def test_with_original_granularity_multiple_aggregations(sample_df):
step = AggregateStep(
name='aggregate',
keepOriginalGranularity=True,
on=['Group'],
aggregations=[
Aggregation(aggfunction='min',
columns=['Value1'],
newcolumns=['min_Value1']),
Aggregation(aggfunction='max',
columns=['Value2'],
newcolumns=['max_Value2']),
],
)
df_result = execute_aggregate(step, sample_df)
assert_dataframes_equals(
df_result,
DataFrame({
'Label':
['Label 1', 'Label 2', 'Label 3', 'Label 4', 'Label 5', 'Label 6'],
'Group': ['Group 1'] * 3 + ['Group 2'] * 3,
'Value1': [13, 7, 20, 1, 10, 5],
'min_Value1': [7] * 3 + [1] * 3,
'Value2': [10, 21, 4, 17, 12, 2],
'max_Value2': [21] * 3 + [17] * 3,
}),
)
def test_count_with_null():
df = DataFrame({
'Label':
['Label 1', 'Label 2', 'Label 3', 'Label 4', 'Label 5', 'Label 6'],
'Group': ['Group 1'] * 3 + [None] * 3, # type: ignore
'Value1': [13, 7, 20, 1, 10, 5],
})
step = AggregateStep(
name='aggregate',
on=['Group'],
aggregations=[
Aggregation(
aggfunction='count distinct including empty',
columns=['Group'],
newcolumns=['__VQB_COUNT'],
),
],
)
df_result = execute_aggregate(step, df)
assert_dataframes_equals(
df_result.sort_values(by=['Group']),
DataFrame({
'Group': ['Group 1', np.nan],
'__VQB_COUNT': [3, 3],
}).sort_values(by=['Group']),
)
def test_simple_aggregate(sample_df):
step = AggregateStep(
name='aggregate',
on=['Group'],
aggregations=[
Aggregation(
aggfunction='sum',
columns=['Value1', 'Value2'],
newcolumns=['Sum-Value1', 'Sum-Value2'],
),
Aggregation(aggfunction='avg',
columns=['Value1'],
newcolumns=['Avg-Value1']),
],
)
df_result = execute_aggregate(step, sample_df)
assert_dataframes_equals(
df_result.sort_values(by=['Group']),
DataFrame({
'Group': ['Group 1', 'Group 2'],
'Sum-Value1': [40, 16],
'Sum-Value2': [35, 31],
'Avg-Value1': [np.average([13, 7, 20]),
np.average([1, 10, 5])],
}).sort_values(by=['Group']),
)
def test_simple_aggregate_with_null():
df = DataFrame({
'Label':
['Label 1', 'Label 2', 'Label 3', 'Label 4', 'Label 5', 'Label 6'],
'Group': ['Group 1'] * 3 + [None] * 3, # type: ignore
'Value1': [13, 7, 20, 1, 10, 5],
})
step = AggregateStep(
name='aggregate',
on=['Group'],
aggregations=[
Aggregation(
aggfunction='sum',
columns=['Value1'],
newcolumns=['Sum-Value1'],
),
],
)
df_result = execute_aggregate(step, df)
assert_dataframes_equals(
df_result.sort_values(by=['Group']),
DataFrame({
'Group': ['Group 1', None],
'Sum-Value1': [40, 16],
}).sort_values(by=['Group']),
)
def test_duplicate_aggregation_columns():
df = DataFrame({
'Label': ['Label 1', 'Label 2', 'Label 3'],
'Group': ['Group 1'] * 3, # type: ignore
'Value1': [13, 7, 20],
})
with pytest.raises(DuplicateColumnError):
step = AggregateStep(
name='aggregate',
on=['Group'],
aggregations=[
Aggregation(
aggfunction='count distinct including empty',
columns=['Group', 'Group'],
newcolumns=['__VQB_COUNT'],
),
],
)
execute_aggregate(step, df)
def test_keep_original_granularity_empty_on(sample_df):
step = AggregateStep(
name='aggregate',
on=[],
keepOriginalGranularity=True,
aggregations=[
Aggregation(aggfunction='count',
columns=['Group'],
newcolumns=['__vqb_count__']),
],
)
df_result = execute_aggregate(step, sample_df)
assert_dataframes_equals(df_result, sample_df.assign(__vqb_count__=6))
def test_count_distinct(sample_df):
step = AggregateStep(
name='aggregate',
keepOriginalGranularity=False,
on=[],
aggregations=[
Aggregation(
aggfunction='count distinct',
columns=['Group'],
newcolumns=['Group_CD'],
)
],
)
df_result = execute_aggregate(step, sample_df)
assert_dataframes_equals(df_result, DataFrame({'Group_CD': [2]}))
def test_without_on(sample_df):
step = AggregateStep(
name='aggregate',
keepOriginalGranularity=False,
on=[],
aggregations=[
Aggregation(aggfunction='sum',
columns=['Value1'],
newcolumns=['sum_value']),
],
)
df_result = execute_aggregate(step, sample_df)
assert_dataframes_equals(
df_result,
DataFrame({'sum_value': [56]}),
)
def test_count(sample_df):
step = AggregateStep(
name='aggregate',
keepOriginalGranularity=False,
on=['Group'],
aggregations=[
Aggregation(aggfunction='count',
columns=['Label'],
newcolumns=['count']),
],
)
df_result = execute_aggregate(step, sample_df)
assert_dataframes_equals(
df_result, DataFrame({
'Group': ['Group 1', 'Group 2'],
'count': [3, 3]
}))
def test_legacy_syntax(sample_df):
step = AggregateStep(
name='aggregate',
keepOriginalGranularity=False,
on=[],
aggregations=[
Aggregation(**{
'aggfunction': 'sum',
'column': 'Value1',
'newcolumn': 'sum_value'
}),
],
)
df_result = execute_aggregate(step, sample_df)
assert_dataframes_equals(
df_result,
DataFrame({'sum_value': [56]}),
)