Skip to content

About

Who This Is For

This manual is for Python developers and data engineers who connect to multiple database backends — PostgreSQL, Snowflake, BigQuery, DuckDB, and others — or Apache Iceberg catalogs, and want a single, consistent API instead of per-database boilerplate. You might be building data pipelines that need to run unchanged against dev, staging, and production backends, or you might be a platform engineer standardizing connectivity for every warehouse in the organisation. If you write code that talks to more than one database, this is for you.

What You Should Already Know

You should be comfortable with:

  • Python at an intermediate level — protocols, dataclasses, and decorators in particular
  • Basic SQL and database concepts: schemas, tables, catalogs
  • At least one Python database client library (e.g. psycopg2, snowflake-connector-python, duckdb)
  • General ideas around connection pooling and credential management

What You'll Get Out of This

After working through this manual, you'll know how to:

  • Connect to any supported database with a typed settings object and start querying immediately
  • Switch between dev, staging, and production backends by swapping a single settings object — no code changes
  • Inspect catalogs, namespaces, tables, and columns through a uniform metadata API that works the same across every backend
  • Configure authentication for real-world setups: OAuth2, IAM, service accounts, certificates
  • Extend the library with new database backends by implementing the Backend protocol and registering settings classes
  • Use IcebergBackend to connect to Apache Iceberg catalogs (REST, Hive, Glue) alongside your SQL backends

How to Navigate

  • Read in order — chapters are arranged in dependency order, so each builds on what came before
  • Use search — the search bar (top right) indexes every page; use it to jump to a specific class or concept
  • Try the MicroSims — interactive simulations are the fastest way to build intuition for how the pieces fit together
  • Check the Learning Graph — the Learning Graph shows how concepts relate and where each fits in the bigger picture

About Mountainash Data

Mountainash Data is the intelligent textbook for mountainash-data, a unified Python library for connecting to any supported database or lakehouse. It wraps every Ibis dialect — DuckDB, PostgreSQL, Snowflake, BigQuery, Trino, ClickHouse, and more — behind a single IbisBackend class, and provides IcebergBackend for Apache Iceberg catalog access. A runtime-checkable Backend protocol gives every backend the same connect, inspect, and query surface, and typed settings classes with auto-registration handle connection configuration and authentication.

Within the mountainash ecosystem, mountainash-data owns the physical layer — connections, authentication, schema inspection, and table enumeration. It deliberately does not build queries or expressions; that logical layer belongs to mountainash-expressions, which consumes the connections and metadata this library produces.

This manual explains the design as much as the API: why there is a Backend protocol instead of a class hierarchy, why settings classes auto-register instead of being wired up by hand, and why inspection returns frozen dataclasses. Understanding those decisions is what lets you extend the library or debug a connection with confidence, rather than just calling methods you don't fully trust.

Source code and issues: github.com/mountainash-io/mountainash-data

License

Copyright © 2026 Nathaniel Ramm. Licensed under CC BY-NC-SA 4.0 for non-commercial use — see the License page for the full terms. Commercial licensing is reserved by the copyright holder; see Contact for inquiries.

Author

Nathaniel Ramm