def test_total_sales_day_by_day_by_shop_and_by_payment(elasticsearch_sale):
# Total sales day by day, by shop and by payment type
# This type of query is not possible with FQuery
search = get_search()
agg = search.aggs.bucket(
'timestamp',
'date_histogram',
field='timestamp',
interval='1d',
)
agg.bucket(
'shop_id',
'terms',
field='shop_id',
).metric(
'total_sales',
'sum',
field='price',
)
agg.bucket(
'payment_type',
'terms',
field='payment_type',
).metric(
'total_sales',
'sum',
field='price',
)
write_output(search, 'total_sales_day_by_day_by_shop_and_by_payment')
def test_total_sales_by_shop(elasticsearch_sale):
# Total sales by shop
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
Sale.shop_id, ),
'total_sales_by_shop',
)
def test_avg_product_price_by_product_type(elasticsearch_sale):
# Average product price by product type
write_fquery_output(
FQuery(get_search()).values(avg_product_price=Avg(
Sale.product_price), ).group_by(Sale.product_type, ),
'avg_product_price_by_product_type',
)
def test_nb_sales_by_shop_limited_size(elasticsearch_sale):
# Nb sales by shop limited size
write_fquery_output(
FQuery(get_search(),
default_size=2).values(Count(Sale), ).group_by(Sale.shop_id, ),
'nb_sales_by_shop_limited_size',
)
def test_avg_part_price_by_part(elasticsearch_sale):
# Average part price by part
write_fquery_output(
FQuery(get_search()).values(avg_part_price=Avg(
Sale.part_price), ).group_by(Sale.part_id, ),
'avg_part_price_by_part',
)
def test_total_sales_by_payment_type(elasticsearch_sale):
# Total sales by payment type
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
Sale.payment_type, ),
'total_sales_by_payment_type',
)
def test_nb_sales_by_product_type_by_part_id_filter_product_type_1(
elasticsearch_sale):
# Number of sales, by product type, by part id, for product_type_1
a = A('filter', term={'products.product_type': 'product_type_1'})
a.bucket(
'parts',
'nested',
path='products.parts',
).bucket(
'part_id',
'terms',
field='products.parts.part_id',
).metric(
'reverse_nested_root',
'reverse_nested',
)
search = get_search()
search.aggs.bucket(
'products',
'nested',
path='products',
).bucket(
'product_type_1',
a,
)
write_output(search,
'nb_sales_by_product_type_by_part_id_filter_product_type_1')
def test_total_sales_by_shop_limited_size(elasticsearch_sale):
# Total sales by shop limited size
write_fquery_output(
FQuery(get_search(),
default_size=2).values(total_sales=Sum(Sale.price), ).group_by(
Sale.shop_id, ),
'total_sales_by_shop_limited_size',
)
def test_total_in_traffic_and_total_out_traffic(elasticsearch_traffic):
# Total in traffic and out traffic
write_fquery_output(
FQuery(get_search()).values(
Sum(TrafficCount.incoming_traffic),
Sum(TrafficCount.outgoing_traffic),
),
'total_in_traffic_and_total_out_traffic',
)
def test_nb_sales_by_product_type(elasticsearch_sale):
# Nb sales by product_type
write_fquery_output(
FQuery(get_search()).values(ReverseNested(
Sale,
Count(Sale),
), ).group_by(Sale.product_type, ),
'nb_sales_by_product_type',
)
def test_total_in_traffic_and_total_out_traffic_by_shop(elasticsearch_traffic):
# Total in traffic and out traffic by shop id
write_fquery_output(
FQuery(get_search()).values(
Sum(TrafficCount.incoming_traffic),
Sum(TrafficCount.outgoing_traffic),
).group_by(TrafficCount.shop_id, ),
'total_in_traffic_and_total_out_traffic_by_shop',
)
def test_total_sales_and_avg_sales(elasticsearch_sale):
# Total sales and avg sales, no aggregations
write_fquery_output(
FQuery(get_search()).values(
total_sales=Sum(Sale.price),
avg_sales=Avg(Sale.price),
),
'total_sales_and_avg_sales',
)
def test_avg_sales_by_grouped_shop(elasticsearch_sale, shops_by_group):
# Avg price by grouped shop id
write_fquery_output(
FQuery(get_search()).values(avg_sales=Avg(Sale.price), ).group_by(
GroupedField(
Sale.shop_id,
groups=shops_by_group,
), ),
'avg_sales_by_grouped_shop',
)
def test_total_sales_by_shop_range_by_payment_type(elasticsearch_sale):
# Total sales by shop range by payment_type
ranges = [[1, 5], [5, 11], [11, 15]]
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
FieldWithRanges(Sale.shop_id, ranges=ranges),
Sale.payment_type,
),
'total_sales_by_shop_range_by_payment_type',
)
def test_nb_sales_by_grouped_shop(elasticsearch_sale, shops_by_group):
# Nb sales by grouped shop id
write_fquery_output(
FQuery(get_search()).values(Count(Sale), ).group_by(
GroupedField(
Sale.shop_id,
groups=shops_by_group,
), ),
'nb_sales_by_grouped_shop',
)
def test_total_sales_every_four_days(elasticsearch_sale):
# Total sales every four days
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
DateHistogram(
Sale.timestamp,
interval='4d',
), ),
'total_sales_every_four_days',
)
def test_nb_sales_by_date_range_with_keys(elasticsearch_sale,
date_ranges_with_keys):
write_fquery_output(
FQuery(get_search()).values(Count(Sale), ).group_by(
DateRange(
Sale.timestamp,
ranges=date_ranges_with_keys,
), ),
'nb_sales_by_date_range_with_keys',
)
def test_total_sales_by_period(elasticsearch_sale, interval, pretty_period):
# Total sales period by period
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
DateHistogram(
Sale.timestamp,
interval=interval,
), ),
'total_sales_{}_by_{}'.format(pretty_period, pretty_period),
)
def test_total_sales_by_price_histogram(elasticsearch_sale):
# Total sales by price histogram
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
Histogram(
Sale.price,
interval=100,
), ),
'total_sales_by_price_histogram',
)
def test_total_sales_by_day_offset(elasticsearch_sale):
# Total sales by day, with offset
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
DateHistogram(
Sale.timestamp,
interval='1d',
offset='+8h',
), ),
'total_sales_by_day_offset_8hours',
)
def test_total_and_avg_sales_by_product_type(elasticsearch_sale):
# Average sale price by product type
write_fquery_output(
FQuery(get_search()).values(
ReverseNested(
Sale,
avg_sales=Avg(Sale.price),
total_sales=Sum(Sale.price),
), ).group_by(Sale.product_type, ),
'total_and_avg_sales_by_product_type',
)
def test_generate_performance_output(elasticsearch_sale):
write_fquery_output(
FQuery(get_search()).values(Count(Sale), ).group_by(
DateHistogram(
Sale.timestamp,
interval='15m',
min=start,
max=end,
)),
'nb_sales_by_15_minutes_performance',
)
def test_nb_sales_by_payment_type_by_date_range(elasticsearch_sale,
date_ranges_with_keys):
# Nb sales by payment type by date range
write_fquery_output(
FQuery(get_search()).values(Count(Sale), ).group_by(
Sale.payment_type,
DateRange(
Sale.timestamp,
ranges=date_ranges_with_keys,
),
),
'nb_sales_by_payment_type_by_date_range',
)
def test_avg_price_filter_shop_id_1(elasticsearch_sale):
# Avg price for shop_id 1
a = A('filter', term={'shop_id': 1})
a.bucket(
'avg_price',
'avg',
field='price',
)
search = get_search()
search.aggs.bucket(
'shop_id_1',
a,
)
write_output(search, 'avg_price_filter_shop_id_1')
def test_flatten_performance():
fquery = FQuery(get_search()).values(Count(Sale), ).group_by(
DateHistogram(
Sale.timestamp,
interval='15m',
min=start,
max=end,
), )
result = load_output('nb_sales_by_15_minutes_performance')
fquery._flatten_result(
result,
remove_nested_aggregations=fquery._contains_nested_expressions(),
)
def test_total_sales_by_shop_range(elasticsearch_sale):
# Total sales by shop range
ranges = [{
'from': 1,
'to': 5,
'key': '1 - 5',
}, {
'from': 5,
'key': '5+',
}]
write_fquery_output(
FQuery(get_search()).values(total_sales=Sum(Sale.price), ).group_by(
FieldWithRanges(Sale.shop_id, ranges=ranges), ),
'total_sales_by_shop_range',
)
def test_avg_part_price_by_shop_range_by_part_id(elasticsearch_sale):
# Average part price by shop range by part id
ranges = [{
'from': 1,
'to': 5,
'key': '1 - 5',
}, {
'from': 5,
'key': '5+',
}]
write_fquery_output(
FQuery(get_search()).values(avg_part_price=Avg(
Sale.part_price), ).group_by(
FieldWithRanges(Sale.shop_id, ranges=ranges),
Sale.part_id,
),
'avg_part_price_by_shop_range_by_part_id',
)
def test_nb_sales_by_date_range_without_keys(elasticsearch_sale):
# Nb sales by date range without keys
ranges_without_keys = [
{
'from': datetime(2016, 1, 1),
'to': datetime(2016, 1, 15),
},
{
'from': datetime(2016, 1, 15),
'to': datetime(2016, 1, 31),
},
]
write_fquery_output(
FQuery(get_search()).values(Count(Sale), ).group_by(
DateRange(
Sale.timestamp,
ranges=ranges_without_keys,
), ),
'nb_sales_by_date_range_without_keys',
)
def test_total_sales_by_shop_and_by_payment(elasticsearch_sale):
# Total sales by shop and by payment type
# This type of query is not possible with FQuery
search = get_search()
search.aggs.bucket(
'shop_id',
'terms',
field='shop_id',
).metric(
'total_sales',
'sum',
field='price',
)
search.aggs.bucket(
'payment_type',
'terms',
field='payment_type',
).metric(
'total_sales',
'sum',
field='price',
)
write_output(search, 'total_sales_by_shop_and_by_payment')
def test_avg_part_price_by_product_and_by_part(elasticsearch_sale):
# Average part price by product and by part
# This type of query is not possible with FQuery
search = get_search()
products_bucket = search.aggs.bucket(
'products',
'nested',
path='products',
)
products_bucket.bucket(
'product_id',
'terms',
field='products.product_id',
).bucket(
'parts',
'nested',
path='products.parts',
).metric(
'avg_part_price',
'avg',
field='products.parts.part_price',
)
products_bucket.bucket(
'parts',
'nested',
path='products.parts',
).bucket(
'part_id',
'terms',
field='products.parts.part_id',
).metric(
'avg_part_price',
'avg',
field='products.parts.part_price',
)
write_output(search, 'avg_part_price_by_product_and_by_part')
