Build multi-tenant RAG with Neon's database-per-user model — no nosy neighbors, max isolation, minimal costs
Docs/Integrations/SQLAlchemy

Schema migration with Neon Postgres and SQLAlchemy

Manage database migrations in your Python project with SQLAlchemy and Alembic

SQLAlchemy is a popular SQL toolkit and Object-Relational Mapping (ORM) library for Python. SQLAlchemy provides a powerful way to interact with databases and manage database schema changes using Alembic, a lightweight database migration tool.

This guide demonstrates how to use SQLAlchemy/Alembic to manage schema migrations for a Neon Postgres database. We create a simple API using the FastAPI web framework and define database models using SQLAlchemy. We then generate and run migrations to manage schema changes over time.

Prerequisites

To follow along with this guide, you will need:

  • A Neon account. If you do not have one, sign up at Neon. Your Neon project comes with a ready-to-use Postgres database named neondb. We'll use this database in the following examples.
  • Python installed on your local machine. We recommend using a newer version of Python, 3.8 or higher.

Setting up your Neon database

Initialize a new project

  1. Log in to the Neon Console and navigate to the Projects section.
  2. Select a project or click the New Project button to create a new one.

Retrieve your Neon database connection string

Navigate to the Connection Details section to find your database connection string. It should look similar to this:

postgresql://alex:AbC123dEf@ep-cool-darkness-123456.us-east-2.aws.neon.tech/dbname?sslmode=require

Keep your connection string handy for later use.

note

Neon supports both direct and pooled database connection strings, which can be copied from the Connection Details widget on your Neon Project Dashboard. A pooled connection string connects your application to the database via a PgBouncer connection pool, allowing for a higher number of concurrent connections. However, using a pooled connection string for migrations can be prone to errors. For this reason, we recommend using a direct (non-pooled) connection when performing migrations. For more information about direct and pooled connections, see Connection pooling.

Setting up the Web application

Set up the Python environment

To manage our project dependencies, we create a new Python virtual environment. Run the following commands in your terminal to set it up.

python -m venv myenv

Activate the virtual environment by running the following command:

# On macOS and Linux
source myenv/bin/activate

# On Windows
myenv\Scripts\activate

With the virtual environment activated, we can create a new directory for our FastAPI project and install the required packages:

mkdir guide-neon-sqlalchemy && cd guide-neon-sqlalchemy
pip install sqlalchemy alembic "psycopg2-binary"
pip install fastapi uvicorn python-dotenv
pip freeze > requirements.txt

We installed SQLAlchemy, Alembic, and the psycopg2-binary package to connect to the Neon Postgres database. We the installed the FastAPI package to create the API endpoints and uvicorn as the web server. We then saved the installed packages to a requirements.txt file so the project can be easily recreated in another environment.

Set up the Database configuration

Create a .env file in the project root directory and add the DATABASE_URL environment variable to it. Use the connection string that you obtained from the Neon Console earlier:

# .env
DATABASE_URL=NEON_POSTGRES_CONNECTION_STRING

We create an app directory at the project root to store the database models and configuration files.

mkdir app
touch guide-neon-sqlalchemy/app/__init__.py

Next, create a new file named database.py in the app subdirectory and add the following code:

# app/database.py

import os

import dotenv
from sqlalchemy import create_engine
from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy.orm import sessionmaker

dotenv.load_dotenv()
SQLALCHEMY_DATABASE_URL = os.getenv("DATABASE_URL")

engine = create_engine(SQLALCHEMY_DATABASE_URL)
SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)

Base = declarative_base()

This code sets up the database connection using SQLAlchemy. It reads the DATABASE_URL environment variable, creates a database engine, and defines a SessionLocal class for database sessions. The Base class is used as a base class for defining database models.

Defining data models and running migrations

Specify the data model

Create a new file named models.py in the app subdirectory and define the database models for your application:

# app/models.py

from sqlalchemy import Column, Integer, String, Text, DateTime, ForeignKey
from sqlalchemy.orm import relationship
from sqlalchemy.sql import func

from .database import Base

class Author(Base):
    __tablename__ = "authors"

    id = Column(Integer, primary_key=True, index=True)
    name = Column(String(100), nullable=False)
    bio = Column(Text)
    created_at = Column(DateTime(timezone=True), server_default=func.now())

    books = relationship("Book", back_populates="author")

class Book(Base):
    __tablename__ = "books"

    id = Column(Integer, primary_key=True, index=True)
    title = Column(String(200), nullable=False)
    author_id = Column(Integer, ForeignKey("authors.id"), nullable=False)
    created_at = Column(DateTime(timezone=True), server_default=func.now())

    author = relationship("Author", back_populates="books")

This code defines two models: Author and Book. The Author model represents an author with fields for name, bio, and a created_at timestamp. The Book model represents a book with fields for title, author (as a foreign key to the Author model), and a created_at timestamp. The relationship function is used to define the one-to-many relationship between Author and Book.

Initialize Alembic

To initialize Alembic for managing database migrations, run the following command in your terminal:

alembic init alembic

This command creates a new directory named alembic with the necessary files for managing migrations. Open the env.py file in the alembic directory and update the target_metadata variable to include the models defined in the models.py file:

# alembic/env.py

from app.models import Base

target_metadata = Base.metadata

We update the alembic/env.py file again to load the database URL from the .env file at project root and set it as the sqlalchemy.url configuration option.

# alembic/env.py

import dotenv
import os

dotenv.load_dotenv()

config.set_main_option('sqlalchemy.url', os.getenv('DATABASE_URL', ""))

Generate the initial migration

To generate the initial migration based on the defined models, run the following command:

alembic revision --autogenerate -m "init-setup"

This command detects the Author and Book models and generates a new migration file in the alembic/versions directory.

Apply the migration

To apply the migration and create the corresponding tables in the Neon Postgres database, run the following command:

alembic upgrade head

This command executes the migration file and creates the necessary tables in the database.

Seed the database

To seed the database with some initial data, create a new file named seed.py in the project root and add the following code:

# seed.py

from database import SessionLocal
from models import Author, Book

def seed_data():
    db = SessionLocal()

    # Create authors
    authors = [
        Author(
            name="J.R.R. Tolkien",
            bio="The creator of Middle-earth and author of The Lord of the Rings."
        ),
        Author(
            name="George R.R. Martin",
            bio="The author of the epic fantasy series A Song of Ice and Fire."
        ),
        Author(
            name="J.K. Rowling",
            bio="The creator of the Harry Potter series."
        ),
    ]
    db.add_all(authors)
    db.commit()

    # Create books
    books = [
        Book(title="The Fellowship of the Ring", author=authors[0]),
        Book(title="The Two Towers", author=authors[0]),
        Book(title="The Return of the King", author=authors[0]),
        Book(title="A Game of Thrones", author=authors[1]),
        Book(title="A Clash of Kings", author=authors[1]),
        Book(title="Harry Potter and the Philosopher's Stone", author=authors[2]),
        Book(title="Harry Potter and the Chamber of Secrets", author=authors[2]),
    ]
    db.add_all(books)
    db.commit()

    print("Data seeded successfully.")

if __name__ == "__main__":
    seed_data()

Now, run the seed.py script to seed the database with the initial data:

python seed.py

Implement the web application

Create API endpoints

Create a file named main.py in the project root directory and define the FastAPI application with endpoints for interacting with authors and books:

# main.py

from fastapi import FastAPI, Depends
from sqlalchemy.orm import Session
import uvicorn

from app.models import Author, Book, Base
from app.database import SessionLocal, engine

Base.metadata.create_all(bind=engine)

app = FastAPI()

def get_db():
    db = SessionLocal()
    try:
        yield db
    finally:
        db.close()

@app.get("/authors/")
def read_authors(db: Session = Depends(get_db)):
    authors = db.query(Author).all()
    return authors


@app.get("/books/{author_id}")
def read_books(author_id: int, db: Session = Depends(get_db)):
    books = db.query(Book).filter(Book.author_id == author_id).all()
    return books

if __name__ == "__main__":
    uvicorn.run(app, host="127.0.0.1", port=8000)

This code defines endpoints for creating and retrieving authors and books. It uses SQLAlchemy's Session to interact with the database and Pydantic models (schemas) for request and response data validation and serialization.

Run the FastAPI server

To start the FastAPI server using uvicorn and test the application, run the following command:

python main.py

Now, you can navigate to http://localhost:8000/authors in your browser to view the list of authors. To view the books by a specific author, navigate to http://localhost:8000/books/{author_id} where {author_id} is the ID of the author.

Applying schema changes

Let's demonstrate how to handle schema changes by adding a new field country to the Author model, to store the author's country of origin.

Update the data model

Open the models.py file and add a new field to the Author model:

# models.py
class Author(Base):
    __tablename__ = "authors"

    id = Column(Integer, primary_key=True, index=True)
    name = Column(String(100), nullable=False)
    bio = Column(Text)
    country = Column(String(100))
    created_at = Column(DateTime(timezone=True), server_default=func.now())

    books = relationship("Book", back_populates="author")

Generate and run the migration

To generate a new migration file for the schema change, run the following command:

alembic revision --autogenerate -m "add-country-to-author"

This command detects the updated Author model and generates a new migration file to add the new field to the corresponding table in the database.

Now, to apply the migration, run the following command:

alembic upgrade head

Test the schema change

Restart the FastAPI development server.

python main.py

Navigate to http://localhost:8000/authors in your browser to view the list of authors. You should see the new country field included in each author's record, reflecting the schema change.

Conclusion

In this guide, we demonstrated how to set up a FastAPI project with Neon Postgres, define database models using SQLAlchemy, generate migrations using Alembic, and run them. Alembic makes it easy to interact with the database and manage schema evolution over time.

Source code

You can find the source code for the application described in this guide on GitHub.

Resources

For more information on the tools and concepts used in this guide, refer to the following resources:

Need help?

Join our Discord Server to ask questions or see what others are doing with Neon. Users on paid plans can open a support ticket from the console. For more details, see Getting Support.

Last updated on

Was this page helpful?