def test_set_up_deletes_in_not_append_mode(self):
# Create individual type CSV (called Family.csv)
target_file_path = Mapper.get_target_file_path(TargetCSVType.FAMILY,
self.test_file_path)
csvops.create_file(target_file_path, TargetCSVType.INDIVIDUAL)
Mapper.set_up(target_file_path, TargetCSVType.FAMILY, Mode.Mode.CREATE)
# Ensure that the old individual CSV was deleted and replaced with a file with Family CSV headers
self.assertTrue(
csvops.check_headers_match(target_file_path, TargetCSVType.FAMILY))
def test_set_up_does_not_delete_in_append_mode(self):
# Create individual type CSV (called Family.csv)
target_file_path = Mapper.get_target_file_path(TargetCSVType.FAMILY,
self.test_file_path)
csvops.create_file(target_file_path, TargetCSVType.INDIVIDUAL)
# Should not delete the file, if it does it won't be an INDIVIDUAL type file
Mapper.set_up(target_file_path, TargetCSVType.FAMILY, Mode.Mode.APPEND)
# Ensure that the old individual CSV was not deleted
self.assertTrue(
csvops.check_headers_match(target_file_path,
TargetCSVType.INDIVIDUAL))
def test_batch_ids_are_all_unique(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_data = self.fb.map(df, None)
unique_ids_size = batch_data['BatchID'].value_counts(dropna=True).size
ids_size = batch_data['BatchID'].values.size
self.assertEqual(ids_size, unique_ids_size)
def test_ids_are_unique(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
attribute_notes = self.anb.map(df, SourceCSVType.ATTRIBUTE_NOTES)
unique_ids_size = attribute_notes['note_id'].value_counts(
dropna=True).size
ids_size = attribute_notes['note_id'].values.size
self.assertEqual(ids_size, unique_ids_size)
def test_all_contribution_ids_are_unique(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
unique_ids_size = contributions_data['ContributionID'].value_counts(
dropna=True).size
ids_size = contributions_data['ContributionID'].values.size
self.assertEqual(ids_size, unique_ids_size)
def test_data_is_not_null(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
attribute_notes = self.anb.map(df, SourceCSVType.ATTRIBUTE_NOTES)
self.assertFalse(
pd.isnull(attribute_notes['individual_id_1']).any()
and pd.isnull(attribute_notes['attribute_group_name'].any())
and pd.isnull(attribute_notes['attribute_name'])
and pd.isnull(attribute_notes['comment']).any())
def test_batch_shared_data_columns_are_correct_types(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_date_is_date = self.fb.batch_data[
'Batch_Date'].dtype == Timestamp
batch_id_is_int = self.fb.batch_data[
'Id'].dtype == np.int64 or self.fb.batch_data['Id'].dtype == int
types_correct = batch_date_is_date and batch_id_is_int
self.assertTrue(types_correct)
def test_number_of_batches_equals_number_of_group_by_batch_name_date(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_data = self.fb.map(df, None)
ids_size = batch_data['BatchID'].values.size
# This seems weird but does work bizarrely
group_by_size = batch_data.groupby(['BatchName',
'BatchDate']).size().size
self.assertEqual(ids_size, group_by_size)
def run(source_file_path, target_file_type: TargetCSVType,
source_type: SourceCSVType):
mode = source_type.mode
set_up(source_file_path, target_file_type, mode)
data = CSVOperations.read_file_without_check(source_file_path)
builder = BuilderFactory.get_builder(target_file_type, source_type)
if builder is None:
raise Exception('No matching builder was found')
output_data = builder.map(data, source_type)
target_file_path = get_target_file_path(target_file_type, source_file_path)
if output_data is None:
return
if mode == Mode.APPEND:
CSVOperations.delete_write(target_file_path, output_data)
else:
CSVOperations.write_file(
get_target_file_path(target_file_type, source_file_path),
output_data)
def test_contributions_ids_are_ints(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
contribution_id_is_int = contributions_data[
'ContributionID'].dtype == int
contribution_batch_id_is_int = contributions_data[
'ContributionBatchID'].dtype == int
contribution_individual_id_is_int = contributions_data[
'IndividualID'].dtype == int
self.assertTrue(contribution_id_is_int and contribution_batch_id_is_int
and contribution_individual_id_is_int)
def test_number_of_shared_batch_data_equals_number_of_contributions_group_by_batch_name_date(
self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
cloned_df = df.copy()
cloned_df = self.fb.fill_missing_batch_data(cloned_df)
self.fb.map(df, None)
ids_size = self.fb.batch_data['Id'].values.size
# This seems weird but does work bizarrely
group_by_size = cloned_df.groupby(['Batch_Name',
'Batch_Date']).size().size
self.assertEqual(ids_size, group_by_size)
def set_up(source_file_path, target_file_type, mode):
target_file_path = get_target_file_path(target_file_type, source_file_path)
if not CSVOperations.check_file_exists(source_file_path):
print(source_file_path)
raise MappingFileNotFound(
'Mapping File Not Found at' + source_file_path, source_file_path)
if CSVOperations.check_file_exists(
target_file_path) and mode != Mode.APPEND:
CSVOperations.delete_file(target_file_path)
if not CSVOperations.check_file_exists(target_file_path):
CSVOperations.create_file(target_file_path, target_file_type)
def test_contributions_amount_equals_contribution_amount(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
self.assertEqual(contributions_data['Amount'].sum(), 6084.99)
def test_all_contributions_received_dates_are_dates(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
self.assertTrue(contributions_data['ReceivedDate'].dtype == Timestamp)
def test_all_check_contributions_have_check_number(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
check_contributions = contributions_data.loc[
contributions_data['ContributionTypeName'] == 'Check']
self.assertFalse(pd.isnull(check_contributions['CheckNumber']).any())
def test_batch_columns_match(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_data = self.fb.map(df, None)
npt.assert_array_equal(batch_data.columns.values,
TargetCSVType.BATCH.columns)
def test_all_contributions_have_funds(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
self.assertFalse(pd.isnull(contributions_data['FundName']).any())
def test_batch_amounts_are_all_decimal(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_data = self.fb.map(df, None)
self.assertTrue(batch_data['BatchAmount'].dtype == decimal.Decimal)
def test_read_attribute_data_has_correct_columns(self):
df = csvops.read_file_without_check(
THIS_DIR + '/testdata/A2501E_ConnectionStepsAttributes.csv')
npt.assert_array_equal(df.columns.values,
SourceCSVType.ATTRIBUTES.columns)
def test_batches_all_have_amounts(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_data = self.fb.map(df, None)
self.assertFalse(pd.isnull(batch_data['BatchAmount']).any())
def test_batch_dates_are_all_timestamps(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_data = self.fb.map(df, None)
self.assertTrue(batch_data['BatchDate'].dtype == Timestamp)
def test_date_parse_mmddyy(self):
correct_date = datetime.date(1957, 2, 23)
self.assertEqual(correct_date, csvops.parse_date("02/23/57"))
def test_all_contribution_batch_ids_have_batches(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
npt.assert_array_equal(
contributions_data['ContributionBatchID'].unique(),
self.fb.batch_data['Id'])
def test_contributions_columns_match(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
npt.assert_array_equal(contributions_data.columns.values,
TargetCSVType.CONTRIBUTION.columns)
def test_read_individual_household_has_correct_columns(self):
df = csvops.read_file_without_check(
THIS_DIR + '/testdata/X9400_no_attributes.csv')
npt.assert_array_equal(df.columns.values,
SourceCSVType.INDIVIDUAL_HOUSEHOLD.columns)
def test_batch_shared_data_columns_are_all_populated(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
self.assertFalse(self.fb.batch_data.isnull().values.any())
def test_batch_shared_data_contains_correct_number_of_batches(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
unique_size = self.fb.batch_data['Id'].value_counts(dropna=True).size
self.assertEqual(unique_size, 5)
def test_number_of_contributions(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
contributions_data = self.fb.build_contributions(df)
self.assertEquals(len(contributions_data.index), 36)
def test_batches_ids_are_all_ints(self):
df = csvops.read_file_without_check(THIS_DIR + TEST_DATA_FILENAME)
self.fb.map(df, None)
batch_data = self.fb.map(df, None)
self.assertTrue(batch_data['BatchID'].dtype == int)
def test_date_parse_mmddyyyy(self):
correct_date = datetime.date(2001, 12, 31)
self.assertEqual(correct_date, csvops.parse_date("12/31/2001"))
