Skip to content
This repository
file 1209 lines (925 sloc) 29.445 kb

Getting Started

Update Your Gemfile

If you're using Rails, you'll need to change the required version of factory_girl_rails:

gem "factory_girl_rails", "~> 4.0"

If you're not using Rails, you'll just have to change the required version of factory_girl:

gem "factory_girl", "~> 4.0"

JRuby users: factory_girl works with JRuby starting with 1.6.7.2 (latest stable, as per July 2012). JRuby has to be used in 1.9 mode, for that, use JRUBY_OPTS environment variable:

export JRUBY_OPTS=--1.9

Once your Gemfile is updated, you'll want to update your bundle.

Configure your test suite

# rspec
RSpec.configure do |config|
  config.include FactoryGirl::Syntax::Methods
end

# Test::Unit
class Test::Unit::TestCase
  include FactoryGirl::Syntax::Methods
end

# Cucumber
World(FactoryGirl::Syntax::Methods)

# Spinach
class Spinach::FeatureSteps
  include FactoryGirl::Syntax::Methods
end

# MiniTest
class MiniTest::Unit::TestCase
  include FactoryGirl::Syntax::Methods
end

# MiniTest::Spec
class MiniTest::Spec
  include FactoryGirl::Syntax::Methods
end

# minitest-rails
class MiniTest::Rails::ActiveSupport::TestCase
  include FactoryGirl::Syntax::Methods
end

If you do not include FactoryGirl::Syntax::Methods in your test suite, then all factory_girl methods will need to be prefaced with FactoryGirl.

Linting Factories

factory_girl allows for linting known factories:

FactoryGirl.lint

FactoryGirl.lint builds each factory and subsequently calls #valid? on it (if #valid? is defined); if any calls to #valid? return false, FactoryGirl::InvalidFactoryError is raised with a list of the offending factories. Recommended usage of FactoryGirl.lint is to invoke this once before the test suite is run.

With RSpec:

# spec/support/factory_girl.rb
RSpec.configure do |config|
  # additional factory_girl configuration

  config.before(:suite) do
    begin
      DatabaseCleaner.start
      FactoryGirl.lint
    ensure
      DatabaseCleaner.clean
    end
  end
end

After calling FactoryGirl.lint, you'll likely want to clear out the database, as built factories will create associated records.

Defining factories

Each factory has a name and a set of attributes. The name is used to guess the class of the object by default, but it's possible to explicitly specify it:

# This will guess the User class
FactoryGirl.define do
  factory :user do
    first_name "John"
    last_name  "Doe"
    admin false
  end

  # This will use the User class (Admin would have been guessed)
  factory :admin, class: User do
    first_name "Admin"
    last_name  "User"
    admin      true
  end
end

It is highly recommended that you have one factory for each class that provides the simplest set of attributes necessary to create an instance of that class. If you're creating ActiveRecord objects, that means that you should only provide attributes that are required through validations and that do not have defaults. Other factories can be created through inheritance to cover common scenarios for each class.

Attempting to define multiple factories with the same name will raise an error.

Factories can be defined anywhere, but will be automatically loaded if they are defined in files at the following locations:

test/factories.rb
spec/factories.rb
test/factories/*.rb
spec/factories/*.rb

Using factories

factory_girl supports several different build strategies: build, create, attributes_for and build_stubbed:

# Returns a User instance that's not saved
user = build(:user)

# Returns a saved User instance
user = create(:user)

# Returns a hash of attributes that can be used to build a User instance
attrs = attributes_for(:user)

# Returns an object with all defined attributes stubbed out
stub = build_stubbed(:user)

# Passing a block to any of the methods above will yield the return object
create(:user) do |user|
  user.posts.create(attributes_for(:post))
end

No matter which strategy is used, it's possible to override the defined attributes by passing a hash:

# Build a User instance and override the first_name property
user = build(:user, first_name: "Joe")
user.first_name
# => "Joe"

Lazy Attributes

Most factory attributes can be added using static values that are evaluated when the factory is defined, but some attributes (such as associations and other attributes that must be dynamically generated) will need values assigned each time an instance is generated. These "lazy" attributes can be added by passing a block instead of a parameter:

factory :user do
  # ...
  activation_code { User.generate_activation_code }
  date_of_birth   { 21.years.ago }
end

Aliases

Aliases allow you to use named associations more easily.

factory :user, aliases: [:author, :commenter] do
  first_name    "John"
  last_name     "Doe"
  date_of_birth { 18.years.ago }
end

factory :post do
  author
  # instead of
  # association :author, factory: :user
  title "How to read a book effectively"
  body  "There are five steps involved."
end

factory :comment do
  commenter
  # instead of
  # association :commenter, factory: :user
  body "Great article!"
end

Dependent Attributes

Attributes can be based on the values of other attributes using the evaluator that is yielded to lazy attribute blocks:

factory :user do
  first_name "Joe"
  last_name  "Blow"
  email { "#{first_name}.#{last_name}@example.com".downcase }
end

create(:user, last_name: "Doe").email
# => "joe.doe@example.com"

Transient Attributes

There may be times where your code can be DRYed up by passing in transient attributes to factories.

factory :user do
  ignore do
    rockstar true
    upcased  false
  end

  name  { "John Doe#{" - Rockstar" if rockstar}" }
  email { "#{name.downcase}@example.com" }

  after(:create) do |user, evaluator|
    user.name.upcase! if evaluator.upcased
  end
end

create(:user, upcased: true).name
#=> "JOHN DOE - ROCKSTAR"

Static and dynamic attributes can be ignored. Ignored attributes will be ignored within attributes_for and won't be set on the model, even if the attribute exists or you attempt to override it.

Within factory_girl's dynamic attributes, you can access ignored attributes as you would expect. If you need to access the evaluator in a factory_girl callback, you'll need to declare a second block argument (for the evaluator) and access ignored attributes from there.

Associations

It's possible to set up associations within factories. If the factory name is the same as the association name, the factory name can be left out.

factory :post do
  # ...
  author
end

You can also specify a different factory or override attributes:

factory :post do
  # ...
  association :author, factory: :user, last_name: "Writely"
end

The behavior of the association method varies depending on the build strategy used for the parent object.

# Builds and saves a User and a Post
post = create(:post)
post.new_record?        # => false
post.author.new_record? # => false

# Builds and saves a User, and then builds but does not save a Post
post = build(:post)
post.new_record?        # => true
post.author.new_record? # => false

To not save the associated object, specify strategy: :build in the factory:

factory :post do
  # ...
  association :author, factory: :user, strategy: :build
end

# Builds a User, and then builds a Post, but does not save either
post = build(:post)
post.new_record?        # => true
post.author.new_record? # => true

Please note that the strategy: :build option must be passed to an explicit call to association, and cannot be used with implicit associations:

factory :post do
  # ...
  author strategy: :build    # <<< this does *not* work; causes author_id to be nil

Generating data for a has_many relationship is a bit more involved, depending on the amount of flexibility desired, but here's a surefire example of generating associated data.

FactoryGirl.define do

  # post factory with a `belongs_to` association for the user
  factory :post do
    title "Through the Looking Glass"
    user
  end

  # user factory without associated posts
  factory :user do
    name "John Doe"

    # user_with_posts will create post data after the user has been created
    factory :user_with_posts do
      # posts_count is declared as an ignored attribute and available in
      # attributes on the factory, as well as the callback via the evaluator
      ignore do
        posts_count 5
      end

      # the after(:create) yields two values; the user instance itself and the
      # evaluator, which stores all values from the factory, including ignored
      # attributes; `create_list`'s second argument is the number of records
      # to create and we make sure the user is associated properly to the post
      after(:create) do |user, evaluator|
        create_list(:post, evaluator.posts_count, user: user)
      end
    end
  end
end

This allows us to do:

create(:user).posts.length # 0
create(:user_with_posts).posts.length # 5
create(:user_with_posts, posts_count: 15).posts.length # 15

Inheritance

You can easily create multiple factories for the same class without repeating common attributes by nesting factories:

factory :post do
  title "A title"

  factory :approved_post do
    approved true
  end
end

approved_post = create(:approved_post)
approved_post.title    # => "A title"
approved_post.approved # => true

You can also assign the parent explicitly:

factory :post do
  title "A title"
end

factory :approved_post, parent: :post do
  approved true
end

As mentioned above, it's good practice to define a basic factory for each class with only the attributes required to create it. Then, create more specific factories that inherit from this basic parent. Factory definitions are still code, so keep them DRY.

Sequences

Unique values in a specific format (for example, e-mail addresses) can be generated using sequences. Sequences are defined by calling sequence in a definition block, and values in a sequence are generated by calling generate:

# Defines a new sequence
FactoryGirl.define do
  sequence :email do |n|
    "person#{n}@example.com"
  end
end

generate :email
# => "person1@example.com"

generate :email
# => "person2@example.com"

Sequences can be used as attributes:

factory :user do
  email
end

Or in lazy attributes:

factory :invite do
  invitee { generate(:email) }
end

And it's also possible to define an in-line sequence that is only used in a particular factory:

factory :user do
  sequence(:email) { |n| "person#{n}@example.com" }
end

You can also override the initial value:

factory :user do
  sequence(:email, 1000) { |n| "person#{n}@example.com" }
end

Without a block, the value will increment itself, starting at its initial value:

factory :post do
  sequence(:position)
end

Sequences can also have aliases. The sequence aliases share the same counter:

factory :user do
  sequence(:email, 1000, aliases: [:sender, :receiver]) { |n| "person#{n}@example.com" }
end

# will increase value counter for :email which is shared by :sender and :receiver
generate(:sender)

Define aliases and use default value (1) for the counter

factory :user do
  sequence(:email, aliases: [:sender, :receiver]) { |n| "person#{n}@example.com" }
end

Setting the value:

factory :user do
  sequence(:email, 'a', aliases: [:sender, :receiver]) { |n| "person#{n}@example.com" }
end

The value just needs to support the #next method. Here the next value will be 'a', then 'b', etc.

Traits

Traits allow you to group attributes together and then apply them to any factory.

factory :user, aliases: [:author]

factory :story do
  title "My awesome story"
  author

  trait :published do
    published true
  end

  trait :unpublished do
    published false
  end

  trait :week_long_publishing do
    start_at { 1.week.ago }
    end_at   { Time.now }
  end

  trait :month_long_publishing do
    start_at { 1.month.ago }
    end_at   { Time.now }
  end

  factory :week_long_published_story,    traits: [:published, :week_long_publishing]
  factory :month_long_published_story,   traits: [:published, :month_long_publishing]
  factory :week_long_unpublished_story,  traits: [:unpublished, :week_long_publishing]
  factory :month_long_unpublished_story, traits: [:unpublished, :month_long_publishing]
end

Traits can be used as attributes:

factory :week_long_published_story_with_title, parent: :story do
  published
  week_long_publishing
  title { "Publishing that was started at #{start_at}" }
end

Traits that define the same attributes won't raise AttributeDefinitionErrors; the trait that defines the attribute latest gets precedence.

factory :user do
  name "Friendly User"
  login { name }

  trait :male do
    name   "John Doe"
    gender "Male"
    login { "#{name} (M)" }
  end

  trait :female do
    name   "Jane Doe"
    gender "Female"
    login { "#{name} (F)" }
  end

  trait :admin do
    admin true
    login { "admin-#{name}" }
  end

  factory :male_admin,   traits: [:male, :admin]   # login will be "admin-John Doe"
  factory :female_admin, traits: [:admin, :female] # login will be "Jane Doe (F)"
end

You can also override individual attributes granted by a trait in subclasses.

factory :user do
  name "Friendly User"
  login { name }

  trait :male do
    name   "John Doe"
    gender "Male"
    login { "#{name} (M)" }
  end

  factory :brandon do
    male
    name "Brandon"
  end
end

Traits can also be passed in as a list of symbols when you construct an instance from factory_girl.

factory :user do
  name "Friendly User"

  trait :male do
    name   "John Doe"
    gender "Male"
  end

  trait :admin do
    admin true
  end
end

# creates an admin user with gender "Male" and name "Jon Snow"
create(:user, :admin, :male, name: "Jon Snow")

This ability works with build, build_stubbed, attributes_for, and create.

create_list and build_list methods are supported as well. Just remember to pass the number of instances to create/build as second parameter, as documented in the "Building or Creating Multiple Records" section of this file.

factory :user do
  name "Friendly User"

  trait :admin do
    admin true
  end
end

# creates 3 admin users with gender "Male" and name "Jon Snow"
create_list(:user, 3, :admin, :male, name: "Jon Snow")

Traits can be used with associations easily too:

factory :user do
  name "Friendly User"

  trait :admin do
    admin true
  end
end

factory :post do
  association :user, :admin, name: 'John Doe'
end

# creates an admin user with name "John Doe"
create(:post).user

When you're using association names that're different than the factory:

factory :user do
  name "Friendly User"

  trait :admin do
    admin true
  end
end

factory :post do
  association :author, :admin, factory: :user, name: 'John Doe'
  # or
  association :author, factory: [:user, :admin], name: 'John Doe'
end

# creates an admin user with name "John Doe"
create(:post).author

Finally, traits can be used within other traits to mix in their attributes.

FactoryGirl.define do
  factory :order do
    trait :completed do
      completed_at { 3.days.ago }
    end

    trait :refunded do
      completed
      refunded_at { 1.day.ago }
    end
  end
end

Callbacks

factory_girl makes available four callbacks for injecting some code:

  • after(:build) - called after a factory is built (via FactoryGirl.build, FactoryGirl.create)
  • before(:create) - called before a factory is saved (via FactoryGirl.create)
  • after(:create) - called after a factory is saved (via FactoryGirl.create)
  • after(:stub) - called after a factory is stubbed (via FactoryGirl.build_stubbed)

Examples:

# Define a factory that calls the generate_hashed_password method after it is built
factory :user do
  after(:build) { |user| generate_hashed_password(user) }
end

Note that you'll have an instance of the user in the block. This can be useful.

You can also define multiple types of callbacks on the same factory:

factory :user do
  after(:build)  { |user| do_something_to(user) }
  after(:create) { |user| do_something_else_to(user) }
end

Factories can also define any number of the same kind of callback. These callbacks will be executed in the order they are specified:

factory :user do
  after(:create) { this_runs_first }
  after(:create) { then_this }
end

Calling create will invoke both after_build and after_create callbacks.

Also, like standard attributes, child factories will inherit (and can also define) callbacks from their parent factory.

Multiple callbacks can be assigned to run a block; this is useful when building various strategies that run the same code (since there are no callbacks that are shared across all strategies).

factory :user do
  callback(:after_stub, :before_create) { do_something }
  after(:stub, :create) { do_something_else }
  before(:create, :custom) { do_a_third_thing }
end

To override callbacks for all factories, define them within the FactoryGirl.define block:

FactoryGirl.define do
  after(:build) { |object| puts "Built #{object}" }
  after(:create) { |object| AuditLog.create(attrs: object.attributes) }

  factory :user do
    name "John Doe"
  end
end

You can also call callbacks that rely on Symbol#to_proc:

# app/models/user.rb
class User < ActiveRecord::Base
  def confirm!
    # confirm the user account
  end
end

# spec/factories.rb
FactoryGirl.define do
  factory :user do
    after :create, &:confirm!
  end
end

create(:user) # creates the user and confirms it

Modifying factories

If you're given a set of factories (say, from a gem developer) but want to change them to fit into your application better, you can modify that factory instead of creating a child factory and adding attributes there.

If a gem were to give you a User factory:

FactoryGirl.define do
  factory :user do
    full_name "John Doe"
    sequence(:username) { |n| "user#{n}" }
    password "password"
  end
end

Instead of creating a child factory that added additional attributes:

FactoryGirl.define do
  factory :application_user, parent: :user do
    full_name     "Jane Doe"
    date_of_birth { 21.years.ago }
    gender        "Female"
    health        90
  end
end

You could modify that factory instead.

FactoryGirl.modify do
  factory :user do
    full_name     "Jane Doe"
    date_of_birth { 21.years.ago }
    gender        "Female"
    health        90
  end
end

When modifying a factory, you can change any of the attributes you want (aside from callbacks).

FactoryGirl.modify must be called outside of a FactoryGirl.define block as it operates on factories differently.

A caveat: you can only modify factories (not sequences or traits) and callbacks still compound as they normally would. So, if the factory you're modifying defines an after(:create) callback, you defining an after(:create) won't override it, it'll just get run after the first callback.

Building or Creating Multiple Records

Sometimes, you'll want to create or build multiple instances of a factory at once.

built_users   = build_list(:user, 25)
created_users = create_list(:user, 25)

These methods will build or create a specific amount of factories and return them as an array. To set the attributes for each of the factories, you can pass in a hash as you normally would.

twenty_year_olds = build_list(:user, 25, date_of_birth: 20.years.ago)

There's also a set of *_pair methods for creating two records at a time:

built_users   = build_pair(:user) # array of two built users
created_users = create_pair(:user) # array of two created users

Custom Construction

If you want to use factory_girl to construct an object where some attributes are passed to initialize or if you want to do something other than simply calling new on your build class, you can override the default behavior by defining initialize_with on your factory. Example:

# user.rb
class User
  attr_accessor :name, :email

  def initialize(name)
    @name = name
  end
end

# factories.rb
sequence(:email) { |n| "person#{n}@example.com" }

factory :user do
  name "Jane Doe"
  email

  initialize_with { new(name) }
end

build(:user).name # Jane Doe

Although factory_girl is written to work with ActiveRecord out of the box, it can also work with any Ruby class. For maximum compatibiltiy with ActiveRecord, the default initializer builds all instances by calling new on your build class without any arguments. It then calls attribute writer methods to assign all the attribute values. While that works fine for ActiveRecord, it actually doesn't work for almost any other Ruby class.

You can override the initializer in order to:

  • Build non-ActiveRecord objects that require arguments to initialize
  • Use a method other than new to instantiate the instance
  • Do crazy things like decorate the instance after it's built

When using initialize_with, you don't have to declare the class itself when calling new; however, any other class methods you want to call will have to be called on the class explicitly.

For example:

factory :user do
  name "John Doe"

  initialize_with { User.build_with_name(name) }
end

You can also access all public attributes within the initialize_with block by calling attributes:

factory :user do
  ignore do
    comments_count 5
  end

  name "John Doe"

  initialize_with { new(attributes) }
end

This will build a hash of all attributes to be passed to new. It won't include ignored attributes, but everything else defined in the factory will be passed (associations, evalued sequences, etc.)

You can define initialize_with for all factories by including it in the FactoryGirl.define block:

FactoryGirl.define do
  initialize_with { new("Awesome first argument") }
end

When using initialize_with, attributes accessed from within the initialize_with block are assigned only in the constructor; this equates to roughly the following code:

FactoryGirl.define do
  factory :user do
    initialize_with { new(name) }

    name { 'value' }
  end
end

build(:user)
# runs
User.new('value')

This prevents duplicate assignment; in versions of factory_girl before 4.0, it would run this:

FactoryGirl.define do
  factory :user do
    initialize_with { new(name) }

    name { 'value' }
  end
end

build(:user)
# runs
user = User.new('value')
user.name = 'value'

Custom Strategies

There are times where you may want to extend behavior of factory_girl by adding a custom build strategy.

Strategies define two methods: association and result. association receives a FactoryGirl::FactoryRunner instance, upon which you can call run, overriding the strategy if you want. The second method, result, receives a FactoryGirl::Evaluation instance. It provides a way to trigger callbacks (with notify), object or hash (to get the result instance or a hash based on the attributes defined in the factory), and create, which executes the to_create callback defined on the factory.

To understand how factory_girl uses strategies internally, it's probably easiest to just view the source for each of the four default strategies.

Here's an example of composing a strategy using FactoryGirl::Strategy::Create to build a JSON representation of your model.

class JsonStrategy
  def initialize
    @strategy = FactoryGirl.strategy_by_name(:create).new
  end

  delegate :association, to: :@strategy

  def result(evaluation)
    @strategy.result(evaluation).to_json
  end
end

For factory_girl to recognize the new strategy, you can register it:

FactoryGirl.register_strategy(:json, JsonStrategy)

This allows you to call

FactoryGirl.json(:user)

Finally, you can override factory_girl's own strategies if you'd like by registering a new object in place of the strategies.

Custom Callbacks

Custom callbacks can be defined if you're using custom strategies:

class JsonStrategy
  def initialize
    @strategy = FactoryGirl.strategy_by_name(:create).new
  end

  delegate :association, to: :@strategy

  def result(evaluation)
    result = @strategy.result(evaluation)
    evaluation.notify(:before_json, result)

    result.to_json.tap do |json|
      evaluation.notify(:after_json, json)
      evaluation.notify(:make_json_awesome, json)
    end
  end
end

FactoryGirl.register_strategy(:json, JsonStrategy)

FactoryGirl.define do
  factory :user do
    before(:json)                { |user| do_something_to(user) }
    after(:json)                 { |user_json| do_something_to(user_json) }
    callback(:make_json_awesome) { |user_json| do_something_to(user_json) }
  end
end

Custom Methods to Persist Objects

By default, creating a record will call save! on the instance; since this may not always be ideal, you can override that behavior by defining to_create on the factory:

factory :different_orm_model do
  to_create { |instance| instance.persist! }
end

To disable the persistence method altogether on create, you can skip_create for that factory:

factory :user_without_database do
  skip_create
end

To override to_create for all factories, define it within the FactoryGirl.define block:

FactoryGirl.define do
  to_create { |instance| instance.persist! }


  factory :user do
    name "John Doe"
  end
end

ActiveSupport Instrumentation

In order to track what factories are created (and with what build strategy), ActiveSupport::Notifications are included to provide a way to subscribe to factories being run. One example would be to track factories based on a threshold of execution time.

ActiveSupport::Notifications.subscribe("factory_girl.run_factory") do |name, start, finish, id, payload|
  execution_time_in_seconds = finish - start

  if execution_time_in_seconds >= 0.5
    $stderr.puts "Slow factory: #{payload[:name]} using strategy #{payload[:strategy]}"
  end
end

Another example would be tracking all factories and how they're used throughout your test suite. If you're using RSpec, it's as simple as adding a before(:suite) and after(:suite):

config.before(:suite) do
  @factory_girl_results = {}
  ActiveSupport::Notifications.subscribe("factory_girl.run_factory") do |name, start, finish, id, payload|
    factory_name = payload[:name]
    strategy_name = payload[:strategy]
    @factory_girl_results[factory_name] ||= {}
    @factory_girl_results[factory_name][strategy_name] ||= 0
    @factory_girl_results[factory_name][strategy_name] += 1
  end
end

config.after(:suite) do
  puts @factory_girl_results
end

Rails Preloaders and RSpec

When running RSpec with a Rails preloader such as spring or zeus, it's possible to encounter an ActiveRecord::AssociationTypeMismatch error when creating a factory with associations, as below:

FactoryGirl.define do
  factory :united_states, class: Location do
    name 'United States'
    association :location_group, factory: :north_america
  end

  factory :north_america, class: LocationGroup do
    name 'North America'
  end
end

The error occurs during the run of the test suite:

Failure/Error: united_states = create(:united_states)
ActiveRecord::AssociationTypeMismatch:
  LocationGroup(#70251250797320) expected, got LocationGroup(#70251200725840)

The two possible solutions are to either run the suite without the preloader, or to add FactoryGirl.reload to the RSpec configuration, like so:

RSpec.configure do |config|
  config.before(:suite) { FactoryGirl.reload }
end

Using Without Bundler

If you're not using Bundler, be sure to have the gem installed and call:

require 'factory_girl'

Once required, assuming you have a directory structure of spec/factories or test/factories, all you'll need to do is run

FactoryGirl.find_definitions

If you're using a separate directory structure for your factories, you can change the definition file paths before trying to find definitions:

FactoryGirl.definition_file_paths = %w(custom_factories_directory)
FactoryGirl.find_definitions

If you don't have a separate directory of factories and would like to define them inline, that's possible as well:

require 'factory_girl'

FactoryGirl.define do
  factory :user do
    name 'John Doe'
    date_of_birth { 21.years.ago }
  end
end
Something went wrong with that request. Please try again.