`airflow.providers.openlineage.plugins.facets`¶

Module Contents¶

Classes¶

`AirflowMappedTaskRunFacet`	Run facet containing information about mapped tasks.
`AirflowJobFacet`	Composite Airflow job facet.
`AirflowStateRunFacet`	Airflow facet providing state information.
`AirflowRunFacet`	Composite Airflow run facet.
`UnknownOperatorInstance`	Describes an unknown operator.
`UnknownOperatorAttributeRunFacet`	RunFacet that describes unknown operators in an Airflow DAG.

class airflow.providers.openlineage.plugins.facets.AirflowMappedTaskRunFacet[source]¶

Bases: openlineage.client.facet.BaseFacet

Run facet containing information about mapped tasks.

mapIndex: int[source]¶

operatorClass: str[source]¶

classmethod from_task_instance(task_instance)[source]¶

class airflow.providers.openlineage.plugins.facets.AirflowJobFacet[source]¶

Bases: openlineage.client.facet.BaseFacet

Composite Airflow job facet.

This facet encapsulates all the necessary information to re-create full scope of an Airflow DAG logic, enabling reconstruction, visualization, and analysis of DAGs in a comprehensive manner. It includes detailed representations of the tasks, task groups, and their hierarchical relationships, making it possible to draw a graph that visually represents the entire DAG structure (like in Airflow UI). It also indicates whether a task should emit an OpenLineage (OL) event, enabling consumers to anticipate the number of events and identify the tasks from which they can expect these events.

Attributes:: taskTree: A dictionary representing the hierarchical structure of tasks in the DAG. taskGroups: A dictionary that contains information about task groups within the DAG. tasks: A dictionary detailing individual tasks within the DAG.

taskTree: dict[source]¶

taskGroups: dict[source]¶

tasks: dict[source]¶

class airflow.providers.openlineage.plugins.facets.AirflowStateRunFacet[source]¶

Bases: openlineage.client.facet.BaseFacet

Airflow facet providing state information.

This facet is designed to be sent at a completion event, offering state information about the DAG run and each individual task. This information is crucial for understanding the execution flow and comprehensive post-run analysis and debugging, including why certain tasks did not emit events, which can occur due to the use of control flow operators like the BranchOperator.

Attributes:: dagRunState: This indicates the final status of the entire DAG run (e.g., “success”, “failed”). tasksState: A dictionary mapping task IDs to their respective states. (e.g., “failed”, “skipped”).