demo-twitter-metrics

Simple demo application gathering and displaying some metrics from twitter api.

Using Deno, mostly because it's a simpler approach to being able to create something like this quickly.

Dependencies

• Docker - either docker desktop or community to run the compose file declarations.
• Deno - JavaScript/TypeScript runtime for the backend api and metrics applications.
• Node - JavaScript runtime and build tools for the frontend ui application.

Running

• Copy .env.example to .env.
  • Edit the file to include relevant keys for use with Twitter's API
• docker compose up -d

You can start the redis instance specifically with docker compose up -d redis and the separate applications can each be started similarly, or developed directly, as they will each search for their given .env file upward.

The project/ui will run with npm start and reverse-proxy to the API on the default backend port.

Projects

Each application will be under project/* as self-contained.

See README.md in each project directory.

• redis - Backend data store.
• api - Backend API to retrieve rollup counts.
• twitter-metrics - Backend metrics gathering from twitter through api.
• ui - React/TypeScript UI with Vite.js

Other considerations

I feel the first versions of most things should be done in a scripted language/environment and with containerization and APIs it's easy enough to swap out such services as necessary if per-instance performance needs to be optimized for a given area.

• node - similar approach to deno, slightly more complex app management
• rust - likely optimal in terms of absolute throughput with minimal overhead
• golang - maximum portability, single executable, slightly slower than rust
• C# (.Net) - fair portability, similar performance to golang

License

MIT License