Prisma Migration and Seeding

Prisma is a powerful ORM (Object-Relational Mapping) tool for Node.js that makes working with databases a breeze. Prisma provides an easy way to define your database schema, generate migrations, and seed your database with initial data. In this guide, we'll cover migrations and seeding using Prisma, and explore how they work in a Docker environment.

Migrations and Seeding Overview

• Migrations: Prisma migrations allow you to evolve your database schema over time by applying changes incrementally.
• Seeding: Seeding is the process of populating your database with initial or test data to help with development or testing.

Wonder How Migrations and Seeding Work in Docker?

When we ran npm run docker:compose, Prisma automatically executed migrations to populate our User table, and also seeded it with initial data. This was made possible by a script called entrypoint-dev.sh, which was called from the Dockerfile.dev.

The entrypoint file is needed because it defines the series of commands that should be executed when the container starts. In this case, it makes sure that our Prisma migrations and seeding are applied as soon as the Docker container spins up.

Practicing Migration and Seeding

Let's practice generating a new migration and seeding data. I won't go too deep into the Prisma syntax, but this should give you a clear idea of the steps involved.

Step 1: Update Your Prisma Schema

Navigate to your Prisma schema file at prisma/schema.prisma and add an email column to the User model.

model User {
  id        Int      @id @default(autoincrement())
  username  String
  email     String?  @unique
  createdAt DateTime @default(now())
  updatedAt DateTime @updatedAt
}

Step 2: Generate Prisma Client

Exit the terminal running your backend, and run:

npm run prisma:generate

This command generates the Prisma client with the latest schema changes, ensuring that your TypeScript code on the host machine is in sync with the updated schema.

Step 3: Update Your Seeding Script

Now, let's update the seeding script in prisma/seed.ts to add email values to our users:

Note: If you don't want to seed data every time you run Docker Compose, you can simply comment out the npx prisma db seed command in the entrypoint-dev.sh file.

async function createUsers() {
  const usernames = ['goat', 'strawberry'];
  const emails = ['goat@example.com', 'strawberry@example.com'];
 
  for (let i = 0; i < usernames.length; i++) {
    await prisma.user.create({
      data: {
        username: usernames[i],
        email: emails[i],
      },
    });
  }
}

Note: The line await prisma.user.deleteMany(); deletes all user data every time we run docker-compose up. If you don't want to lose existing data, simply comment out this line.

Step 4: Update the Entry Point Script

Go to the entrypoint-dev.sh file in your root directory, and update the migration command as follows:

npx prisma migrate dev --name added_email

This command creates a new migration with the name added_email that adds the email field to the User model.

Step 5: Run Docker Compose

Finally, run the docker compose command:

npm run docker:compose

It will stop at the point shown in the following warning:

This is because we added a @unique constraint in the schema, and Prisma is asking for your confirmation to continue. To do this, we need to interact with Docker by entering the Y command. To proceed, follow these steps:

1. Open a new terminal and type docker ps to list the currently running containers.

2. Type docker attach <CONTAINER ID of express-server> i.e. docker attach 33e27b4273e7

3. Type y to confirm. Your migration will be applied, and finally, the seeding process will proceed.

4. A new migration file will be generated under prisma/migrations in both the host and the Docker container.

• Host:

• Docker container (Files tab -> usr -> src -> app):

Step 6: Check your endpoint

Open a browser and go to /v1/users (opens in a new tab); the seeded data will be displayed with updated schema.

Wrap-Up

And there you have it! You've updated your schema, applied a new migration, regenerated Prisma client files, and seeded your database. These steps allow you to easily keep your schema, migrations, and data in sync during development.