PostgreSQL

By default, LineaPy uses SQLite for artifact store, which keeps the package light and simple. Given the limitations of SQLite (e.g., single write access to a database at a time), however, we may want to use a more advanced database such as PostgreSQL.

Run PostgreSQL with Docker

The easiest way to use PostgreSQL with LineaPy is through Docker. We can spin up a PostgreSQL instance with Docker using the following command:

$ docker run --name lineaPostgres -p 15432:5432 -e POSTGRES_USER=postgresuser -e POSTGRES_PASSWORD=postgrespwd -e POSTGRES_DB=postgresdb -d postgres

where

  • --name lineaPostgres sets the Docker container name as lineaPostgres

  • -p 15432:5432 exposes PostgreSQL at port 15432 (5432 is the default PostgreSQL port within the Docker image)

  • -e POSTGRES_USER=postgresuser sets the username as postgresuser

  • -e POSTGRES_PASSWORD=postgrespwd sets the password as postgrespwd

  • -e POSTGRES_DB=postgresdb sets the database name as postgresdb

  • -d postgres specifies the name of the official PostgreSQL Docker image

To validate whether the Docker container has been started successfully, run:

$ docker ps -a | grep lineaPostgres

which will show the container information if it exists, as the following:

1b68ae97e029   postgres   "docker-entrypoint.s…"   6 hours ago   Up 6 hours   0.0.0.0:15432->5432/tcp   lineaPostgres

Note

You may experience trouble launching the container if the container name or port is already occupied. If so, simply change the container name or port and relaunch the container as instructed above. Or, you can remove the conflicting container with docker rm -f CONTAINER_ID where CONTAINER_ID is the ID of the conflicting container to remove (e.g., 1b68ae97e029 above).

Install Database Adapter

To support interaction between Python and PostgreSQL, we need to install a database adapter. A popular choice is a Python package called psycopg2, which can be installed as follows:

$ pip install psycopg2

Connect LineaPy with PostgreSQL

Now that the new database is in place, we need to make LineaPy recognize and use it. We can do this by exporting the database connection string into an environmental variable LINEAPY_DATABASE_URL, like so:

$ export LINEAPY_DATABASE_URL=postgresql://postgresuser:[email protected]:15432/postgresdb

Note that this has to be done prior to using LineaPy so that the environment variable exists in runtime.

Note

If you want to use PostgreSQL as the default backend, you can make the environment variable persist across sessions by defining it in .bashrc or .zshrc.

You can check the connection between LineaPy and PostgreSQL with:

>>> from lineapy.db.db import RelationalLineaDB
>>> print(RelationalLineaDB.from_environment().url)

which will print:

postgresql://postgresuser:[email protected]:15432/postgresdb

if successful. Otherwise, it will default back to SQLite and print:

sqlite:///.lineapy/db.sqlite

Known Issues

If you are using PostgreSQL as your database, you might encounter the following error:

NoSuchModuleError: Can't load plugin: sqlalchemy.dialects:postgres

This is caused by a change in SQLAlchemy where they dropped support for DB URLs of the form postgres://. Using postgresql:// instead should fix this error.