Files
2022-01-04 16:01:14 +01:00

55 lines
2.0 KiB
Markdown

# SQLAlchemy
**FastAPI Users** provides the necessary tools to work with SQL databases thanks to [SQLAlchemy ORM with asyncio](https://docs.sqlalchemy.org/en/14/orm/extensions/asyncio.html).
!!! warning
The previous adapter using `encode/databases` is now deprecated but can still be installed using `fastapi-users[sqlalchemy]`.
## Asynchronous driver
To work with your DBMS, you'll need to install the corresponding asyncio driver. The common choices are:
* For PostgreSQL: `pip install asyncpg`
* For SQLite: `pip install aiosqlite`
For the sake of this tutorial from now on, we'll use a simple SQLite databse.
## Setup User table
Let's declare our SQLAlchemy `User` table.
```py hl_lines="15 16"
--8<-- "docs/src/db_sqlalchemy.py"
```
As you can see, **FastAPI Users** provides a mixin that will include base fields for our `User` table. You can of course add you own fields there to fit to your needs!
## Implement a function to create the tables
We'll now create an utility function to create all the defined tables.
```py hl_lines="23-25"
--8<-- "docs/src/db_sqlalchemy.py"
```
This function can be called, for example, during the initialization of your FastAPI app.
!!! warning
In production, it's strongly recommended to setup a migration system to update your SQL schemas. See [Alembic](https://alembic.sqlalchemy.org/en/latest/).
## Create the database adapter dependency
The database adapter of **FastAPI Users** makes the link between your database configuration and the users logic. It should be generated by a FastAPI dependency.
```py hl_lines="28-34"
--8<-- "docs/src/db_sqlalchemy.py"
```
Notice that we define first a `get_async_session` dependency returning us a fresh SQLAlchemy session to interact with the database.
It's then used inside the `get_user_db` dependency to generate our adapter. Notice that we pass it three things:
* A reference to your [`UserDB` model](../models.md).
* The `session` instance we just injected.
* The `UserTable` variable, which is the actual SQLAlchemy model.