def test_transform_dataset_shape_strips_whitespace(self):
dataset_shape_transformer = dataset_shape.DatasetShapeTransformer(
MEMBER_ORGANIZATION_ID,
TEST_SCHEMA,
TEST_COLUMN_MAPPING,
row_format=True,
multiple_val_delimiter=",",
)
dataset = pd.DataFrame(data={"field1": [" 1", "2 ", " 3 "]})
actual_shaped_dataset = dataset_shape_transformer.transform_dataset_shape(
dataset
)
expected_shaped_dataset = pd.DataFrame(
data={
"field1": ["1", "2", "3"],
"MemberOrganization": ["sample_id", "sample_id", "sample_id"],
"ForceOverWrite": ["1", "1", "1"],
}
)
pd.util.testing.assert_frame_equal(
expected_shaped_dataset, actual_shaped_dataset
)
def test_transform_dataset_shape_multiple_values_blank_values(self):
dataset_shape_transformer = dataset_shape.DatasetShapeTransformer(
MEMBER_ORGANIZATION_ID,
TEST_SCHEMA,
TEST_COLUMN_MAPPING,
row_format=True,
multiple_val_delimiter=",",
)
dataset = pd.DataFrame(
data={
"internal_column_name1": ["1", "2"],
"internal_column_name2": ["3", "4"],
"internal_column_name3": ["", "4"],
}
)
actual_shaped_dataset = dataset_shape_transformer.transform_dataset_shape(
dataset
)
expected_shaped_dataset = pd.DataFrame(
data={
"field1": ["1", "2"],
"field2": ["3", "4"],
"field3": [[""], ["4"]],
"MemberOrganization": ["sample_id", "sample_id"],
"ForceOverWrite": ["1", "1"],
}
)
pd.util.testing.assert_frame_equal(
expected_shaped_dataset, actual_shaped_dataset
)
def test_transform_dataset_shape_row_format_some_columns_present(self):
dataset_shape_transformer = dataset_shape.DatasetShapeTransformer(
MEMBER_ORGANIZATION_ID, TEST_SCHEMA, TEST_COLUMN_MAPPING, row_format=True
)
dataset = pd.DataFrame(
data={"internal_column_name1": [1, 2], "internal_column_name2": [3, 4]}
)
actual_shaped_dataset = dataset_shape_transformer.transform_dataset_shape(
dataset
)
expected_column_names = [
"field1",
"field2",
"MemberOrganization",
"ForceOverWrite",
]
self.assertTrue(
all(
[
a == b
for a, b in zip(
actual_shaped_dataset.columns.values, expected_column_names
)
]
)
)
def test_transform_dataset_shape_col_format(self):
dataset_shape_transformer = dataset_shape.DatasetShapeTransformer(
MEMBER_ORGANIZATION_ID, TEST_SCHEMA_COL, {}, row_format=False
)
dataset = pd.DataFrame(
data={
"field1": ["field1_1", "field1_2"],
"Intakefield2": ["field2_1", "field2_2"],
"Intakefield3": ["field3_1", "field3_2"],
"Exitfield3": ["field3_3", "field3_4"],
"actual_field4": ["field4_1", "field4_1"],
}
)
actual_shaped_dataset = dataset_shape_transformer.transform_dataset_shape(
dataset
)
expected_shaped_dataset = pd.DataFrame(
data={
"MilestoneFlag": [
"Intake",
"Intake",
"Exit",
"Exit",
"NinetyDays",
"NinetyDays",
],
"field1": [
"field1_1",
"field1_2",
"field1_1",
"field1_2",
"field1_1",
"field1_2",
],
"field2": ["field2_1", "field2_2", "", "", "", ""],
"field3": ["field3_1", "field3_2", "field3_3", "field3_4", "", ""],
"field4": ["", "", "", "", "field4_1", "field4_1"],
"MemberOrganization": [
"sample_id",
"sample_id",
"sample_id",
"sample_id",
"sample_id",
"sample_id",
],
"ForceOverWrite": ["1", "1", "1", "1", "1", "1"],
}
)
pd.util.testing.assert_frame_equal(
expected_shaped_dataset, actual_shaped_dataset
)
def test_transform_dataset_shape_datset_is_empty(self):
dataset_shape_transformer = dataset_shape.DatasetShapeTransformer(
MEMBER_ORGANIZATION_ID, TEST_SCHEMA, TEST_COLUMN_MAPPING, row_format=True
)
empty_dataset = pd.DataFrame(columns=[])
actual_shaped_dataset = dataset_shape_transformer.transform_dataset_shape(
empty_dataset
)
expected_shaped_dataset = pd.DataFrame(columns=[])
pd.util.testing.assert_frame_equal(
expected_shaped_dataset, actual_shaped_dataset
)