curtus

Curtus

A fast and lightweight URL shortener. You can either deploy it as a standalone service or use as a library. Two ways to store data: either using a built-in synchronous SQLite3 (a good compromise) or an in-memory database (very fast but data is lost in case of restart).

Configuration options

• server.port — a port to listen to for HTTP requests
• server.token — authorization token string, set to null if not used
• salt — a random number from 0 to 10^9, must be constant for a given database
• inMemory — allows running with no file-based DB at all
• redirectOnError — a URL to redirect to when unpacking fails
• cleanup.checkOnEvery — a number of HTTP requests after which to check for outdated URLs
• cleanup.lifetime — a number in seconds for a minimal URL lifetime

Setup (npm package only)

1. npm i curtus --save
2. Test code

const curtus = require('curtus')

curtus.init({/* config object */})

const
  quickCode = curtus.db.saveUrl('https://someurl.com/trololo'),
  restoredUrl = curtus.db.getCodeUrl(quickCode)

console.log('Code is', quickCode)
console.log('Restored URL is', restoredUrl)

// run this when you need to wipe out old records
curtus.db.removeOldUrls()

// you may run this on your app shutdown to keep things clean
curtus.db.dbShutdown()

Setup (server only)

1. Create config.json file with the full or partial structure of config.dist.json if you need to change the default configuration. At least we recommend changing the salt.
2. npm i to install the packages.
3. npm run server to start a local server.

Authorization (server only)

If necessary shortening can be protected when used via a built-in server. Configure server.token value (see above) and pass the token in a HTTP request header: Authorization: Bearer YOUR_TOKEN_HERE.

Clean server install

If you want to install Curtus to a completely new server feel free to use the installation script provided. Simply run:

docs/linux-install.sh

To do

• cover with tests
• add performance tests
• type annotations and perfect code
• extend with optional statistics collection