Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Her is an ORM (Object Relational Mapper) that maps REST resources to Ruby objects. It is designed to build applications that are powered by a RESTful API instead of a database.
Ruby Other

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
examples
lib
spec
.gitignore
.travis.yml
Gemfile
Guardfile
LICENSE
README.md
Rakefile
her.gemspec

README.md

Her

Her is an ORM (Object Relational Mapper) that maps REST resources to Ruby objects. It is designed to build applications that are powered by a RESTful API instead of a database.

Build Status

Installation

In your Gemfile, add:

gem "her"

That’s it!

Usage

First, you have to define which API your models will be bound to. For example, with Rails, you would create a new config/initializers/her.rb file with this line:

# config/initializers/her.rb
Her::API.setup :base_uri => "https://api.example.com"

And then to add the ORM behavior to a class, you just have to include Her::Model in it:

class User
  include Her::Model
end

After that, using Her is very similar to many ActiveModel-like ORMs:

User.all
# GET https://api.example.com/users and return an array of User objects

User.find(1)
# GET https://api.example.com/users/1 and return a User object

@user = User.create(:fullname => "Tobias Fünke")
# POST "https://api.example.com/users" with the data and return a User object

@user = User.new(:fullname => "Tobias Fünke")
@user.occupation = "actor"
@user.save
# POST https://api.example.com/users with the data and return a User object

@user = User.find(1)
@user.fullname = "Lindsay Fünke"
@user.save
# PUT https://api.example.com/users/1 with the data and return+update the User object

Middleware

Since Her relies on Faraday to send HTTP requests, you can add additional middleware to handle requests and responses. Using a block in the setup call, you have access to Faraday’s builder object and are able to customize the middleware stack used on each request and response.

Authentication

Her doesn’t support any kind of authentication. However, it’s very easy to implement one with a request middleware. Using the builder block, we add it to the default list of middleware.

class MyAuthentication < Faraday::Middleware
  def initialize(app, options={})
    @options = options
  end

  def call(env)
    env[:request_headers]["X-API-Token"] = @options[:token] if @options.include?(:token)
    @app.call(env)
  end
end

Her::API.setup :base_uri => "https://api.example.com" do |builder|
  # This token could be stored in the client session
  builder.use MyAuthentication, :token => "bb2b2dd75413d32c1ac421d39e95b978d1819ff611f68fc2fdd5c8b9c7331192"
end

Now, each HTTP request made by Her will have the X-API-Token header.

Parsing JSON data

By default, Her handles JSON data. It expects the resource/collection data to be returned at the first level.

Note: Before 0.2, Her expected the resource/collection data to be returned in a data key within the JSON object. If you want the old behavior, you can use the Her::Middleware::SecondLevelParseJSON middleware.

// The response of GET /users/1
{ "id" : 1, "name" : "Tobias Fünke" }

// The response of GET /users
[{ "id" : 1, "name" : "Tobias Fünke" }]

However, you can define your own parsing method, using a response middleware. The middleware is expected to set env[:body] to a hash with three keys: data, errors and metadata. The following code enables parsing JSON data and treating the result as first-level properties. Using the builder block, we then replace the default parser with our custom parser.

class MyCustomParser < Faraday::Response::Middleware
  def on_complete(env)
    json = MultiJson.load(env[:body], :symbolize_keys => true)
    env[:body] = {
      :data => json[:result],
      :errors => json[:errors],
      :metadata => json[:metadata]
    }
  end
end

Her::API.setup :base_uri => "https://api.example.com" do |builder|
  # We use the `swap` method to replace Her’s default parser middleware
  builder.swap Her::Middleware::DefaultParseJSON, MyCustomParser
end
# User.find(1) will now expect "https://api.example.com/users/1" to return something like '{ "result" => { "id": 1, "name": "Tobias Fünke" }, "errors" => [] }'

OAuth

Using the faraday_middleware and simple_oauth gems, it’s fairly easy to use OAuth authentication with Her.

In your Gemfile:

gem "her"
gem "faraday_middleware"
gem "simple_oauth"

In your Ruby code:

# Create an application on `https://dev.twitter.com/apps` to set these values
TWITTER_CREDENTIALS = {
  :consumer_key => "",
  :consumer_secret => "",
  :token => "",
  :token_secret => ""
}

Her::API.setup :base_uri => "https://api.twitter.com/1/" do |builder|
  # We need to insert the middleware at the beginning of the stack (hence the `insert 0`)
  builder.insert 0, FaradayMiddleware::OAuth, TWITTER_CREDENTIALS
end

class Tweet
  include Her::Model
end

@tweets = Tweet.get("/statuses/home_timeline.json")

Caching

Again, using the faraday_middleware makes it very easy to cache requests and responses:

In your Gemfile:

gem "her"
gem "faraday_middleware"

In your Ruby code:

class MyCache
  def initialize
    @cache = {}
  end

  def write(key, value)
    @cache[key] = value
  end

  def read(key)
    @cache[key]
  end

  def fetch(key, &block)
    return value = read(key) if value.nil?
    write key, yield
  end
end

# A cache system must respond to `#write`, `#read` and `#fetch`.
# We should be probably using something like Memcached here, not a global object
$cache = MyCache.new

Her::API.setup :base_uri => "https://api.example.com" do |builder|
  builder.use FaradayMiddleware::Caching, $cache
end

class User
  include Her::Model
end

@user = User.find(1)
# GET /users/1

@user = User.find(1)
# This request will be fetched from the cache

Relationships

You can define has_many, has_one and belongs_to relationships in your models. The relationship data is handled in two different ways. If there’s relationship data when parsing a resource, it will be used to create new Ruby objects.

If no relationship data was included when parsing a resource, calling a method with the same name as the relationship will fetch the data (providing there’s an HTTP request available for it in the API).

For example, with this setup:

class User
  include Her::Model
  has_many :comments
  has_one :role
  belongs_to :organization
end

class Comment
  include Her::Model
end

class Role
  include Her::Model
end

class Organization
  include Her::Model
end

If there’s relationship data in the resource, no extra HTTP request is made when calling the #comments method and an array of resources is returned:

@user = User.find(1) # { :data => { :id => 1, :name => "George Michael Bluth", :comments => [{ :id => 1, :text => "Foo" }, { :id => 2, :text => "Bar" }], :role => { :id => 1, :name => "Admin" }, :organization => { :id => 2, :name => "Bluth Company" } }}
@user.comments # => [#<Comment id=1>, #<Comment id=2>] fetched directly from @user
@user.role # => #<Role id=1> fetched directly from @user
@user.organization # => #<Organization id=2> fetched directly from @user

If there’s no relationship data in the resource, an extra HTTP request (to GET /users/1/comments) is made when calling the #comments method:

@user = User.find(1) # { :data => { :id => 1, :name => "George Michael Bluth" }}
@user.comments # => [#<Comment id=1>, #<Comment id=2>] fetched from /users/1/comments

For has_one relationships, an extra HTTP request (to GET /users/1/role) is made when calling the #role method:

@user = User.find(1) # { :data => { :id => 1, :name => "George Michael Bluth" }}
@user.role # => #<Role id=1> fetched from /users/1/role

For belongs_to relationships, an extra HTTP request (to GET /organizations/2) is made when calling the #organization method:

@user = User.find(1) # { :data => { :id => 1, :name => "George Michael Bluth", :organization_id => 2 }}
@user.organization # => #<Organization id=2> fetched from /organizations/2

However, subsequent calls to #comments, #role and #organization will not trigger extra HTTP requests as the data has already been fetched.

Hooks

You can add before and after hooks to your models that are triggered on specific actions (save, update, create, destroy):

class User
  include Her::Model
  before_save :set_internal_id

  def set_internal_id
    self.internal_id = 42 # Will be passed in the HTTP request
  end
end

@user = User.create(:fullname => "Tobias Fünke")
# POST /users&fullname=Tobias+Fünke&internal_id=42

Custom requests

You can easily define custom requests for your models using custom_get, custom_post, etc.

class User
  include Her::Model
  custom_get :popular, :unpopular
  custom_post :from_default
end

User.popular # => [#<User id=1>, #<User id=2>]
# GET /users/popular

User.unpopular # => [#<User id=3>, #<User id=4>]
# GET /users/unpopular

User.from_default(:name => "Maeby Fünke") # => #<User id=5>
# POST /users/from_default?name=Maeby+Fünke

You can also use get, post, put or delete (which maps the returned data to either a collection or a resource).

class User
  include Her::Model
end

User.get(:popular) # => [#<User id=1>, #<User id=2>]
# GET /users/popular

User.get(:single_best) # => #<User id=1>
# GET /users/single_best

Also, get_collection (which maps the returned data to a collection of resources), get_resource (which maps the returned data to a single resource) or get_raw (which yields the parsed data return from the HTTP request) can also be used. Other HTTP methods are supported (post_raw, put_resource, etc.).

class User
  include Her::Model

  def self.popular
    get_collection(:popular)
  end

  def self.total
    get_raw(:stats) do |parsed_data|
      parsed_data[:data][:total_users]
    end
  end
end

User.popular # => [#<User id=1>, #<User id=2>]
User.total # => 42

You can also use full request paths (with strings instead of symbols).

class User
  include Her::Model
end

User.get("/users/popular") # => [#<User id=1>, #<User id=2>]
# GET /users/popular

Custom paths

You can define custom HTTP paths for your models:

class User
  include Her::Model
  collection_path "/hello_users/:id"
end

@user = User.find(1)
# GET /hello_users/1

You can also include custom variables in your paths:

class User
  include Her::Model
  collection_path "/organizations/:organization_id/users"
end

@user = User.find(1, :_organization_id => 2)
# GET /organizations/2/users/1

@user = User.all(:_organization_id => 2)
# GET /organizations/2/users

@user = User.new(:fullname => "Tobias Fünke", :organization_id => 2)
@user.save
# POST /organizations/2/users

Multiple APIs

It is possible to use different APIs for different models. Instead of calling Her::API.setup, you can create instances of Her::API:

# config/initializers/her.rb
$my_api = Her::API.new
$my_api.setup :base_uri => "https://my_api.example.com"

$other_api = Her::API.new
$other_api.setup :base_uri => "https://other_api.example.com"

You can then define which API a model will use:

class User
  include Her::Model
  uses_api $my_api
end

class Category
  include Her::Model
  uses_api $other_api
end

User.all
# GET https://my_api.example.com/users

Category.all
# GET https://other_api.example.com/categories

Things to be done

  • Better error handling
  • Better API documentation (using YARD)

Contributors

Feel free to contribute and submit issues/pull requests on GitHub like these fine folks did:

Take a look at the spec folder before you do, and make sure bundle exec rake spec passes after your modifications :)

License

Her is © 2012 Rémi Prévost and may be freely distributed under the MIT license. See the LICENSE file.

Something went wrong with that request. Please try again.