def test_dataset_default(configuration_data):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("some_new_key.csv", "some_bucket")
ds = config.dataset(dataset_file)
assert ds.data_frequency == "D"
assert ds.dataset_type == DatasetType.TARGET_TIME_SERIES
assert ds.dataset_domain == DatasetDomain.RETAIL
assert ds.dataset_name == "some_new_key"
assert ds.dataset_schema == {
"Attributes": [
{
"AttributeName": "item_id",
"AttributeType": "string"
},
{
"AttributeName": "timestamp",
"AttributeType": "timestamp",
},
{
"AttributeName": "demand",
"AttributeType": "float"
},
]
}
def test_dataset_import_timestamp_format_none(configuration_data,
forecast_stub):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("RetailDemandTRM.csv", "some_bucket")
dataset = config.dataset(dataset_file)
forecast_stub.add_response("list_dataset_import_jobs",
{"DatasetImportJobs": []})
dataset.cli = forecast_stub.client
assert dataset.timestamp_format == None
def test_dataset_status_lifecycle(configuration_data, forecast_stub):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("RetailDemandTRM.csv", "some_bucket")
dataset = config.dataset(dataset_file)
forecast_stub.add_client_error("describe_dataset", "ResourceNotFoundException")
forecast_stub.add_response("describe_dataset", {"Status": "ACTIVE"})
dataset.cli = forecast_stub.client
assert dataset.status == Status.DOES_NOT_EXIST
assert dataset.status == "ACTIVE"
def test_dataset_create_noop_errors(configuration_data, forecast_stub):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("RetailDemandTRM.csv", "some_bucket")
dataset = config.dataset(dataset_file)
configured_dataset = configuration_data.get("RetailDemandTRM").get(
"Datasets")[2]
params = {
"DatasetType": configured_dataset.get("DatasetType"),
"DatasetName": "RetailDemandTRM",
"Domain": configured_dataset.get("Domain"),
"Schema": configured_dataset.get("Schema"),
"DataFrequency": configured_dataset.get("DataFrequency"),
}
create_params = deepcopy(params)
create_params["Tags"] = [{"Key": "SolutionId", "Value": "SOL0123"}]
forecast_stub.add_response(
"describe_dataset",
params,
)
forecast_stub.add_response("create_dataset", {"DatasetArn": dataset.arn},
create_params)
forecast_stub.add_response(
"describe_dataset",
params,
)
dataset.cli = forecast_stub.client
dataset.create()
# clobber the values to trigger some exceptions
# this is likey caused by a user changing configuration unexpectedly
dataset._dataset_type = DatasetType.RELATED_TIME_SERIES
dataset._dataset_domain = DatasetDomain.WORK_FORCE
dataset._data_frequency = DataFrequency("1min")
dataset._dataset_schema = {}
with pytest.raises(ValueError) as excinfo:
dataset.create()
assert "dataset type" in str(excinfo.value)
assert "dataset domain" in str(excinfo.value)
assert "data frequency" in str(excinfo.value)
assert "dataset schema" in str(excinfo.value)
def test_dataset_create(configuration_data, forecast_stub):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("RetailDemandTRM.csv", "some_bucket")
dataset = config.dataset(dataset_file)
configured_dataset = configuration_data.get("RetailDemandTRM").get("Datasets")[2]
forecast_stub.add_client_error("describe_dataset", "ResourceNotFoundException")
forecast_stub.add_response("create_dataset", {"DatasetArn": "arn:"})
# should not call anything
dataset.cli = forecast_stub.client
dataset.create()
forecast_stub.assert_no_pending_responses()
def test_dataset_import_timestamp_format(configuration_data, forecast_stub,
format):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("RetailDemandTRM.csv", "some_bucket")
dataset = config.dataset(dataset_file)
forecast_stub.add_response(
"list_dataset_import_jobs",
{
"DatasetImportJobs": [{
"DatasetImportJobArn": "arn:something",
"LastModificationTime": datetime(2015, 1, 1),
}]
},
)
forecast_stub.add_response("describe_dataset_import_job",
{"TimestampFormat": format})
dataset.cli = forecast_stub.client
assert dataset.timestamp_format == format
def test_dataset_create_noop_errors(configuration_data, forecast_stub):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("RetailDemandTRM.csv", "some_bucket")
dataset = config.dataset(dataset_file)
configured_dataset = configuration_data.get("RetailDemandTRM").get("Datasets")[2]
for i in range(0, 2):
forecast_stub.add_response(
"describe_dataset",
{
"DatasetType": configured_dataset.get("DatasetType"),
"DatasetName": "RetailDemandTRM",
"Domain": configured_dataset.get("Domain"),
"Schema": configured_dataset.get("Schema"),
"DataFrequency": configured_dataset.get("DataFrequency"),
},
)
# should not call anything
dataset.cli = forecast_stub.client
dataset.create()
# clobber the values to trigger some exceptions
# this is likey caused by a user changing configuration unexpectedly
dataset._dataset_type = DatasetType.RELATED_TIME_SERIES
dataset._dataset_domain = DatasetDomain.WORK_FORCE
dataset._data_frequency = DataFrequency("1min")
dataset._dataset_schema = {}
with pytest.raises(ValueError) as excinfo:
dataset.create()
assert "dataset type" in str(excinfo.value)
assert "dataset domain" in str(excinfo.value)
assert "data frequency" in str(excinfo.value)
assert "dataset schema" in str(excinfo.value)
def test_dataset_imports(configuration_data, forecast_stub):
config = Config()
config.config = configuration_data
dataset_file = DatasetFile("RetailDemandTRM.csv", "some_bucket")
dataset = config.dataset(dataset_file)
forecast_stub.add_response(
"list_dataset_import_jobs",
{
"DatasetImportJobs": [
{
"DatasetImportJobArn": "arn::",
"DatasetImportJobName": "middle_job",
"LastModificationTime": datetime(2018, 1, 1),
},
{
"DatasetImportJobArn": "arn::",
"DatasetImportJobName": "end_job",
"LastModificationTime": datetime(2019, 1, 1),
},
{
"DatasetImportJobArn": "arn::",
"DatasetImportJobName": "early_job",
"LastModificationTime": datetime(2017, 1, 1),
},
]
},
)
dataset.cli = forecast_stub.client
ds_imports = dataset.imports
assert ds_imports[0].get("DatasetImportJobName") == "end_job"
assert ds_imports[1].get("DatasetImportJobName") == "middle_job"
assert ds_imports[2].get("DatasetImportJobName") == "early_job"