend, React (vite-swc) Frontend.
• Backend serves API and static files: (build of React Frontend, downloadables, etc)
• Mongo

Folder Structure

• scripts/ -- standalone scripts
• data/ -- imported (partially processed / split) data, mostly used by backend
• shared/ -- source code imported by both front- and backend
• api/ -- backend
• web/ -- frontend
• package.json -- monorepo wrapper scripts

For more info, see READMEs in each root folder