jeanlange / named_seeds

Replace ActiveRecord::Fixtures With #{your_factory_lib}

NamedSeeds

Make your tests fast by augmenting them with transactional fixtures powered by your favorite factory library!

We all know that ActiveRecord::Fixtures suck because they are authored in YAML files. But Rails did get something right with transactional tests and easy helper methods to access fixtures by name. NamedSeeds aims to be a drop-in replacement for Rails fixtures or an enhancement to RSpec and Cucumber while using any object generator of your choice!

The idea is to leverage your tests' existing factories to generate fixtures that will be populated before testing starts and to use a database transaction strategy around each test. In this way you have a populated story that can be accessed via convienient helper methods like users(:admin) which in turn yields much faster test runs. Database fixtures, even those seeded by factories, are not a panacea and we highly suggested that you continue to use factories in your tests when it makes sense to do so. For those that think this is mad or have FUD to share, please see my blog articles:

• Inprogress...
• Inprogress...

Installation

Add the named_seeds gem to your Rails' Gemfile in both the development and test group as shown below. This is needed as the NamedSeeds gem exposes rake tasks needed in both environments.

group :development, :test do
  gem 'named_seeds'
end

Usage

NamedSeeds requires that you create a db/test/seeds.rb file. The contents of this file can be anything you want. We recommend using a factory library like FactoryGirl or Machinist.

require 'factory_girl'
FactoryGirl.find_definitions rescue true

@bob = FactoryGirl.create :user, :id => NamedSeeds.identify(:bob), :email => '[email protected]'

Use the NamedSeeds.identify method to give a name to the identity used for this record. You will be able to find this record using that name later on.

Rake Files

Using these tasks manually should not be needed as both are hooked into the proper test:prepare and db:setup process for you.

NamedSeeds includes two rake tasks. The db:test:seeds is the one that does all the work and is automatically called after Rails' test:prepare for you. So running your rake test tasks will create your test database and seed it for you automatically before your tests run.

The other task is db:development:seed. This task invokes the normal Rails db:seed task, then loads the db/test/seeds.rb file while still in development mode. We automatically call this task after db:setup for you. This task provides a way for a developer to populate their development database with the same fixture story used for testing. This makes it easy for developers to learn your application as the test story is a 1 to 1 mapping of data in local development.

$ rake db:test:seed          # Run the seed data from db/test/seeds.rb and 
                             # optionally your Rails db/seeds.rb if you have 
                             # configured `app_load_seed` below.

$ rake db:development:seed   # Runs the normal Rails `db:seed` task then 
                             # loads the db/test/seeds.rb file.

Rails

By default, Rails' ActiveSupport::TestCase has set the use_transactional_fixtures to true. So all you need to do is declare which tables have NamedSeeds keys.

ENV["RAILS_ENV"] = "test"
require File.expand_path('../../config/environment', __FILE__)
require 'rails/test_help'

class ActiveSupport::TestCase
  named_seeds :users, :posts
end

Now you can use both the users and posts helper methods to find any named identity from your seed file.

require 'test_helper'
class UserTest < ActiveSupport::TestCase
  setup { @user = users(:bob) }
  tests "should work" do
    assert_equal '[email protected]', @user.posts
  end
end

RSpec

Coming soon...

Cucumber

Coming soon...

Advanced Usage

Review how helper methods may map to custom table names. Like Rails did with #fixture_table_names...

Configurations

NamedSeeds is a Rails::Railtie that exposes a few config options. So open up the config/environments/development.rb (yes in development.rb) and use the config.named_seeds options below.

• app_load_seed - Load your Rails application's db/seeds.rb file into the test database. This is done before db/test/seeds.rb is loaded. Default is false.
• engines_with_load_seed - Some Rails engines provide a load seed hook. If you want NamedSeed to call the engine's seed method into your tests database, push the engine constant to this array. Any object responding to load_seed sould work here too. Default is an empty array.

My::Application.configure do
  config.named_seeds.app_load_seed = true
  config.named_seeds.engines_with_load_seed += [GeoData::Engine, OurLookupTables]
end

NamedSeeds uses DatabaseCleaner to clean the database before seeding it. Use the config.named_seeds.db_cleaner options below to configure its behavior. Please see the DatabaseCleaner documentation for full details.

• orm - The ORM module to use. Default is :active_record.
• connection - The connection name to use. Default is :test.
• strategy - Strategy to clean the database with. Default is :truncation.
• strategy_args - Args to be passed to the strategy. Default is an empty hash.

My::Application.configure do
  config.named_seeds.db_cleaner.orm           = :active_record
  config.named_seeds.db_cleaner.connection    = :test
  config.named_seeds.db_cleaner.strategy      = :truncation
  config.named_seeds.db_cleaner.strategy_args = {:except => ['geodata', 'lookuptable']}
end

Todo

Show Rails implementation using ActiveSupport::TestCase or RSpec.

Show examples with these 3 factory libraries.

• https://github.com/thoughtbot/factory_girl
• https://github.com/paulelliott/fabrication
• https://github.com/notahat/machinist

Set up a dummy_application in test.

How are we different from:

• https://github.com/mbleigh/seed-fu
• https://github.com/sevenwire/bootstrapper/
• https://github.com/franciscotufro/environmental_seeder

Talk about http://railscasts.com/episodes/179-seed-data