def test_convert_with_nested_expectations(self):
# arrange
schema = DataLoader.load_schema_with_name(
"schema_nested_expectation_json.json")
expected_schema = DataLoader.load_schema_with_name(
"schema_nested_expectation_result_json.json")
# act
result = self.parser.convert_expectations(
SchemaDefinition.create(schema, False))
# assert
self.assertStingEqualAsDict(result.schema_content, expected_schema)
def test_inspect_with_both_schema_formats(self):
# arrange
schema_json = DataLoader.load_schema_with_name("schema_diff_json.json")
schema_avro = DataLoader.load_schema_with_name("schema_diff_avro.json")
samples = DataLoader.load_samples()
# act
result_json = self.inspector.inspect(
samples, SchemaDefinition.create(schema_json, False))
result_avro = self.inspector.inspect(
samples, SchemaDefinition.create(schema_avro, False))
# assert
self.assertEqual(result_json, result_avro)
def test_inspect_with_multiple_expectations_asyncapi_style_json(self):
# arrange
schema = DataLoader.load_schema_with_name(
"schema_expectation_asyncapi_style_json.json")
samples = [
{
'random_integer': 1,
'random_string': 'id_1'
},
{
'random_integer': 2,
'random_string': 'foo'
}, # no match (string)
{
'random_integer': 3,
'random_string': 'id_3'
},
{
'random_integer': 4,
'random_string': 'id_4'
}, # no match (integer)
{
'random_integer': 5,
'random_string': 'foo'
}, # no match (integer, string)
]
# act
metrics = self.inspector.inspect(
samples, SchemaDefinition.create(schema, False))
# assert
self.assertAlmostEqual(6 / 10, metrics.attribute_integrity, 3)
def test_inspect_with_inferred_schemas(self):
# arrange
schema = DataLoader.load_schema_with_name("schema_registry_json.json")
schema_definition = SchemaDefinition.create(schema, True)
samples = DataLoader.load_samples()
# act
result = self.inspector.inspect(samples, schema_definition)
# assert
self.assertEqual(1.0, result.attribute_integrity)
self.assertEqual(.0, result.attribute_specification)
self.assertEqual(.5, result.attribute_quality_index)
def test_load_required_types_for_deeply_nested_schema(self):
# arrange
schema = DataLoader.load_schema_with_name(
"schema_registry_avro_complex.json")
schema_obj = json.loads(schema)
# act
type_definitions, _ = self.parser.load_required_types_from_schema(
schema_obj)
# assert
self.assertListEqual([
"complex/subtypeString", "complex/subtypeComplex/subtypeNumber",
"simpleNumber"
], type_definitions)
def test_load_required_types_for_deeply_nested_schema(self):
# arrange
schema = DataLoader.load_schema_with_name(
"schema_inferred_complex.json")
schema_obj = json.loads(schema)
# act
type_definitions, _ = JsonSchemaParser(
).load_required_types_from_schema(schema_obj)
# assert
# assert
self.assertListEqual(type_definitions, [
"base", "complex/type1number",
"complex/type3complex/subtype1number"
])
def test_integrity_for_complex_type(self):
# arrange
schema = DataLoader.load_schema_with_name("schema_registry_avro.json")
samples = [
{
"timestamp": 1595601702,
"iss_position": {
"longitude": "-42.2948",
"latitude": "-40.3670"
},
"message": "success"
},
{
"timestamp": 1595601702,
"iss_position": {
"latitude": "-40.3670"
},
"message": "success"
},
{
"timestamp": "wrong",
"iss_position": {
"longitude": 666,
"latitude": "-40.0283"
},
"message": "success"
},
]
# act
result = self.inspector.inspect_attributes(
samples, SchemaDefinition.create(schema, False))
# assert - only message is not mandatory so 3 out of 12 (3*4) are missing or wrong
invalid_elements = 3
all_elements = 12
expected_integrity = (all_elements - invalid_elements) / all_elements
self.assertAlmostEqual(
expected_integrity, result.attribute_integrity, 3,
f"Integrity must be {expected_integrity * 100}%")
def test_specification_from_toeggelomat(self):
# arrange
samples = DataLoader.load_samples_from_file("samples_toeggelomat.json")
# act
schema = DataLoader.load_schema_with_name("schema_toeggelomat.json")
result = self.inspector.inspect(samples,
SchemaDefinition.create(schema, False))
# assert
self.assertEqual(53, len(result.attribute_details.keys()),
"There should be 53 keys in the dictionary")
for attribute_metric in result.attribute_details.keys():
self.assertEqual(
1.0, result.attribute_details[attribute_metric].
attribute_specification,
f"Attribute specification must be 100% ({attribute_metric})")
self.assertEqual(
1.0,
result.attribute_details[attribute_metric].attribute_integrity,
f"Attribute integrity must be 100% ({attribute_metric})")
def test_inspect_with_min_max_range_expectation(self):
# arrange
schema = DataLoader.load_schema_with_name("schema_with_min_max.json")
samples = [
{
'random_integer': 3
},
{
'random_integer': 11
},
{
'random_integer': 3
},
{
'random_integer': 8
},
{
'random_integer': 3
},
{
'random_integer': -5
},
{
'random_integer': 3
},
{
'random_integer': 10
},
]
# act
metrics = self.inspector.inspect(
samples, SchemaDefinition.create(schema, False))
# assert
self.assertEqual((6 / 8), metrics.attribute_integrity,
f"Attribute integrity must be {(6 / 8) * 100}%")
