orm-loader¶

A lightweight, reusable foundation for building and validating SQLAlchemy-based data models.

orm-loader provides infrastructure for SQLAlchemy-based data models. It is the shared plumbing layer, not the place where model-specific rules live.

It focuses on:

ORM table introspection
safe bulk ingestion patterns
file-based loading via staging tables
model-agnostic validation scaffolding
operational helpers for supported backends

It currently ships with backend implementations for SQLite and PostgreSQL.

Core Concepts¶

Tables are structural — semantics live downstream
Mixins define capabilities
Protocols decouple infrastructure from implementations
Ingestion is explicit and staged

API Reference¶

Design Philosophy¶

orm-loader is intentionally conservative. It gives downstream libraries the machinery to load, inspect, and validate data without deciding what the data means.

The library is designed to sit below:

OMOP CDM (initial scope)
extension to O3, MCODE, etc.
custom clinical schemas
research data marts

and above:

raw SQLAlchemy
database-specific ingestion scripts

What this library does not do¶

No domain validation
No schema enforcement
No migrations
No concurrency guarantees
No support yet for arbitrary database dialects

Why mixins and protocols?¶

Mixins provide reusable behaviour.

Protocols provide: - static typing - runtime structural checks - decoupling between infrastructure and models

This allows downstream libraries to: - replace base classes - mock implementations - incrementally adopt features