beijingrb / hydra_shield

A Rails Engine to generate and consume Hydra APIs

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

HydraShield

Usage

Mount the engine in your routes:

# config/routes.rb
Rails.application.routes.draw do

  # Your routes ...

  mount HydraShield::Engine => "/"
end

In your controller, use the shield helper:

# app/controllers/user_controller.rb
class UserController < ApplicationController

  def index
    users = find_users
    # You can explicitly indicate which shield should be used
    shield users: users, with: UserCollectionShield
  end

  def show
    user = find_user
    # Or it can be inferred from the first object class name. User => UserShield
    shield user: user
  end
end

Finally, define the shields:

# app/shields/user_shield.rb
class UserShield < HydraShield::ResourceShield

  # All items you passed in to the shield helper (except `:with`) get set as
  # instance variables. You'll probably want access to them directly.
  attr_reader :user

  # Set the required fields.
  self.id    = -> { user_path(user) }
  self.type  = "User"
  self.label = "A user is a person or agent who can interact with the app"

  # Declare the properties that should be exposed, their type, and a description.
  # Type and description are used in the vocab.
  property :login,     :string, "The username for the user", :required
  property :full_name, :string, "The user's full name",      :required

  property :last_login_at, :timestamp, "The last time the user was seen",  :readonly
  property :active,        :boolean,   "Has the user logged in recently?", :readonly

  # There's a shortcut if you want `created_at` and `deleted_at`:
  include_timestamps

  # All those properties have to be methods on the shield. The Forwardable module
  # is automatically included for the simple cases:
  delegate [:login, :full_name, :last_login_at, :created_at, :updated_at] => :user

  # For the ones involving logic, just add a real method.
  def active
    last_login_at > 1.month.ago
  end

  # The operations we can perform on the resource
  operation :retrieve_user,
            method: "GET",
            description: "Retrives the user",
            returns: type,
            status_codes: {
              404: "The User wasn't found"
            }

  operation :update_user,
            method: "PUT",
            description: "Updates the user's attributes",
            expects: type,
            returns: type,
            status_codes: {
              404: "The User wasn't found"
            }

  link :account,
       "The account of which the user is a member",
       operations: [ AccountShield[:retrieve_account] ]

  # `link`s are properties, too
  def account
    user_account_path(user)
  end

end

TODOs

  • Property types should be a schema.org type, or a object that can be expanded in the context:
{
    "@context": {
        "Event": "http://schema.org/Event",
        "description": "http://schema.org/description",
        "end_date": {
            "@id": "http://schema.org/endDate",
            "@type": "http://www.w3.org/2001/XMLSchema#dateTime"
        },
        "hydra": "http://www.w3.org/ns/hydra/core#",
        "name": "http://schema.org/name",
        "start_date": {
            "@id": "http://schema.org/startDate",
            "@type": "http://www.w3.org/2001/XMLSchema#dateTime"
        },
        "vocab": "http://www.markus-lanthaler.com/hydra/event-api/vocab#"
    }
}

About

A Rails Engine to generate and consume Hydra APIs

License:MIT License


Languages

Language:Ruby 83.7%Language:HTML 10.9%Language:CSS 2.9%Language:JavaScript 2.5%