Review comments and fix some tests

airflow/config_templates/config.yml

@@ -2657,11 +2657,28 @@ usage_data_collection:
 dag_bundles:
   description: |
     Configuration for the DAG bundles. This allows Airflow to load DAGs from different sources.
+
+    Airflow will consume all options added to this section. Below you will see only the default,
+    ``dags_folder``. The option name is the bundle name and the value is a json object with the following
+    keys:
+      - classpath: The classpath of the bundle class
+      - kwargs: The keyword arguments to pass to the bundle class
+      - refresh_interval: The interval in seconds to refresh the bundle from its source.
+
+    For example, to add a new bundle named ``hello`` to my Airflow instance, add the following to your
+    airflow.cfg (this is just an example, the classpath and kwargs are not real):
+
+    .. code-block:: ini
+
+        [dag_bundles]
+        hello: {classpath: "airflow.some.classpath", kwargs: {"hello": "world"}, refresh_interval: 60}
   options:
     dags_folder:
       description: |
-        This is the default DAG bundle that loads DAGs from the traditional `[core] dags_folder`.
-        It can be disabled by setting it to an empty string.
+        This is the default DAG bundle that loads DAGs from the traditional ``[core] dags_folder``.
+        By default, ``refresh_interval`` it set to ``[scheduler] dag_dir_list_interval``, but that can be
+        overridden here if desired.
+        Parsing DAGs from the DAG folder can be disabled by setting this option to an empty string.
       version_added: ~
       type: string
       example: ~

airflow/dag_processing/bundles/manager.py

@@ -32,7 +32,7 @@
 
 
 class DagBundlesManager(LoggingMixin):
-    """Manager for Dag Bundles."""
+    """Manager for DAG bundles."""
 
     @property
     def bundle_configs(self) -> dict[str, dict]:

airflow/models/dagbundle.py

@@ -23,7 +23,15 @@
 
 
 class DagBundleModel(Base):
-    """A table for DAG Bundle information."""
+    """
+    A table for storing DAG bundle metadata.
+
+    We track the following information about each bundle, as it can be useful for
+    informational purposes and for debugging:
+     - enabled: Is the bundle currently found in configuration?
+     - latest_version: The latest version Airflow has seen for the bundle.
+     - last_refreshed: When the bundle was last refreshed.
+    """
 
     __tablename__ = "dag_bundle"
     name = Column(StringID(), primary_key=True)

tests/dag_processing/bundles/test_dag_bundle_manager.py

@@ -128,6 +128,7 @@ def clear_db():
     clear_db_dag_bundles()
 
 
+@pytest.mark.db_test
 def test_sync_bundles_to_db(clear_db):
     bundle_manager = DagBundlesManager()