Seed Data Generator Skill

Generate realistic test data for database development, testing, and demos.

Instructions

You are a test data generation expert. When invoked:

1. Analyze Schema:

   • Identify tables and relationships
   • Understand column types and constraints
   • Detect foreign key dependencies
   • Recognize data patterns (email, phone, dates, etc.)

2. Generate Realistic Data:

   • Use faker libraries for realistic data
   • Maintain referential integrity
   • Follow business logic constraints
   • Create diverse but realistic scenarios

3. Seed Database:

   • Insert data in correct order (respect foreign keys)
   • Handle different database systems
   • Provide both SQL and ORM-based seeders
   • Support incremental seeding

4. Customize Generation:

   • Allow quantity specification
   • Support different data scenarios (edge cases, happy path)
   • Enable data relationships customization
   • Provide reproducible seeds (with random seed values)

Supported Tools

• JavaScript/TypeScript: Faker.js, Chance.js, Casual
• Python: Faker, Factory Boy, Mimesis
• Ruby: Faker, FactoryBot
• Raw SQL: Generate INSERT statements
• ORMs: Prisma, TypeORM, Sequelize, Django, Rails

Usage Examples

@seed-data-generator
@seed-data-generator --count 100
@seed-data-generator --table users
@seed-data-generator --scenario e-commerce
@seed-data-generator --realistic-relationships

SQL Seed Data

PostgreSQL - Basic Insert

-- seed/001_users.sql
INSERT INTO users (username, email, password_hash, active, created_at)
VALUES
  ('john_doe', 'john@example.com', '$2b$10$...', true, '2024-01-15 10:00:00'),
  ('jane_smith', 'jane@example.com', '$2b$10$...', true, '2024-01-16 11:30:00'),
  ('bob_wilson', 'bob@example.com', '$2b$10$...', true, '2024-01-17 09:15:00'),
  ('alice_brown', 'alice@example.com', '$2b$10$...', false, '2024-01-18 14:45:00'),
  ('charlie_davis', 'charlie@example.com', '$2b$10$...', true, '2024-01-19 16:20:00');

-- seed/002_categories.sql
INSERT INTO categories (name, slug, parent_id)
VALUES
  ('Electronics', 'electronics', NULL),
  ('Computers', 'computers', 1),
  ('Laptops', 'laptops', 2),
  ('Desktops', 'desktops', 2),
  ('Accessories', 'accessories', 1),
  ('Clothing', 'clothing', NULL),
  ('Men', 'men', 6),
  ('Women', 'women', 6);

-- seed/003_products.sql
INSERT INTO products (name, description, price, stock_quantity, category_id, created_at)
VALUES
  (
    'MacBook Pro 16"',
    'Powerful laptop with M3 chip, 16GB RAM, 512GB SSD',
    2499.99,
    15,
    3,
    NOW()
  ),
  (
    'Dell XPS 13',
    'Compact laptop with Intel i7, 16GB RAM, 512GB SSD',
    1299.99,
    20,
    3,
    NOW()
  ),
  (
    'Gaming Desktop',
    'High-performance desktop with RTX 4080, 32GB RAM',
    2999.99,
    8,
    4,
    NOW()
  ),
  (
    'Wireless Mouse',
    'Ergonomic wireless mouse with precision tracking',
    29.99,
    100,
    5,
    NOW()
  ),
  (
    'Mechanical Keyboard',
    'RGB mechanical keyboard with Cherry MX switches',
    149.99,
    45,
    5,
    NOW()
  );

-- seed/004_orders.sql
INSERT INTO orders (user_id, total_amount, status, created_at)
VALUES
  (1, 2529.98, 'completed', '2024-01-20 10:30:00'),
  (2, 1299.99, 'completed', '2024-01-21 14:15:00'),
  (3, 179.98, 'processing', '2024-01-22 09:45:00'),
  (1, 2999.99, 'pending', '2024-01-23 16:00:00'),
  (4, 29.99, 'completed', '2024-01-24 11:20:00');

-- seed/005_order_items.sql
INSERT INTO order_items (order_id, product_id, quantity, price)
VALUES
  -- Order 1
  (1, 1, 1, 2499.99),
  (1, 4, 1, 29.99),
  -- Order 2
  (2, 2, 1, 1299.99),
  -- Order 3
  (3, 4, 1, 29.99),
  (3, 5, 1, 149.99),
  -- Order 4
  (4, 3, 1, 2999.99),
  -- Order 5
  (5, 4, 1, 29.99);

Generated Seed with Functions

-- Generate random users
CREATE OR REPLACE FUNCTION generate_users(count INTEGER)
RETURNS void AS $$
DECLARE
  i INTEGER;
BEGIN
  FOR i IN 1..count LOOP
    INSERT INTO users (username, email, password_hash, active, created_at)
    VALUES (
      'user_' || i,
      'user' || i || '@example.com',
      '$2b$10$fakehashedpassword',
      random() > 0.1, -- 90% active
      NOW() - (random() * 365 || ' days')::INTERVAL
    );
  END LOOP;
END;
$$ LANGUAGE plpgsql;

-- Generate random products
CREATE OR REPLACE FUNCTION generate_products(count INTEGER)
RETURNS void AS $$
DECLARE
  i INTEGER;
  categories INTEGER[];
BEGIN
  SELECT ARRAY_AGG(id) INTO categories FROM categories;

  FOR i IN 1..count LOOP
    INSERT INTO products (name, description, price, stock_quantity, category_id, created_at)
    VALUES (
      'Product ' || i,
      'Description for product ' || i,
      (random() * 1000 + 10)::NUMERIC(10,2),
      (random() * 100)::INTEGER,
      categories[1 + floor(random() * array_length(categories, 1))::INTEGER],
      NOW() - (random() * 180 || ' days')::INTERVAL
    );
  END LOOP;
END;
$$ LANGUAGE plpgsql;

-- Generate random orders
CREATE OR REPLACE FUNCTION generate_orders(count INTEGER)
RETURNS void AS $$
DECLARE
  i INTEGER;
  user_ids INTEGER[];
  product_ids INTEGER[];
  order_id INTEGER;
  item_count INTEGER;
  j INTEGER;
  total NUMERIC(10,2);
BEGIN
  SELECT ARRAY_AGG(id) INTO user_ids FROM users WHERE active = true;
  SELECT ARRAY_AGG(id) INTO product_ids FROM products WHERE stock_quantity > 0;

  FOR i IN 1..count LOOP
    -- Random number of items (1-5)
    item_count := 1 + floor(random() * 5)::INTEGER;
    total := 0;

    INSERT INTO orders (user_id, total_amount, status, created_at)
    VALUES (
      user_ids[1 + floor(random() * array_length(user_ids, 1))::INTEGER],
      0, -- Will update after adding items
      (ARRAY['pending', 'processing', 'completed', 'cancelled'])[1 + floor(random() * 4)::INTEGER],
      NOW() - (random() * 90 || ' days')::INTERVAL
    )
    RETURNING id INTO order_id;

    -- Add order items
    FOR j IN 1..item_count LOOP
      DECLARE
        product_price NUMERIC(10,2);
        quantity INTEGER;
      BEGIN
        SELECT price INTO product_price
        FROM products
        WHERE id = product_ids[1 + floor(random() * array_length(product_ids, 1))::INTEGER];

        quantity := 1 + floor(random() * 3)::INTEGER;

        INSERT INTO order_items (order_id, product_id, quantity, price)
        VALUES (
          order_id,
          product_ids[1 + floor(random() * array_length(product_ids, 1))::INTEGER],
          quantity,
          product_price
        );

        total := total + (product_price * quantity);
      END;
    END LOOP;

    -- Update order total
    UPDATE orders SET total_amount = total WHERE id = order_id;
  END LOOP;
END;
$$ LANGUAGE plpgsql;

-- Usage
SELECT generate_users(100);
SELECT generate_products(50);
SELECT generate_orders(200);

JavaScript/TypeScript Seeders

Faker.js + Prisma

// prisma/seed.ts
import { PrismaClient } from '@prisma/client';
import { faker } from '@faker-js/faker';

const prisma = new PrismaClient();

async function main() {
  console.log('Seeding database...');

  // Clear existing data
  await prisma.orderItem.deleteMany();
  await prisma.order.deleteMany();
  await prisma.review.deleteMany();
  await prisma.product.deleteMany();
  await prisma.category.deleteMany();
  await prisma.user.deleteMany();

  // Create categories
  const categories = await Promise.all([
    prisma.category.create({
      data: {
        name: 'Electronics',
        slug: 'electronics',
      },
    }),
    prisma.category.create({
      data: {
        name: 'Clothing',
        slug: 'clothing',
      },
    }),
    prisma.category.create({
      data: {
        name: 'Home & Garden',
        slug: 'home-garden',
      },
    }),
  ]);

  // Create users
  const users = await Promise.all(
    Array.from({ length: 20 }, async () => {
      const firstName = faker.person.firstName();
      const lastName = faker.person.lastName();

      return prisma.user.create({
        data: {
          username: faker.internet.userN