#Guild Bounty Tracker
This tool is for guilds that like to find all the Bounty NPCs before starting the Guild Mission. It allows you to register your guild, then provide your guild members with a password they can use to log in and update their hunt progress.
The reference implementation is here:
The first step is to register your guild. This will create a guild account, with the Admin login being for you (and other officers, if you share the admin password with them), and the Member login being for regular members. Both logins use the same guild name, it's only the password that decides which one a user gets logged in as, so don't choose the same password for both. I should probably add a validation check to prevent that, come to think of it.
All regular members can do is mark various NPCs as Found or not, and enter their names in the Player boxes. Admins, on the other hand, can change the Member password, and change the Admin email address or password. Admins also have a Reset button they can use to clear all names and Found statuses from the table in one go.
If you want to set the app up on your own server, you're of course welcome to do so, but you'll need some experience dealing with Node applications. You'll also need, at minimum a working Redis database.
The general procedure for getting the app up and running is:
- Install dependencies
- Download the app
- Install the app's dependencies
- Configure the web server to be a proxy for the app
The dependencies in step 1 are:
- Git – version control and source control
- Node – runs the actual application
- Redis – key/value-based data store
- JSMin – JavaScript minifier
- Clean-CSS – CSS minifier
- Forever – restarts the application server if it crashes
Each item links to a page explaining how to install it. The guides I've chosen here are generally for Ubuntu, as that's what my staging and production servers run.
The last three items are only necessary for staging/production – in a development environment the minified files aren't used, and obviously it makes no sense to keep automatically restarting the app after a crash while you're still working on it.
Once you have the dependencies installed (and your Redis server is running), you
need to decide where you're going to install the app. In development, just pick
a directory where you normally develop stuff, e.g. ~/code/, then run:
$ git clone git://github.com/caerphoto/bounty_tracker.git
$ npm install
You can then start the server with node app.js and then visit
http://localhost:3000/ to see the app running.
In production, /var/www is a common install location. You'll need to perform
the following operations as a user with the required permissions; this
SuperUser post explains how to set up a user group that has the necessary
permissions, so you don't have to keep typing sudo (or worse, logging in as
root).
To download and install:
$ cd /var/www
$ git clone git://github.com/caerphoto/bounty_tracker.git
$ npm install
$ ./compile
$ NODE_ENV=production forever start app.js
The ./compile step may not be necessary, as the compiled JS and CSS files are
included in the Git repository, but it's best to run it anyway, just in case
I've forgotten to update them before a commit.
Once the app is running, you may be able to see it running at
http://yourdomain.tld:3000/, but clearly this isn't ideal from an end-user
point of view. That's where the final step comes in.
This is a potentially very complex topic, as the possibilities are endless, but if you're using nginx, this guide on the Arg! Team blog is a good starting point, and explains the reasoning behind using nginx to serve static assets (images, CSS, JS) whilst passing other requests to the Node app (tl;dr spoiler: it's faster).
If you're using Apache, this page looks easy enough to follow, though I can't vouch for it.
- NPC list needs maintenance.
- Dulfy links might be out of date - possibly better to have links to maps of paths directly?