GENERATE.md

Generate SQLAlchemy models from existing DB

Prerequisites:

• A running ClassTranscribe PostgreSQL Database
• Docker and or Python/pip

Run a Docker container with Python 3 (if necessary):

$ docker run --net=host -it --rm -v $(pwd):/usr/app -w /usr/app python:3 bash

Wherever pip and python are available, install sqlacodegen and the psycopg2 pre-built binary:

% pip install psycopg2_binary sqlacodegen==3.0.0b2   # Needed to support SQLA > v1.3

NOTE: this specific (beta) version is currently needed to work with SQLAlchemy

The psycopg2 tool allows us to connect to postgres, and the sqlacodegen utility will translate your existing PostgreSQL database tables into usable SQLAlchemy models:

% sqlacodegen postgresql+psycopg2://${POSTGRES_USER}:${POSTGRES_PASSWORD}@localhost:5432/ct2019db --outfile ./entities.py