def test_strings_NaNs_are_concatenated(self):
self.df["new_concat_field_ab"] = concat_fieldvalues(self.df, ['a', 'b'])
self.df["new_concat_field_abc"] = concat_fieldvalues(self.df, ['a', 'b', 'c'])
expected_result_ab = pd.DataFrame({'new_concat_field_ab': ['x y', 'y', 'x']})
expected_result_abc = pd.DataFrame({'new_concat_field_abc': ['x y z', 'y z', 'x']})
assert_series_equal(self.df["new_concat_field_ab"], expected_result_ab['new_concat_field_ab'])
assert_series_equal(self.df["new_concat_field_abc"], expected_result_abc['new_concat_field_abc'])
def test_leading_trailing_whitespaces_in_fields_are_stripped(self):
""" any leading/trailing whitespaces within a column will be stripped before concatenation occurs"""
self.df["new_concat_field_ae"] = concat_fieldvalues(self.df, ['a', 'e'])
expected_result_ae = pd.DataFrame({'new_concat_field_ae': ['x y12', 'y', 'x']})
assert_series_equal(self.df["new_concat_field_ae"], expected_result_ae['new_concat_field_ae'])
def test_concat_NaNs_returns_NaN(self):
mydf = self.df.copy(deep=True)
mydf["new_concat_field_bc"] = concat_fieldvalues(self.df, ['b', 'c'])
expected_result_bc = pd.DataFrame({'new_concat_field_bc': ['y z', 'y z', np.nan]})
assert_series_equal(mydf["new_concat_field_bc"], expected_result_bc["new_concat_field_bc"])
def test_single_field_is_returned(self):
# any fields with NaN will still come out NaN if not concatenated with other fields
self.df["new_concat_field_a"] = concat_fieldvalues(self.df, ['a'])
expected_result_ab = pd.DataFrame({'new_concat_field_a': ['x', np.nan, 'x']})
assert_series_equal(self.df["new_concat_field_a"], expected_result_ab['new_concat_field_a'])
def test_strings_NaNs_numbers_are_concatenated(self):
self.df["new_concat_field_abd"] = concat_fieldvalues(self.df, ['a', 'b', 'd'])
expected_result_abd = pd.DataFrame({'new_concat_field_abd': ['x y x', 'y', 'x 1']})
assert_series_equal(self.df["new_concat_field_abd"], expected_result_abd['new_concat_field_abd'])