class-to-sql

Tiny helper to map simple Python class-like metadata to SQLite tables. This repo contains a minimal, easy-to-read implementation suitable for small projects and examples.

Quickstart

1. (Optional) set the database location:

export CLASS_TO_SQL_DB=$(pwd)/db/peppermint.db

2. Run the example or import in Python (ensure src/ is on PYTHONPATH or install editable):

python examples/basic.py

or in Python:

from class_to_sql.base_dataclass import Base

base = Base()
meta = {
	'id': int,
	'name': str,
	'super': {'primary_key': ['id'], 'defaults': {}}
}
Base.create_table('users', meta)
base.callback_create('users', {'id': 1, 'name': 'Alice'}, meta)

Development

Create a virtual environment and install test deps:

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pip install black ruff

Run linter/formatter checks and tests:

ruff check .
black --check .
pytest -q

Notes

• By default the package writes the SQLite DB to db/peppermint.db in the repository root when CLASS_TO_SQL_DB is not set.
• The implementation uses parameterized SQL for DML to reduce injection risks. DDL still interpolates table/column names and should be used with trusted metadata.
• Tests use a temporary DB file and set CLASS_TO_SQL_DB to keep runs hermetic.