airflow-cron

Generate Airflow DAG models from cron configuration.

Overview

airflow-cron converts the user-crontab models from cron-pydantic into airflow-pydantic DAG and Bash task models. Each enabled cron job becomes one DAG containing one Bash task.

from airflow_cron import CronAirflowConfiguration, create_dags

configuration = CronAirflowConfiguration.model_validate(
    {
        "environment": {"PATH": "/usr/local/bin:/usr/bin"},
        "job": {
            "backup": {
                "schedule": "0 2 * * *",
                "command": "/opt/bin/backup",
            },
            "cleanup": {
                "schedule": "@weekly",
                "command": "/opt/bin/cleanup",
                "enabled": False,
            },
        },
    }
)

dag_models = create_dags(configuration)

The result contains a backup DAG scheduled at 0 2 * * *. Its run task uses the cron command as its Bash command and appends the configured environment to the worker environment. Disabled jobs are omitted.

To expose generated models from an Airflow DAG file:

for dag_id, model in dag_models.items():
    globals()[dag_id] = model.instantiate()

Common Airflow settings can be supplied declaratively:

job:
  backup:
    schedule: "0 2 * * *"
    command: /opt/bin/backup
dag_args:
  max_active_runs: 1
  tags: [cron]
task_args:
  execution_timeout: 02:00:00

catchup defaults to false, matching cron's behavior of not backfilling missed runs. Job schedule and command always take precedence over defaults.

Compatibility

Airflow cannot faithfully represent every crontab feature. Conversion rejects:

• system crontabs because Airflow executor user selection is deployment-specific
• @reboot schedules
• Cronie random ranges using ~
• CRON_TZ and RANDOM_DELAY scheduler environment settings
• unescaped %, which cron converts into command input

@annually maps to @yearly, and @midnight maps to @daily.

[!NOTE] This library was generated using copier from the Base Python Project Template repository.