Reputation: 1203
I am running a FastAPI application with PostgreSQL database and SQLAlchemy. I used sessionmaker
to create and use database sessions, and I have a custom alembic migration script that needs to be performed on the database.
from app.core.config import DATABASE_URL
from app.models import User, Passwords
from app.db import Base
from alembic import context
config = context.config
config.set_main_option('sqlalchemy.url', DATABASE_URL)
fileConfig(config.config_file_name)
target_metadata = Base.metadata
def run_migrations_offline():
url = config.get_main_option("sqlalchemy.url")
context.configure(
url=url,
target_metadata=target_metadata,
literal_binds=True,
dialect_opts={"paramstyle": "named"},
)
with context.begin_transaction():
context.run_migrations()
I want to be able to create a test database, apply the same migration, and teardown after the test. I have tried the following in the conftest.py
# Apply migrations at beginning and end of testing session
@pytest.fixture(scope="session")
def apply_migrations():
warnings.filterwarnings("ignore", category=DeprecationWarning)
environ["TEST"] = "True"
config = Config("alembic.ini")
alembic.command.upgrade(config, "head")
yield
alembic.command.downgrade(config, "base")
@pytest.fixture(scope="session")
def create_test_db():
url = f"{str(DATABASE_URL)}_test"
if database_exists(url):
drop_database(url)
create_database(url)
apply_migrations()
engine = create_engine(url, echo=True)
yield
drop_database(url)
But it does not seem to do anything. I am new to testing and I can't find a decent guide on how to set this up properly. So any kind of help would be greatly appreciated.
Upvotes: 8
Views: 10355
Reputation: 31
I'm not sure about putting apply_migrations()
in the DB creation - as that is driven by Alembic and you're here trying to setup the environment for running Alembic (i.e., that looks circular to me), but for just setting up the DB:
@pytest.fixture(scope="session")
def alembic_engine():
set_up_empty_test_db()
...
Where set_up_empty_test_db()
is a function that does what its name implies. In my case that function runs a docker container that starts Postgres and then creates a blank/empty database.
See https://pytest-alembic.readthedocs.io/en/latest/setup.html#setting-up-fixtures
Upvotes: 0