PG Docker

A python package to provide containerized postgres databases in python

Why would you need this?

If you are using postgres and want to write tests that run against a real database, then this will make your life easier.

Installation

Install via pip:

pip install pg-docker

You will also need to have docker.

Usage

Note: This package is mainly built with pytest in mind, but you can use the context managers documented below with other testing frameworks as well.

Example

With pytest:

import psycopg2


pytest_plugins = ["pg_docker"]


def test_using_a_database(pg_database):
    db_connection = psycopg2.connect(**pg_database.connection_kwargs())
    cursor = db_connection.cursor()
    cursor.execute("SELECT 'hello world!'")

    assert cursor.fetchone() == ("hello world!",)

Usage with pytest

You first need to enable the plugin. To do this add a conftest.py to the root directory of your tests and add:

pytest_plugins = ["pg_docker"]

You can find more details on how to activate plugins in the pytest docs

The plugin The following fixtures:

• pg_database: DatabaseParams for a clean database.
• pg_database_pool: A DatabasePool instance. Use this if you need more than one database in your tests at a time.

Configuring Database Migrations

Use the below template in your conftest.py to configure how your databases are set up.

def setup_db(pg_params):
    """Add any setup logic for your database in here."""
    pass

@pytest.fixture(scope="session")
def pg_setup_db():
    return setup_db

Note: You might be inclined to edit the above code to nest the setup_db function inside of the fixture function. This will not work, because the fixture result needs to be pickleable!

Advanced Usage (and other testing frameworks)

For other use cases you can use the database_pool context manager:

with database_pool() as db_pool:
    with db_pool.database as db_params:
        connection = psycopg2.connect(**db_params.connection_kwargs())