This is the codebase for api.runemadsen.com, a hypermedia API that returns data about me, Rune Madsen. I was tired of updating my DB model every time I wanted to update my website, so I made this.... and made it public.

In a hypermedia'ish way, the API is explanatory. Do a GET on api.runemadsen.com, and it will list all resources available for unauthenticated users. If you have an authentication token (which you don't, because there's only one and that's for me), it will also list requests you can make while authenticated.

The API tries to conform to the HAL specification as much as possible. This means....

Every resource has a _links object that holds URI's for resource links. Here's a very simple (and fake) example of what that might look like.

{
  "_links" : {
    "self" : { "href" : "/animals" },
    "next" : { "href" : "/animals?page=2" },
    "show" : { "href" : "/animals/{id}", "templated" : true }
  }
}

Each link object has at least a href attribute that can be a simple URI or a URI template, the latter always represented with the templated property set to true.

If the resource comes preloaded with any of the above _links, those resources can be found in the _embedded object. This is especially important when loading a resource collection.

{
  "_links" : {
    "self" : { "href" : "/animals" },
    "next" : { "href" : "/animals?page=2" },
    "show" : { "href" : "/animals/{id}" }
  },
  "_embedded" : {
    "animals" : [
      {
        "name" : "lion",
        "type" : "cat",
        "_links": { "self": { "href": "/animals/lion" }}
      },
      {
        "name" : "tiger",
        "type" : "cat",
        "_links": { "self": { "href": "/animals/tiger" }}
      }
    ]
  }
}

You can paginate, yeah!

Status: Not Implemented

The writings endpoint returns shorter and longer essays written by me.

GET /articles

Status: Not Implemented

The portfolio endpoint returns work done by me.

GET /portfolio

Status: Not Implemented

The courses endpoint returns syllabi and lectures from classes taught by me.

GET /courses

Status: Not Implemented

The presentations endpoint returns slides from presentations given by me.

GET /presentations

Status: Not Implemented

The bio endpoint returns biographical info, like recent professional experience and education.

GET /bio

Status: Not Implemented

The info endpoint returns contact info.

GET /bio

Status: Not Implemented

GET /photos

Status: Not Implemented

The activities endpoint returns recent real world activities, like Foursquare check-ins.

GET /activities

Status: Not Implemented

The code endpoint returns programming activities, like statistics around programming languages and Git commits.

GET /code

Status: Not Implemented

The geo endpoint returns openpaths data

GET /geo

If you want to try run this api on a local machine, you can use vagrant do spin up a development box:

$ gem install berkshelf
$ vagrant plugin install vagrant-berkshelf
$ vagrant plugin install vagrant-omnibus
$ vagrant up

After chef has installed all the goodies, you can SSH into the dev box and run the api:

$ vagrant ssh
$ cd /vagrant
$ goreman start

You now have the api running on localhost:3001 and rethinkdb running on localhost:8081 on your host machine.

Run the test by doing this:

$ cd test
$ go test

• application+json/hal as middleware https://github.com/codegangsta/martini#middleware-handlers