Rails fixtures vs factories for test data

Maya Patel Jan 2026

2 tabs

FactoryBot.define do
  factory :post do
    association :author, factory: :user
    sequence(:title) { |n| "Post Title #{n}" }
    body { Faker::Lorem.paragraphs(number: 3).join("\n\n") }
    status { :draft }
    published_at { nil }

    trait :published do
      status { :published }
      published_at { 1.day.ago }
    end

    trait :with_comments do
      after(:create) do |post|
        create_list(:comment, 3, post: post)
      end
    end

    trait :with_tags do
      after(:create) do |post|
        post.tag_list.add('rails', 'ruby', 'testing')
        post.save
      end
    end
  end
end

require 'test_helper'

class PostTest < ActiveSupport::TestCase
  test 'published scope returns only published posts' do
    published = create(:post, :published)
    draft = create(:post)

    assert_includes Post.published, published
    refute_includes Post.published, draft
  end

  test 'validates title presence' do
    post = build(:post, title: nil)
    refute post.valid?
    assert_includes post.errors[:title], "can't be blank"
  end

  test 'can have multiple comments' do
    post = create(:post, :with_comments)
    assert_equal 3, post.comments.count
  end
end

2 files · ruby Explain with highlit

Fixtures and factories both create test data, but suit different needs. Fixtures load YAML files into the database before tests, providing fast, consistent data. I use fixtures for static reference data like countries or categories. Factories (via FactoryBot) build objects programmatically with attributes and associations. They're more flexible for test-specific scenarios and maintain referential integrity automatically. Factories support traits for variations and sequences for unique values. I prefer factories for most tests since they're more maintainable and explicit. Fixtures excel for large, static datasets. The key is consistency—mixing both approaches causes confusion. For React integration tests, I mock API responses with MSW instead of seeding data.