Link Shortener

KV database studies and proofs of concept

My implementation: "JFA Link Shortener" Terms of Use

Requirements

• A KV-like NoSQL database, like AWS DynamoDB or Cloudflare KV
• A serverless functions provider, like AWS Lambda or Cloudflare Workers
• A way to interact with the API, like cURL or a frontend UI of any kind

How it works

API docs

Example frontend docs

How to deploy

Cloudflare

Prerequisites

• Cloudflare account with Workers enabled
• Wrangler CLI (npm install -g wrangler)
• Two KV namespaces created in Cloudflare:
  • One for URL storage
  • One for rate limiting

Steps

1. Clone and set up:

$ git clone https://github.com/jfalava/link-shortener.git
$ cd link-shortener/cloudflare
$ npm install -g pnpm
$ pnpm install

2. Create KV namespaces in Cloudflare dashboard or via the wrangler cli:

$ pnpx wrangler kv:namespace create "URL_STORAGE"
$ pnpx wrangler kv:namespace create "RATE_LIMIT"

3. Configure wrangler.toml:

name = "<your project name>"
main = "src/index.ts"
compatibility_date = "2025-01-29"
compatibility_flags = ["nodejs_compat"]
routes = [{ pattern = "<your-domain>", custom_domain = true }]

[observability]
enabled = true

[placement]
mode = "smart"

[[kv_namespaces]]
binding = "URL_STORAGE"
id = "<URL_STORAGE id>"

[[kv_namespaces]]
binding = "RATE_LIMIT"
id = "<RATE_LIMIT id>"

4. Deploy to Cloudflare:

$ pnpm run deploy

After deployment, your worker will be available at your-worker.your-subdomain.workers.dev

TODO

• AWS Lambda version of the API endpoint
• IBM Cloud Functions version of the API endpoint
• Self-hosted version of the code and the infrastructure required
• IaaC deployments for AWS and IBM Cloud