🚀 elevata 1.3.1 — Airflow Example Stabilization

This release focuses on stabilizing the Airflow example environment introduced
in 1.3.0 and making it significantly easier to validate elevata orchestration
against real-world target systems.

The Airflow example is intended as a quick way to test lineage-driven execution
without requiring additional infrastructure or complex setup. Version 1.3.1
ensures that this works reliably across different environments and database targets.

🛠️ What changed

The Airflow setup now installs backend-specific dependencies at container startup
instead of baking them into the image. This prevents dependency conflicts with
Airflow itself and allows switching between backends simply by changing:

ELEVATA_SQL_DIALECT

in the .env file.

The example now works out of the box with the supported elevata targets:

• BigQuery
• Databricks
• DuckDB
• Fabric Warehouse
• MSSQL
• PostgreSQL
• Snowflake

without modifying Dockerfiles or rebuilding images.

💡 Why this matters

Airflow and database adapters frequently depend on different versions of
core libraries such as SQLAlchemy. Installing backend dependencies in an
isolated runtime environment ensures that:

• Airflow remains stable
• elevata execution remains deterministic
• users can test orchestration against their own target systems immediately

👥 Who should update

Anyone using the Airflow example introduced in 1.3.0 should upgrade to 1.3.1.

No changes to elevata projects or dataset definitions are required.

Full Changelog: v1.3.0...v1.3.1