def build_project_data():
years = {
'$project': {
'datetime': '$datetime',
'count': '$count',
'sum': '$sum',
'sum2': '$sum2',
'min': '$min',
'max': '$max',
}
}
months = {
'$project': {
'datetime': '$months.datetime',
'count': '$months.count',
'sum': '$months.sum',
'sum2': '$months.sum2',
'min': '$months.min',
'max': '$months.max',
}
}
days = {
'$project': {
'datetime': '$months.days.datetime',
'count': '$months.days.count',
'sum': '$months.days.sum',
'sum2': '$months.days.sum2',
'min': '$months.days.min',
'max': '$months.days.max',
}
}
hours = {
'$project': {
'datetime': '$months.days.hours.datetime',
'count': '$months.days.hours.count',
'sum': '$months.days.hours.sum',
'sum2': '$months.days.hours.sum2',
'min': '$months.days.hours.min',
'max': '$months.days.hours.max',
}
}
return [
(aggregateby.Aggregateby(0), years),
(aggregateby.Aggregateby(1), months),
(aggregateby.Aggregateby(2), days),
(aggregateby.Aggregateby(3), hours),
]
def test_can_parse_valid_str_aggregateby(
self,
int_aggregateby,
aggregation_keys,
):
parsed_aggregateby = aggregateby.Aggregateby(int_aggregateby)
self.assertEqual(parsed_aggregateby.aggregation_keys, aggregation_keys)
def test_aggregateby_freq(
self,
interval,
coef,
expected_freq,
):
parsed_aggregateby = aggregateby.Aggregateby(interval, coef=coef)
self.assertEqual(parsed_aggregateby.freq, expected_freq)
def build_unwind_and_match_data():
years_pipeline = []
months_pipeline = [{
'$unwind': '$months',
}, {
'$match': {
'months.datetime': {
'$gte': datetime(2001, 3, 1),
'$lte': datetime(2001, 4, 1),
}
}
}]
days_pipeline = [{
'$unwind': '$months',
}, {
'$match': {
'months.datetime': {
'$gte': datetime(2001, 3, 1),
'$lte': datetime(2001, 4, 1),
}
}
}, {
'$unwind': '$months.days',
}, {
'$match': {
'months.days.datetime': {
'$gte': datetime(2001, 3, 22),
'$lte': datetime(2001, 4, 2),
}
}
}]
hours_pipeline = [{
'$unwind': '$months',
}, {
'$match': {
'months.datetime': {
'$gte': datetime(2001, 3, 1),
'$lte': datetime(2001, 4, 1),
}
}
}, {
'$unwind': '$months.days',
}, {
'$match': {
'months.days.datetime': {
'$gte': datetime(2001, 3, 22),
'$lte': datetime(2001, 4, 2),
}
}
}, {
'$unwind': '$months.days.hours',
}, {
'$match': {
'months.days.hours.datetime': {
'$gte': datetime(2001, 3, 22, 12),
'$lte': datetime(2001, 4, 2, 0),
}
}
}]
start = datetime(2001, 3, 22, 12)
end = datetime(2001, 4, 2)
return [
(start, end, aggregateby.Aggregateby(0), years_pipeline),
(start, end, aggregateby.Aggregateby(1), months_pipeline),
(start, end, aggregateby.Aggregateby(2), days_pipeline),
(start, end, aggregateby.Aggregateby(3), hours_pipeline),
]
def raw_data_groupbys_and_expected_outputs():
return [
(
# raw data
[],
# aggregateby
aggregateby.Aggregateby(3),
# groupby
[],
# expected df index
pd.Index([], name='datetime'),
# expected df data
[],
),
(
# raw data
[{
'datetime': datetime(1987, 5, 8),
'count': 0,
'sum': 0,
'sum2': 0,
'min': pd.np.inf,
'max': -pd.np.inf,
}],
# aggregateby
aggregateby.Aggregateby(2),
# groupby
[],
# expected df index
pd.Index([datetime(1987, 5, 8)], name='datetime'),
# expected df data
[
[0, pd.np.inf, -pd.np.inf, pd.np.nan, pd.np.nan],
],
),
(
# raw data
[{
'datetime': datetime(1987, 5, 8),
'count': 5,
'sum': 8.2,
'sum2': 14.32,
'min': 1.1,
'max': 2.3,
}],
# aggregateby
aggregateby.Aggregateby(2),
# groupby
[],
# expected df index
pd.Index([datetime(1987, 5, 8)], name='datetime'),
# expected df data
[
[5, 1.1, 2.3, 1.64, 0.417612],
],
),
(
# raw data
[{
'datetime': datetime(1987, 5, 8),
'count': 3,
'sum': 4.1,
'sum2': 5.79,
'min': 1.1,
'max': 1.7,
}, {
'datetime': datetime(1987, 5, 8),
'count': 2,
'sum': 4.1,
'sum2': 8.53,
'min': 1.8,
'max': 2.3,
}],
# aggregateby
aggregateby.Aggregateby(2),
# groupby
[],
# expected df index
pd.Index([datetime(1987, 5, 8)], name='datetime'),
# expected df data
[
[5, 1.1, 2.3, 1.64, 0.417612],
],
),
(
# raw data
[{
'datetime': datetime(1987, 5, 8),
'count': 3,
'sum': 4.1,
'sum2': 5.79,
'min': 1.1,
'max': 1.7,
'plop': 'lol',
}, {
'datetime': datetime(1987, 5, 8),
'count': 2,
'sum': 4.1,
'sum2': 8.53,
'min': 1.8,
'max': 2.3,
'plop': 'mdr',
}],
# aggregateby
aggregateby.Aggregateby(2),
# groupby
['plop'],
# expected df index
pd.MultiIndex.from_product([[
datetime(1987, 5, 8),
], [
'lol',
'mdr',
]],
names=['datetime', 'plop']),
# expected df data
[
[3, 1.1, 1.7, 1.366666, 0.249443],
[2, 1.8, 2.3, 2.05, 0.25],
],
),
(
# raw data
[{
'datetime': datetime(1987, 5, 8),
'count': 3,
'sum': 21.6 * 3,
'sum2': 21.6**2 * 3,
'min': 21.6,
'max': 21.6,
}],
# aggregateby
aggregateby.Aggregateby(2),
# groupby
[],
# expected df index
pd.Index([datetime(1987, 5, 8)], name='datetime'),
# expected df data
[
[3, 21.6, 21.6, 21.6, 0.0],
],
),
(
# raw data
[{
'datetime': datetime(1987, 5, 8),
'count': 3,
'sum': 4.1,
'sum2': 5.79,
'min': 1.1,
'max': 1.7,
}, {
'datetime': datetime(1987, 5, 9),
'count': 2,
'sum': 4.1,
'sum2': 8.53,
'min': 1.8,
'max': 2.3,
}, {
'datetime': datetime(1987, 5, 11),
'count': 1,
'sum': 2.3,
'sum2': 2.3**2,
'min': 2.3,
'max': 2.3,
}],
# aggregateby
aggregateby.Aggregateby(2, coef=2),
# groupby
[],
# expected df index
pd.Index([
datetime(1987, 5, 8),
datetime(1987, 5, 10),
],
name='datetime'),
# expected df data
[
[5, 1.1, 2.3, 1.64, 0.417612],
[1, 2.3, 2.3, 2.3, 0.0],
],
),
(
# raw data
[{
'datetime': datetime(1987, 5, 8),
'count': 3,
'sum': 4.1,
'sum2': 5.79,
'min': 1.1,
'max': 1.7,
'plop': 'A',
}, {
'datetime': datetime(1987, 5, 9),
'count': 2,
'sum': 4.1,
'sum2': 8.53,
'min': 1.8,
'max': 2.3,
'plop': 'A',
}, {
'datetime': datetime(1987, 5, 11),
'count': 1,
'sum': 2.3,
'sum2': 2.3**2,
'min': 2.3,
'max': 2.3,
'plop': 'B',
}],
# aggregateby
aggregateby.Aggregateby(2, coef=2),
# groupby
['plop'],
# expected df index
pd.MultiIndex(
levels=[[
datetime(1987, 5, 8),
datetime(1987, 5, 10),
], [
'A',
'B',
]],
labels=[[0, 1], [0, 1]],
names=['datetime', 'plop'],
),
# expected df data
[
[5, 1.1, 2.3, 1.64, 0.417612],
[1, 2.3, 2.3, 2.3, 0.0],
],
)
]