Database

The Mojodex database is a PostgreSQL database using the pg_vector extension to store and query vector data.

The database is launched as a Docker container and is accessed by the backend and background services. See general documentation for more information.

Database schema

You can have a look at the database schema at dbdiagram.io: https://dbdiagram.io/d/MojodexDatabase-659d0645ac844320ae85b440

Table names are prefixed with md_ standing for "Mojodex".

Initialization of Mojodex data

The create-mojodex-data.sql and init-mojodex-data.sql files respectively contains the SQL script to create the Mojodex schema and data. You can modify this file to add or remove data. For now, all updates are stacked in this file and no sql patch is provided.

Entities

Each table is represented as a python class in mojodex_code/entities/db_base_entities.py. This file is generated using sqlacodegentool:

sqlacodegen postgresql+psycopg2://$DBUSER:$DBPASS@localhost:5432/$DBNAME --outfile mojodex_core/entities/db_base_entities.py

⚠️ Be sure to install sqlacodegen with the appropriate pg_vector features:

python3 -m pip install git+https://github.com/hoomano/sqlacodegen.git@feature-pgvector

Some of those entities are then extended in mojodex_code/entities/ to add specific methods or attributes. To get the most out of an entity in the code, retrieve it by querying then database using the aproppriate daughter class.

Example using User entity, daughter of base entity MdUser:

db_session.query(User).get(user_id)

You can also directly create object using the daughter class:

new_user = User(name="John Doe", email="john@doe.com")
db_session.add(new_user)
db_session.commit()