fix(ingest): fix lingering demo-data source issues (datahub-project#6659

)

metadata-ingestion/docs/sources/demo-data/demo-data_recipe.yml

@@ -0,0 +1,3 @@
+source:
+  type: demo-data
+  config: {}

metadata-ingestion/setup.py

@@ -553,7 +553,7 @@ def get_long_description():
         "presto-on-hive = datahub.ingestion.source.sql.presto_on_hive:PrestoOnHiveSource",
         "pulsar = datahub.ingestion.source.pulsar:PulsarSource",
         "salesforce = datahub.ingestion.source.salesforce:SalesforceSource",
-        "dummy-data = datahub.ingestion.source.dummy_source.DummySource",
+        "demo-data = datahub.ingestion.source.demo_data.DemoDataSource",
         "unity-catalog = datahub.ingestion.source.unity.source:UnityCatalogSource",
     ],
     "datahub.ingestion.sink.plugins": [

metadata-ingestion/src/datahub/ingestion/source/demo_data.py

@@ -0,0 +1,33 @@
+from datahub.configuration.common import ConfigModel
+from datahub.ingestion.api.common import PipelineContext
+from datahub.ingestion.api.decorators import (
+    SupportStatus,
+    config_class,
+    platform_name,
+    support_status,
+)
+from datahub.ingestion.source.file import FileSourceConfig, GenericFileSource
+from datahub.utilities.sample_data import download_sample_data
+
+
+class DemoDataConfig(ConfigModel):
+    # The demo data source does not accept any configuration.
+    pass
+
+
+@platform_name("Demo Data")
+@config_class(ConfigModel)
+@support_status(SupportStatus.UNKNOWN)
+class DemoDataSource(GenericFileSource):
+    """
+    This source loads sample data into DataHub. It is intended for demo and testing purposes only.
+    """
+
+    def __init__(self, ctx: PipelineContext, config: DemoDataConfig):
+        file_config = FileSourceConfig(filename=download_sample_data())
+        super().__init__(ctx, file_config)
+
+    @classmethod
+    def create(cls, config_dict, ctx):
+        config = DemoDataConfig.parse_obj(config_dict or {})
+        return cls(ctx, config)