def test_summary_string():
df = create_data()
results = sample._summarize_data(df)
df_result = "average length of string: " + \
str(round(np.mean([len(str(i)) for i in df['F']]), 1))
test_result = results['summary'][5]
assert df_result == test_result
def test_summary_other():
df = create_data()
results = sample._summarize_data(df)
test_result_datetime = results['summary'][1]
test_result_category = results['summary'][4]
assert test_result_datetime == 'No summary available'
assert test_result_category == 'No summary available'
def test_data_type():
df = create_data()
results = sample._summarize_data(df)
assert df.dtypes.equals(results['data_type'])
def test_column_names():
df = create_data()
results = sample._summarize_data(df)
assert df.columns.equals(results.index)
def test_sample_record():
df = create_data()
results = sample._summarize_data(df)
assert df.iloc[1].equals(results['sample_record'])
def test_summary_int32():
df = create_data()
results = sample._summarize_data(df)
df_result = 'unique values: ' + str(df['D'].unique().size)
test_result = results['summary'][3]
assert df_result == test_result
def test_summary_float32():
df = create_data()
results = sample._summarize_data(df)
df_result = 'median value: ' + str(np.median(df['C']))
test_result = results['summary'][2]
assert df_result == test_result