airflow-config

Declarative Apache Airflow DAGs from Hydra YAML and airflow-pydantic models.

# config/pipeline.yaml
# @package _global_
_target_: airflow_config.Configuration

dags:
  daily-report:
    schedule: "0 6 * * *"
    start_date: "2025-01-01"
    catchup: false
    tasks:
      report:
        _target_: airflow_config.BashTask
        bash_command: python /opt/jobs/report.py

from airflow_config import load_config

config = load_config("config", "pipeline")
config.generate_in_mem()

Configuration can provide shared defaults, environment overrides, templates, extensions, generated Python DAG files, or in-memory DAG registration.

Documentation

• Tutorial: generate a DAG from YAML

• How-to guides

• Why configuration and DAG models are separate

• API reference

Published documentation is available at airflow-laminar.github.io/airflow-config.

Ecosystem

• airflow-pydantic supplies the DAG, task, sensor, schedule, pool, and variable models.

• airflow-supervisor and supervisor-pydantic manage supervisord jobs.

• airflow-systemd and systemd-pydantic manage systemd services.

• airflow-cron and cron-pydantic convert crontabs into DAG models.

• airflow-balancer supplies host and port selection extensions.

• airflow-common supplies common operators and library management.

Note

This library was generated using copier from the Base Python Project Template repository.