Skip to content

README.md

Latest commit

 

History

History
52 lines (41 loc) · 2.17 KB

README.md

File metadata and controls

52 lines (41 loc) · 2.17 KB
Raw

NanoLink - Backend API

This directory contains the core infrastructure and application code for the NanoLink backend.

Microservice Architecture

The backend is split into two independent services that communicate via Apache Kafka:

  1. API Server (index.ts & controllers/): A Fastify server that handles incoming HTTP requests. It resolves short URLs from Redis (cache-aside) and publishes click events to Kafka without blocking the HTTP response.
  2. Analytics Worker (analytics/analyticsWorker.ts): A headless Kafka consumer that runs in the background, batches incoming click events, and performs bulk inserts into PostgreSQL.

API Endpoints

1. Health Check

Checks if the server is up and running.

  • URL: /
  • Method: GET
  • Response: {"Status": "OK"}

2. Shorten URL

Accepts a long URL, validates it using Zod, and generates a short link using Base62.

  • URL: /url
  • Method: POST
  • Body: {"url": "https://www.google.com"}
  • Response: {"Shortened URL": "http://localhost:3000/a1B2c3D"}

3. Redirect (The Core Engine)

Redirects the user to the original website. Behind the scenes: Hits Redis first. If Cache Miss, falls back to Postgres. Publishes a Fire-And-Forget event to Kafka.

  • URL: /:shortId
  • Method: GET
  • Behavior: HTTP 302 Redirect to the original long_url. Returns 404 if not found.

4. View All URLs

Returns all saved URLs in the database with pagination.

  • URL: /all
  • Method: GET
  • Response: {"RESULT": [{"id": 1, "short_code": "a1B2c3D", "long_url": "https://... "}]}

5. View Analytics Dashboard

Returns an aggregated list of all URLs joined with their total click counts.

  • URL: /analytics
  • Method: GET
  • Response: {"results": [{"short_code": "a1B2c3D", "total_clicks": 42}]}

Docker Infrastructure

The backend is fully containerized. The compose.yaml file defines 5 distinct services:

  1. url_shortner: The main Fastify API.
  2. analytics_worker: The headless background worker.
  3. postgres_db: The relational database.
  4. redis_cache: The in-memory cache for ultra-fast reads.
  5. kafka_kraft: The message broker running in KRaft mode (no Zookeeper).