def test_validate(self):
catalog = os.path.join(SAMPLES_DIR, "data.json")
catalog = DataJson(catalog)
distrib_meta = catalog.get_distribution(identifier="125.1")
df = pd.read_csv(distrib_meta["downloadURL"],
parse_dates=["indice_tiempo"
]).set_index("indice_tiempo")
dataset_meta = catalog.get_dataset(
identifier=distrib_meta["dataset_identifier"])
validate_distribution(df, catalog, dataset_meta, distrib_meta)
def test_repeated_field_id(self):
catalog = os.path.join(SAMPLES_DIR, "repeated_field_id.json")
catalog = DataJson(catalog)
identifier = "125.1"
distribution = catalog.get_distribution(identifier=identifier)
dataset = catalog.get_dataset(
identifier=distribution["dataset_identifier"])
df = pd.read_csv(distribution["downloadURL"],
parse_dates=["indice_tiempo"
]).set_index("indice_tiempo")
validate_distribution(df, catalog, dataset, distribution)
def run(self, distribution_model: Distribution, catalog: DataJson):
"""
Valida las distribuciones de series de tiempo de un catálogo
entero a partir de su URL, o archivo fuente
Returns:
bool: True si la distribución pasa las validaciones, False caso contrario
"""
df = self.init_df(distribution_model)
dataset_id = distribution_model.dataset.identifier
if dataset_id is None:
raise ValueError(
NO_DATASET_IDENTIFIER.format(distribution_model.identifier))
dataset = catalog.get_dataset(dataset_id)
distribution = catalog.get_distribution(distribution_model.identifier)
validate_distribution(df, catalog, dataset, distribution)
return True
