Connect to PostgreSQL Using SQLAlchemy & Python

Learn how to connect to a PostgreSQL database using Python and SQLAlchemy. SQLAlchemy allows us to abstract away much of the implementation so we can work with Python to manage our database objects.

You can download the code at the Analyzing Alpha GitHub Repo.

First, import the necessary imports so we can create a database engine and a session from that engine.

from sqlalchemy import create_engine
from sqlalchemy.orm import sessionmaker
from sqlalchemy_utils import database_exists, create_database
from local_settings import postgresql as settings

You’ll notice local_settings contains the PostgreSQL settings. It’s just a dictionary file and you can see an example in the GitHub repo.

Create a get_engine function to make our lives easier.

def get_engine(user, passwd, host, port, db):
    url = f"postgresql://{user}:{passwd}@{host}:{port}/{db}"
    if not database_exists(url):
    engine = create_engine(url, pool_size=50, echo=False)
    return engine

Get the credentials from the local_settings import above.

engine = get_engine(settings['pguser'],

Verify that the engine was created successfully.


Create a new function so that we don’t have to enter our credentials every time.

def get_engine_from_settings():
    keys = ['pguser','pgpasswd','pghost','pgport','pgdb']
    if not all(key in keys for key in settings.keys()):
        raise Exception('Bad config file')

    return get_engine(settings['pguser'],

Create a session function that binds to our previously created engine.

def get_session():
    engine = get_engine_from_settings()
    session = sessionmaker(bind=engine)()
    return session

Create a new session and verify it works.

session = get_session()

Congratulations, you successfully connected to Postgres using Python and SQLALchemy.

When you’re done, make sure you close your session.


You can also close out all currently checked-in sessions.

engine = session.get_bind()
engine.dispose() # Close all checked in sessions