def test_delete_columns_no_data(self):
"""
Negative test
data: None
Checks that delete_columns raises a TypeError when no data is passed.
"""
# 1. Arrange
# 2. Act & 3. Assert
with self.assertRaises(TypeError):
delete_columns()
def test_delete_columns_inplace(self):
"""
Positive test
Checks that the dataframe has 3 columns after applying
delete_columns on it.
"""
# 1. Arrange
df = generate_example_df_divcols()
# 2. Act
delete_columns(df, inplace=True)
# 3. Assert
self.assertTrue(len(df.columns) == 3)
def test_delete_columns_wrong_datatype(self):
"""
Negative test
data: Series (unsupported datatype)
Checks that delete_columns raises a TypeError when the data is passed
as a series.
"""
# 1. Arrange
ser = generate_example_series()
# 2. Act & 3. Assert
with self.assertRaises(TypeError):
delete_columns(ser, inplace=True)
def test_delete_columns_wrong_column(self):
"""
Negative test
columns: ['d', 'e', 'z'] (z doesn't exist as a column in the data)
Checks that delete_columns raises a ValueError when a column name is
passed that doesn't exist in the dataframe.
"""
# 1. Arrange
df = generate_example_df_divcols()
# 2. Act & 3. Assert
with self.assertRaises(ValueError):
delete_columns(df, columns=['d', 'e', 'z'], inplace=True)
def test_delete_columns_threshold(self):
"""
Positive test
threshold: 8
Checks that the dataframe has 6 columns after applying
delete_columns on it. Columns c and h contain less than 8 non-NA
values.
"""
# 1. Arrange
df = generate_example_df_divcols()
# 2. Act
delete_columns(df, threshold=8, inplace=True)
# 3. Assert
self.assertTrue(len(df.columns) == 6)
def test_delete_columns_columns_specified(self):
"""
Positive test
columns: ['d', 'e']
Checks that the dataframe has 7 columns after applying
delete_columns on it. The column d doesn't have any NA values and
should therefore not be deleted.
"""
# 1. Arrange
df = generate_example_df_divcols()
# 2. Act
delete_columns(df, columns=['d', 'e'], inplace=True)
# 3. Assert
self.assertTrue(len(df.columns) == 7)
def test_delete_columns_threshold_and_columns_specified(self):
"""
Positive test
columns: ['h']
threshold: 8
Checks that the dataframe has 7 columns after applying
delete_columns on it. Columns c and h contain less than 8 non-NA
values but only column h is being considered.
"""
# 1. Arrange
df = generate_example_df_divcols()
# 2. Act
delete_columns(df, columns=['h'], threshold=8, inplace=True)
# 3. Assert
self.assertTrue(len(df.columns) == 7)
def test_delete_columns_returning(self):
"""
Positive test
Checks that the original dataframe still has 10 columns and
therefore has not been modified.
Checks that the returned dataframe has 3 columns.
"""
# 1. Arrange
df = generate_example_df_divcols()
# 2. Act
df2 = delete_columns(df)
# 3. Assert
self.assertTrue(len(df.columns) == 8)
self.assertTrue(len(df2.columns) == 3)