def test_self_check(mock_gcs_conn, mock_list_keys, mock_emit):
my_data_connector: InferredAssetGCSDataConnector = InferredAssetGCSDataConnector(
name="my_data_connector",
datasource_name="FAKE_DATASOURCE_NAME",
execution_engine=PandasExecutionEngine(),
default_regex={
"pattern": r"(.+)-(\d+)\.csv",
"group_names": ["data_asset_name", "number"],
},
bucket_or_name="test_bucket",
prefix="",
)
my_data_connector._refresh_data_references_cache()
self_check_report_object = my_data_connector.self_check()
assert self_check_report_object == {
"class_name": "InferredAssetGCSDataConnector",
"data_asset_count": 2,
"example_data_asset_names": ["A", "B"],
"data_assets": {
"A": {
"example_data_references": ["A-100.csv", "A-101.csv"],
"batch_definition_count": 2,
},
"B": {
"example_data_references": ["B-1.csv", "B-2.csv"],
"batch_definition_count": 2,
},
},
"example_unmatched_data_references": [],
"unmatched_data_reference_count": 0,
}
def test_complex_regex_example_with_implicit_data_asset_names(
mock_gcs_conn, mock_list_keys, mock_emit):
my_data_connector: InferredAssetGCSDataConnector = InferredAssetGCSDataConnector(
name="my_data_connector",
datasource_name="FAKE_DATASOURCE_NAME",
execution_engine=PandasExecutionEngine(),
default_regex={
"pattern": r"(\d{4})/(\d{2})/(.+)-\d+\.csv",
"group_names": ["year_dir", "month_dir", "data_asset_name"],
},
bucket_or_name="test_bucket",
prefix="",
)
my_data_connector._refresh_data_references_cache()
assert (len(
my_data_connector.get_batch_definition_list_from_batch_request(
batch_request=BatchRequest(
datasource_name="FAKE_DATASOURCE_NAME",
data_connector_name="my_data_connector",
data_asset_name="alpha",
))) == 3)
assert (len(
my_data_connector.get_batch_definition_list_from_batch_request(
batch_request=BatchRequest(
datasource_name="FAKE_DATASOURCE_NAME",
data_connector_name="my_data_connector",
data_asset_name="beta",
))) == 4)
assert my_data_connector.get_batch_definition_list_from_batch_request(
batch_request=BatchRequest(
datasource_name="FAKE_DATASOURCE_NAME",
data_connector_name="my_data_connector",
data_asset_name="alpha",
data_connector_query={
"batch_filter_parameters": {
"year_dir": "2020",
"month_dir": "03",
}
},
)) == [
BatchDefinition(
datasource_name="FAKE_DATASOURCE_NAME",
data_connector_name="my_data_connector",
data_asset_name="alpha",
batch_identifiers=IDDict(
year_dir="2020",
month_dir="03",
),
)
]
def test_get_batch_definition_list_from_batch_request_with_nonexistent_datasource_name_raises_error(
mock_gcs_conn, mock_list_keys, mock_emit,
empty_data_context_stats_enabled):
my_data_connector = InferredAssetGCSDataConnector(
name="my_data_connector",
datasource_name="FAKE_DATASOURCE_NAME",
execution_engine=PandasExecutionEngine(),
default_regex={
"pattern": r"(.+)/(.+)-(\d+)\.csv",
"group_names": ["data_asset_name", "letter", "number"],
},
bucket_or_name="test_bucket",
prefix="",
)
# Raises error in `DataConnector._validate_batch_request()` due to `datasource_name` in BatchRequest not matching DataConnector `datasource_name`
with pytest.raises(ValueError):
my_data_connector.get_batch_definition_list_from_batch_request(
BatchRequest(
datasource_name="something",
data_connector_name="my_data_connector",
data_asset_name="something",
))
def test_instantiation_without_args(mock_gcs_conn, mock_list_keys,
expected_config_dict):
my_data_connector = InferredAssetGCSDataConnector(
name="my_data_connector",
datasource_name="FAKE_DATASOURCE_NAME",
execution_engine=PandasExecutionEngine(),
default_regex={
"pattern": r"(.+)/(.+)-(\d+)\.csv",
"group_names": ["data_asset_name", "letter", "number"],
},
bucket_or_name="test_bucket",
prefix="",
)
assert my_data_connector.self_check() == expected_config_dict
my_data_connector._refresh_data_references_cache()
assert my_data_connector.get_data_reference_list_count() == 4
assert my_data_connector.get_unmatched_data_references() == []
def test_get_batch_definition_list_from_batch_request_with_unknown_data_connector_raises_error(
mock_gcs_conn, mock_list_keys, mock_emit):
my_data_connector: InferredAssetGCSDataConnector = InferredAssetGCSDataConnector(
name="my_data_connector",
datasource_name="FAKE_DATASOURCE_NAME",
execution_engine=PandasExecutionEngine(),
default_regex={
"pattern": r"(\d{4})/(\d{2})/(.+)-\d+\.csv",
"group_names": ["year_dir", "month_dir", "data_asset_name"],
},
bucket_or_name="test_bucket",
prefix="",
)
my_data_connector._refresh_data_references_cache()
# Raises error in `DataConnector._validate_batch_request()` due to `data-connector_name` in BatchRequest not matching DataConnector name
with pytest.raises(ValueError):
my_data_connector.get_batch_definition_list_from_batch_request(
batch_request=BatchRequest(
datasource_name="FAKE_DATASOURCE_NAME",
data_connector_name="non_existent_data_connector",
data_asset_name="my_data_asset",
))