HappySeed Base Install
=================

### What does this do?

The `happy_seed` application template sets up `haml`, `meta-tags`, disables `turbolinks`, and installs the `happy_seed` generators into your application.

The `happy_seed:base` generator installs `dotenv-rails`, `puma`, http site authentication, and the `setup_controller`.

/setup is used to see the `happy_seed` documentation for each of the installed generators, and copies things into the rails doc directory.  The `setup_controller` is only visible locally.

### Why do you want this?

Haml is great and we like it better than balancing out closing tags.  meta-tags makes it easy to flesh out SEO.

Turbolinks tend to be more trouble than they are worth.  Outside of making web site analytics slightly more cumbersome, we end up having trouble with debugging and testing.  Turbolinks work by having all of the javascript of the page loaded and resident and only replacing the body of the page when the user loads up a new page.  When everything works correctly this makes things faster; when it doesn't work correctly, you end up tracking down issues where DOM event handlers aren't correctly attached.

dotenv-rails mimics the way that heroku starts up the application even when using "rails s", rather than foreman.  This pulls in data from the .env file.  In the case where you only have the web application in the Procfile, its easier to read through the log file when you use rails s rather than foreman.  .env is also a good way to make sure that none of your configuration stuff is stuck away in code somewhere.

Assuming that you aren't hemmoraging memory, using unicorn is a great way to get more out of your dynos.  This setup gives you 3 workers for every dyno.  (We have nothing against Puma.)

Finally we setup some basic HTTP auth and a simple setup controller to help navigate what seed has generated for you.  By default the HTTP auth stuff is off (i.e. no user/pass set) but the basic idea is that before you open things up to the world you can share your app to a limited group of people.


### Environment Variables

The AWS env variables are there for reference

If the `HTTP_AUTH_*` variables are set, `application_controller` will force them to be used for authentication to let people into the site.  (The `happy_seed:splash` generator will not use this.)

`AWS_ACCESS_KEY` stuff isn't used, but is here for reference.

```
  AWS_ACCESS_KEY_ID=
  AWS_SECRET_ACCESS_KEY=
  S3_BUCKET_NAME=
  HTTP_AUTH_USERNAME=
  HTTP_AUTH_PASSWORD=
```

### What needs to be done?

Even though the setup controller only works locally, you should plan on removing the `setup_controller` since it shows the configuration variables.