codonnell/difficulty-api

Difficulty estimator for torn

difficulty-api

Description

difficulty-api is the backend for a script which helps estimate the difficulty of targets in the game Torn. It stores an attack log for each of its users. It uses its stored attacks to give an estimated difficulty rating for a target, given a would-be attacker's battle stats.

In order to achieve this functionality, difficulty-api exposes two endpoints: api-key and difficulties. The api-key endpoint is used for signing up by submitting an api key or just updating the api key for an existing player if it has changed. The difficulties endpoint requires an api key to determine the identity of the caller as well as a list of torn player ids whose difficulties will be estimated.

Installation

You will need a license for Datomic Pro, which you can get here. In order to install it, follow the instructions provided by Datomic under Leiningen setup. You can also set the environment variables $DATOMIC_USERNAME and $DATOMIC_PASSWORD. If you need persistent storage, like on a production server, you will need to provision that storage. This project is set up to use Postgres, which Datomic provides instructions for here. In order to connect to a persistent database, you will need to have a transactor running.

Create a file resources/env.edn which has a map like {:config-file "resources/dev.edn"} which points to the edn file containing your config options.

Usage

Run the application locally

lein repl
(go)

Run the tests

lein test

Packaging and running as standalone jar

Before starting the application in a production setting (with persistent storage), make sure the Datomic transactor is running.

lein do clean, uberjar
java -jar target/difficulty-api-$VERSION-standalone.jar

Connecting to the embedded repl

If you are running with the default settings (and use leiningen), you can connect to the embedded repl with

lein repl :connect localhost:7888

If you have the application running on a production server with your repl port blocked (as it should be), you can connect to the embedded repl using an ssh tunnel with port forwarding. In one terminal, set up the ssh tunnel:

ssh -nNT -L 9000:localhost:7888 username@serveraddress

Now you can connect to the embedded repl with

lein repl :connect localhost:9000