Rename dataset related python variable names to asset (apache#41348)

abhishekbhakat · Sep 30, 2024 · ede7cb2 · ede7cb2
1 parent 2eecca1
commit ede7cb2
Show file tree

Hide file tree

Showing 199 changed files with 5,061 additions and 4,127 deletions.
diff --git a/RELEASE_NOTES.rst b/RELEASE_NOTES.rst
@@ -642,7 +642,7 @@ Dataset URIs are now validated on input (#37005)
 
 Datasets must use a URI that conform to rules laid down in AIP-60, and the value
 will be automatically normalized when the DAG file is parsed. See
-`documentation on Datasets <https://airflow.apache.org/docs/apache-airflow/stable/authoring-and-scheduling/datasets.html>`_ for
+`documentation on Datasets <https://airflow.apache.org/docs/apache-airflow/2.9.0/authoring-and-scheduling/datasets.html>`_ for
 a more detailed description on the rules.
 
 You may need to change your Dataset identifiers if they look like a URI, but are
@@ -3264,7 +3264,7 @@ If you have the producer and consumer in different files you do not need to use
 Datasets represent the abstract concept of a dataset, and (for now) do not have any direct read or write
 capability - in this release we are adding the foundational feature that we will build upon.
 
-For more info on Datasets please see :doc:`/authoring-and-scheduling/datasets`.
+For more info on Datasets please see `Datasets documentation <https://airflow.apache.org/docs/apache-airflow/2.4.0/authoring-and-scheduling/datasets.html>`_.
 
 Expanded dynamic task mapping support
 """""""""""""""""""""""""""""""""""""

diff --git a/airflow/__init__.py b/airflow/__init__.py
@@ -55,7 +55,7 @@
 __all__ = [
     "__version__",
     "DAG",
-    "Dataset",
+    "Asset",
     "XComArg",
 ]
 
@@ -76,7 +76,7 @@
 # Things to lazy import in form {local_name: ('target_module', 'target_name', 'deprecated')}
 __lazy_imports: dict[str, tuple[str, str, bool]] = {
     "DAG": (".models.dag", "DAG", False),
-    "Dataset": (".datasets", "Dataset", False),
+    "Asset": (".assets", "Asset", False),
     "XComArg": (".models.xcom_arg", "XComArg", False),
     "version": (".version", "", False),
     # Deprecated lazy imports
@@ -86,8 +86,8 @@
     # These objects are imported by PEP-562, however, static analyzers and IDE's
     # have no idea about typing of these objects.
     # Add it under TYPE_CHECKING block should help with it.
+    from airflow.models.asset import Asset
     from airflow.models.dag import DAG
-    from airflow.models.dataset import Dataset
     from airflow.models.xcom_arg import XComArg
 
 

diff --git a/airflow/api_connexion/endpoints/dag_run_endpoint.py b/airflow/api_connexion/endpoints/dag_run_endpoint.py
@@ -39,6 +39,10 @@
     format_datetime,
     format_parameters,
 )
+from airflow.api_connexion.schemas.asset_schema import (
+    AssetEventCollection,
+    asset_event_collection_schema,
+)
 from airflow.api_connexion.schemas.dag_run_schema import (
     DAGRunCollection,
     DAGRunCollectionSchema,
@@ -50,10 +54,6 @@
     set_dagrun_note_form_schema,
     set_dagrun_state_form_schema,
 )
-from airflow.api_connexion.schemas.dataset_schema import (
-    DatasetEventCollection,
-    dataset_event_collection_schema,
-)
 from airflow.api_connexion.schemas.task_instance_schema import (
     TaskInstanceReferenceCollection,
     task_instance_reference_collection_schema,
@@ -112,12 +112,12 @@ def get_dag_run(
 
 
 @security.requires_access_dag("GET", DagAccessEntity.RUN)
-@security.requires_access_dataset("GET")
+@security.requires_access_asset("GET")
 @provide_session
 def get_upstream_dataset_events(
     *, dag_id: str, dag_run_id: str, session: Session = NEW_SESSION
 ) -> APIResponse:
-    """If dag run is dataset-triggered, return the dataset events that triggered it."""
+    """If dag run is dataset-triggered, return the asset events that triggered it."""
     dag_run: DagRun | None = session.scalar(
         select(DagRun).where(
             DagRun.dag_id == dag_id,
@@ -130,8 +130,8 @@ def get_upstream_dataset_events(
             detail=f"DAGRun with DAG ID: '{dag_id}' and DagRun ID: '{dag_run_id}' not found",
         )
     events = dag_run.consumed_dataset_events
-    return dataset_event_collection_schema.dump(
-        DatasetEventCollection(dataset_events=events, total_entries=len(events))
+    return asset_event_collection_schema.dump(
+        AssetEventCollection(dataset_events=events, total_entries=len(events))
     )