unshrtn

unshrtn is a LevelDB backed URL unshortening microservice written in JavaScript and Node for quick, asynchronous processing of requests. unshrtn remembers what it has already looked up so it can save you the trouble of keeping track of URLs when you are looking up a lot of them at one time and they haven't necessarily been de-duplicated.

unshrtn uses the metaweb library to do the lookups, which means that in addition to returning the unshortened URL you will also get some potentially useful metadata if the response happens to be HTML:

% curl http://localhost:3000?url=https://bit.ly/348J1DN
{
  "short": "https://bit.ly/348J1DN",
  "long": "https://www.youtube.com/watch?v=oHg5SJYRHA0",
  "canonical": "https://www.youtube.com/watch?v=oHg5SJYRHA0",
  "status": 200,
  "content_type": "text/html; charset=utf-8",
  "title": "RickRoll'D - YouTube",
  "description": "http://www.facebook.com/rickroll548 As long as trolls are still trolling, the Rick will never stop rolling.",
  "image": "https://i.ytimg.com/vi/oHg5SJYRHA0/hqdefault.jpg",
  "publisher": "YouTube",
  "keywords": ["Cotter548", "Shawn", "Cotter", "lol", "gamefaqs", "CE", "reddit", "rettocs", "no", "brb", "afk", "lawl", "pwnt", "Rickroll", "Rickroll'd", "Rick", "Roll", "Duckroll", "Duck", "rick", "roll", "astley", "..."]
}

Install

Probably the easiest way to get unshrtn up and running is with Docker.

docker run -p 3000:3000 docnow/unshrtn
Status: Downloaded newer image for docnow/unshrtn:latest
info: started http://0.0.0.0:3000

You can also install it with npm:

% npm install -g unshrtn
% unshrtn
info: started http://0.0.0.0:3000

Usage

Once installed you can start up the microservice:

% unshrtn start

If you'd like more control over how it starts use the start subcommand:

% unshrtn start --host 0.0.0.0 --port 8080 --log /var/log/unshrtn.log

By default unshrtn writes its database to unshrtn.db in the current working directory. However you can control this with the --database option:

% unshrt start --database /var/unshrtn/unshrtn.db

You can dump the database as line oriented JSON and load it again if you want to back it up or move it around:

% unshrtn dump /path/to/datbase > backup.jsonl
% cat backup.jsonl | unshrtn load /path/to/another/database

If you just want to look up one URL you can also use unshrtn on the command line:

% unshrtn get https://bitly.com/4kb77v