def _apply_agg(self, search_query):
filters = self._get_agg_filters(
search_query.get_context().iter_post_filters_with_meta(),
{self.qf._name, self.name}
)
aggs = {}
if self._compute_enabled:
aggs.update({
self._enabled_agg_name: agg.Filter(self.field != None), # noqa:E711
})
if self._compute_min_max:
stat_aggs = {
self._min_agg_name: agg.Min(self.field),
self._max_agg_name: agg.Max(self.field),
}
if filters:
aggs.update({
self._filter_agg_name: agg.Filter(
Bool.must(*filters), aggs=stat_aggs
)
})
else:
aggs.update(stat_aggs)
return search_query.aggregations(**aggs)
def _extract_orders(self, search_query):
order_aggs = {}
order_by = []
if search_query._order_by:
for i, o in enumerate(search_query._order_by):
order_name = self._order_agg_name(i)
if isinstance(o, Sort):
expr = o.expr
order = o.order
else:
expr = o
order = None
if isinstance(expr, Field):
field_name = expr.get_name()
elif isinstance(expr, AttributedField):
field_name = expr.get_field_name()
else:
field_name = expr
if field_name == '_score':
# default ordering for _score is desc
desc = order is None or order == 'desc'
else:
desc = order == 'desc'
if desc:
if field_name == '_score':
order_aggs[order_name] = agg.Max(script='_score')
else:
order_aggs[order_name] = agg.Max(expr)
order_by.append({order_name: 'desc'})
else:
if field_name == '_score':
order_aggs[order_name] = agg.Min(script='_score')
else:
order_aggs[order_name] = agg.Min(expr)
order_by.append({order_name: 'asc'})
else:
order_aggs = {
self._order_agg_name(0): agg.Max(script='_score')
}
order_by = [{self._order_agg_name(0): 'desc'}]
return order_aggs, order_by
def _apply_agg(self, search_query):
filters = self._get_agg_filters(
search_query.get_context().iter_post_filters_with_meta(), {self.qf._name, self.name}
)
aggs = {}
if self._compute_enabled:
aggs.update({
self._enabled_agg_name: agg.Nested(
path=self.path,
aggs={
self._filter_key_agg_name: agg.Filter(
self.key_expression,
aggs={
self._filter_value_agg_name: agg.Filter(
self.value_field != None
)
}
)
}
)
})
if self._compute_min_max:
stat_aggs = {
self._enabled_agg_name_stat: agg.Nested(
path=self.path,
aggs={
self._filter_key_agg_name: agg.Filter(
self.key_expression,
aggs={
self._min_agg_name: agg.Min(self.value_field),
self._max_agg_name: agg.Max(self.value_field),
}
)
}
)
}
if filters:
aggs.update({
self._filter_agg_name: agg.Filter(Bool.must(*filters), aggs=stat_aggs)
})
else:
aggs.update(stat_aggs)
return search_query.aggregations(**aggs)
def test_aggregations(self):
f = DynamicDocument.fields
sq = SearchQuery().aggregations(min_price=agg.Min(f.price))
self.assert_expression(
sq,
{
"aggregations": {
"min_price": {
"min": {"field": "price"}
}
}
}
)
self.assertEqual(collect_doc_classes(sq), {DynamicDocument})
sq = SearchQuery().aggregations(genders=agg.Terms(f.gender))
self.assert_expression(
sq,
{
"aggregations": {
"genders": {
"terms": {"field": "gender"}
}
}
}
)
self.assertEqual(collect_doc_classes(sq), {DynamicDocument})
sq = (
SearchQuery()
.aggregations(
type=agg.Terms(f.type, aggs={'min_price': agg.Min(f.price)})
)
)
self.assert_expression(
sq,
{
"aggregations": {
"type": {
"terms": {"field": "type"},
"aggregations": {
"min_price": {
"min": {"field": "price"}
}
}
}
}
},
)
self.assertEqual(collect_doc_classes(sq), {DynamicDocument})
sq = (
SearchQuery()
.aggregations(
top_tags=(
agg.Terms(
f.tags,
size=3,
aggs={
'top_tag_hits': agg.TopHits(
sort=f.last_activity_date.desc(),
size=1,
_source=Params(include=[f.title]))
}
)
)
)
)
self.assert_expression(
sq,
{
"aggregations": {
"top_tags": {
"terms": {
"field": "tags",
"size": 3
},
"aggregations": {
"top_tag_hits": {
"top_hits": {
"sort": {
"last_activity_date": "desc"
},
"_source": {
"include": ["title"]
},
"size" : 1
}
}
}
}
}
}
)
self.assertEqual(collect_doc_classes(sq), {DynamicDocument})
sq = (
SearchQuery()
.aggregations({
'top_sites': agg.Terms(
f.domain,
order=Sort('top_hit', 'desc'),
aggs={
'top_tags_hits': agg.TopHits(),
'top_hit': agg.Max(script='_doc.score'),
}
)
})
)
self.assert_expression(
sq,
{
"aggregations": {
"top_sites": {
"terms": {
"field": "domain",
"order": {
"top_hit": "desc"
}
},
"aggregations": {
"top_tags_hits": {
"top_hits": {}
},
"top_hit" : {
"max": {
"script": "_doc.score"
}
}
}
}
}
}
)
self.assertEqual(collect_doc_classes(sq), {DynamicDocument})
def test_aggs(self):
f = DynamicDocument.fields
a = agg.AggExpression()
self.assertRaises(NotImplementedError, a.build_agg_result, {})
a = agg.Avg(f.price)
self.assert_expression(
a,
{
"avg": {"field": "price"}
}
)
res = a.build_agg_result({
'value': 75.3
})
self.assertAlmostEqual(res.value, 75.3)
res = a.build_agg_result({
'value': None
})
self.assertIs(res.value, None)
aa = a.clone()
self.assertIsNot(a, aa)
self.assertEqual(a.__visit_name__, aa.__visit_name__)
self.assertEqual(a.params, aa.params)
a = agg.Min(f.price)
self.assert_expression(
a,
{
"min": {"field": "price"}
}
)
res = a.build_agg_result({
'value': 38
})
self.assertAlmostEqual(res.value, 38)
res = a.build_agg_result({
'value': 1297167619690,
'value_as_string': '2011-02-08T12:20:19.690Z'
})
self.assertAlmostEqual(res.value, 1297167619690)
a = agg.Max(f.price)
self.assert_expression(
a,
{
"max": {"field": "price"}
}
)
res = a.build_agg_result({
'value': 45693.5
})
self.assertAlmostEqual(res.value, 45693.5)
a = agg.Stats(f.grade)
self.assert_expression(
a,
{
"stats": {"field": "grade"}
}
)
a = a.build_agg_result(
{
"count": 6,
"min": 60,
"max": 98,
"avg": 78.5,
"sum": 471
}
)
self.assertEqual(a.count, 6)
self.assertEqual(a.min, 60)
self.assertEqual(a.max, 98)
self.assertAlmostEqual(a.avg, 78.5)
self.assertEqual(a.sum, 471)
a = agg.ExtendedStats(f.grade)
self.assert_expression(
a,
{
"extended_stats": {"field": "grade"}
}
)
a = a.build_agg_result(
{
"count": 6,
"min": 72,
"max": 117.6,
"avg": 94.2,
"sum": 565.2,
"sum_of_squares": 54551.51999999999,
"variance": 218.2799999999976,
"std_deviation": 14.774302013969987
}
)
self.assertEqual(a.count, 6)
self.assertEqual(a.min, 72)
self.assertAlmostEqual(a.max, 117.6)
self.assertAlmostEqual(a.avg, 94.2)
self.assertAlmostEqual(a.sum, 565.2)
self.assertAlmostEqual(a.sum_of_squares, 54551.51999999999)
self.assertAlmostEqual(a.variance, 218.2799999999976)
self.assertAlmostEqual(a.std_deviation, 14.774302013969987)
percentiles_agg = agg.Percentiles(f.load_time, percents=[95, 99, 99.9])
self.assert_expression(
percentiles_agg,
{
"percentiles": {
"field": "load_time",
"percents": [95, 99, 99.9]
}
}
)
a = percentiles_agg.build_agg_result(
{
"values": {
"95.0": 60,
"99.0": 150,
"99.9": 153,
}
}
)
self.assertEqual(
a.values,
[(95.0, 60), (99.0, 150), (99.9, 153)],
)
self.assertEqual(a.get_value(95), 60)
self.assertEqual(a.get_value(95.0), 60)
self.assertEqual(a.get_value(99), 150)
self.assertEqual(a.get_value(99.0), 150)
self.assertEqual(a.get_value(99.9), 153)
a = percentiles_agg.build_agg_result(
{
"values": {
"95.0": 60,
"95.0_as_string": "60",
"99.0": 150,
"99.0_as_string": "150",
"99.9": 153,
"99.9_as_string": "153",
}
}
)
self.assertEqual(
a.values,
[(95.0, 60), (99.0, 150), (99.9, 153)],
)
self.assertEqual(a.get_value(95), 60)
self.assertEqual(a.get_value(95.0), 60)
self.assertEqual(a.get_value(99), 150)
self.assertEqual(a.get_value(99.0), 150)
self.assertEqual(a.get_value(99.9), 153)
percentiles_agg = agg.Percentiles(f.load_time, percents=[50])
self.assert_expression(
percentiles_agg,
{
"percentiles": {
"field": "load_time",
"percents": [50]
}
}
)
a = percentiles_agg.build_agg_result(
{
"values": {
"50.0": "NaN",
}
}
)
self.assertEqual(
len(a.values),
1
)
self.assertAlmostEqual(
a.values[0][0],
50.0
)
self.assertTrue(
math.isnan(a.values[0][1])
)
self.assertTrue(math.isnan(a.get_value(50)))
self.assertTrue(math.isnan(a.get_value(50.0)))
ranks_agg = agg.PercentileRanks(f.load_time, values=[14.8, 30])
self.assert_expression(
ranks_agg,
{
"percentile_ranks": {
"field": "load_time",
"values": [14.8, 30.0]
}
}
)
a = ranks_agg.build_agg_result(
{
"values": {
"14.8": 12.32,
"30": 100,
}
}
)
self.assertEqual(
a.values,
[(14.8, 12.32), (30.0, 100)],
)
self.assertEqual(
a.values,
[(14.8, 12.32), (30.0, 100)],
)
self.assertAlmostEqual(a.get_percent(14.8), 12.32)
self.assertAlmostEqual(a.get_percent(13.7 + 1.1), 12.32)
self.assertAlmostEqual(a.get_percent(30), 100.0)
self.assertAlmostEqual(a.get_percent(30.0), 100.0)
a = ranks_agg.build_agg_result(
{
"values": {
"14.8": 12.32,
"14.8_as_string": "12.32",
"30": 100,
"30_as_string": "100",
}
}
)
self.assertEqual(
a.values,
[(14.8, 12.32), (30.0, 100)],
)
self.assertEqual(
a.values,
[(14.8, 12.32), (30.0, 100)],
)
self.assertAlmostEqual(a.get_percent(14.8), 12.32)
self.assertAlmostEqual(a.get_percent(13.7 + 1.1), 12.32)
self.assertAlmostEqual(a.get_percent(30), 100.0)
self.assertAlmostEqual(a.get_percent(30.0), 100.0)
a = agg.Cardinality(f.author, precision_threshold=100)
self.assert_expression(
a,
{
"cardinality": {
"field": "author",
"precision_threshold": 100
}
}
)
a = a.build_agg_result(
{
"value": 184
}
)
self.assertEqual(a.value, 184)
a = agg.Global()
self.assert_expression(a, {"global": {}})
a = a.build_agg_result(
{"doc_count": 185}
)
self.assertEqual(a.doc_count, 185)
a = agg.Filter(f.company == 1)
self.assert_expression(a, {"filter": {"term": {"company": 1}}})
a2 = a.clone()
self.assertIsNot(a, a2)
self.assert_expression(a2, {"filter": {"term": {"company": 1}}})
a = a.build_agg_result(
{"doc_count": 148}
)
self.assertEqual(a.doc_count, 148)
a = agg.Terms(f.status)
self.assert_expression(
a,
{
"terms": {"field": "status"}
}
)
a1 = a.clone()
self.assertIsNot(a, a1)
a = a.build_agg_result(
{
'buckets': [
{'doc_count': 7353499, 'key': 0},
{'doc_count': 2267139, 'key': 1},
{'doc_count': 1036951, 'key': 4},
{'doc_count': 438384, 'key': 2},
{'doc_count': 9594, 'key': 3},
{'doc_count': 46, 'key': 5}
]
}
)
self.assertEqual(len(a.buckets), 6)
self.assertEqual(list(iter(a)), a.buckets)
self.assertEqual(a.buckets[0].key, 0)
self.assertEqual(a.buckets[0].doc_count, 7353499)
self.assertEqual(repr(a.buckets[0]), '<Bucket key=0 doc_count=7353499>')
self.assertIs(a.buckets[0], a.get_bucket(0))
self.assertEqual(a.buckets[1].key, 1)
self.assertEqual(a.buckets[1].doc_count, 2267139)
self.assertIs(a.buckets[1], a.get_bucket(1))
self.assertEqual(repr(a.buckets[1]), '<Bucket key=1 doc_count=2267139>')
self.assertEqual(a.buckets[2].key, 4)
self.assertEqual(a.buckets[2].doc_count, 1036951)
self.assertIs(a.buckets[2], a.get_bucket(4))
self.assertEqual(repr(a.buckets[2]), '<Bucket key=4 doc_count=1036951>')
self.assertEqual(a.buckets[3].key, 2)
self.assertEqual(a.buckets[3].doc_count, 438384)
self.assertIs(a.buckets[3], a.get_bucket(2))
self.assertEqual(repr(a.buckets[3]), '<Bucket key=2 doc_count=438384>')
self.assertEqual(a.buckets[4].key, 3)
self.assertEqual(a.buckets[4].doc_count, 9594)
self.assertIs(a.buckets[4], a.get_bucket(3))
self.assertEqual(repr(a.buckets[4]), '<Bucket key=3 doc_count=9594>')
self.assertEqual(a.buckets[5].key, 5)
self.assertEqual(a.buckets[5].doc_count, 46)
self.assertIs(a.buckets[5], a.get_bucket(5))
self.assertEqual(repr(a.buckets[5]), '<Bucket key=5 doc_count=46>')
a = agg.Terms(f.is_visible, type=Boolean)
self.assert_expression(
a,
{
"terms": {"field": "is_visible"}
}
)
a = a.build_agg_result(
{
'buckets': [
{'doc_count': 7, 'key': 'T'},
{'doc_count': 2, 'key': 'F'},
]
}
)
self.assertEqual(len(a.buckets), 2)
self.assertEqual(a.buckets[0].key, True)
self.assertEqual(a.buckets[0].doc_count, 7)
self.assertIs(a.buckets[0], a.get_bucket(True))
self.assertEqual(a.buckets[1].key, False)
self.assertEqual(a.buckets[1].doc_count, 2)
self.assertIs(a.buckets[1], a.get_bucket(False))
a = agg.Terms(f.category, type=List(Integer))
self.assert_expression(
a,
{
"terms": {"field": "category"}
}
)
a = a.build_agg_result(
{
'buckets': [
{'doc_count': 792, 'key': 28},
{'doc_count': 185, 'key': 3},
]
}
)
self.assertEqual(len(a.buckets), 2)
self.assertEqual(a.buckets[0].key, 28)
self.assertEqual(a.buckets[0].doc_count, 792)
self.assertIs(a.buckets[0], a.get_bucket(28))
self.assertEqual(a.buckets[1].key, 3)
self.assertEqual(a.buckets[1].doc_count, 185)
self.assertIs(a.buckets[1], a.get_bucket(3))
class ProductDocument(Document):
is_visible = Field(Boolean)
a = agg.Terms(ProductDocument.is_visible)
self.assert_expression(
a,
{
"terms": {"field": "is_visible"}
}
)
a = a.build_agg_result(
{
'buckets': [
{'doc_count': 7, 'key': 'T'},
{'doc_count': 2, 'key': 'F'},
]
}
)
self.assertEqual(len(a.buckets), 2)
self.assertEqual(a.buckets[0].key, True)
self.assertEqual(a.buckets[0].doc_count, 7)
self.assertIs(a.buckets[0], a.get_bucket(True))
self.assertEqual(a.buckets[1].key, False)
self.assertEqual(a.buckets[1].doc_count, 2)
self.assertIs(a.buckets[1], a.get_bucket(False))
a = agg.SignificantTerms(f.crime_type)
self.assert_expression(
a,
{
"significant_terms": {"field": "crime_type"}
}
)
a = a.build_agg_result(
{
"doc_count": 47347,
"buckets" : [
{
"key": "Bicycle theft",
"doc_count": 3640,
"score": 0.371,
"bg_count": 66799,
},
{
"key": "Mobile phone theft",
"doc_count": 27617,
"score": 0.0599,
"bg_count": 53182,
}
]
}
)
self.assertEqual(len(a.buckets), 2)
self.assertEqual(a.buckets[0].key, 'Bicycle theft')
self.assertEqual(a.buckets[0].doc_count, 3640)
self.assertAlmostEqual(a.buckets[0].score, 0.371)
self.assertEqual(a.buckets[0].bg_count, 66799)
self.assertIs(a.buckets[0], a.get_bucket('Bicycle theft'))
self.assertEqual(a.buckets[1].key, 'Mobile phone theft')
self.assertEqual(a.buckets[1].doc_count, 27617)
self.assertAlmostEqual(a.buckets[1].score, 0.0599)
self.assertEqual(a.buckets[1].bg_count, 53182)
self.assertIs(a.buckets[1], a.get_bucket('Mobile phone theft'))
a = agg.Range(
f.price,
ranges=[{'to': 200}, {'from': 200, 'to': 1000}, {'from': 1000}],
type=Integer,
)
self.assert_expression(
a,
{
"range": {
"field": "price",
"ranges": [
{"to": 200},
{"from": 200, "to": 1000},
{"from": 1000}
]
}
}
)
a1 = a.clone()
self.assertIsNot(a1, a)
a = a.build_agg_result(
{
"buckets": [
{
"to": 200,
"doc_count": 12
},
{
"from": 200,
"to": 1000,
"doc_count": 197
},
{
"from": 1000,
"doc_count": 8
}
]
}
)
self.assertEqual(len(a.buckets), 3)
self.assertEqual(a.buckets[0].doc_count, 12)
self.assertEqual(a.buckets[1].doc_count, 197)
self.assertEqual(a.buckets[2].doc_count, 8)
a = agg.Filters([Term(f.body, 'error'), Term(f.body, 'warning')])
self.assert_expression(
a,
{
"filters": {
"filters": [
{"term": {"body": "error"}},
{"term": {"body": "warning"}}
]
}
}
)
a = a.build_agg_result(
{
"buckets": [
{
"doc_count" : 34
},
{
"doc_count" : 439
},
]
}
)
self.assertEqual(len(a.buckets), 2)
self.assertIs(a.buckets[0].key, None)
self.assertEqual(a.buckets[0].doc_count, 34)
self.assertIs(a.buckets[1].key, None)
self.assertEqual(a.buckets[1].doc_count, 439)
self.assertIs(a.get_bucket(None), None)
a = agg.Filters(Params(errors=Term(f.body, 'error'), warnings=Term(f.body, 'warning')))
self.assert_expression(
a,
{
"filters": {
"filters": {
"errors": {"term": {"body": "error"}},
"warnings": {"term": {"body": "warning"}}
}
}
}
)
a = a.build_agg_result(
{
"buckets": {
"errors": {
"doc_count" : 34
},
"warnings": {
"doc_count" : 439
},
}
}
)
self.assertEqual(len(a.buckets), 2)
self.assertIs(a.buckets[0].key, 'errors')
self.assertEqual(a.buckets[0].doc_count, 34)
self.assertIs(a.buckets[0], a.get_bucket('errors'))
self.assertIs(a.buckets[1].key, 'warnings')
self.assertEqual(a.buckets[1].doc_count, 439)
self.assertIs(a.buckets[1], a.get_bucket('warnings'))
a = agg.Nested(f.resellers, aggs={'min_price': agg.Min(f.resellers.price)})
self.assert_expression(
a,
{
"nested": {"path": "resellers"},
"aggregations": {
"min_price": {"min": {"field": "resellers.price"}}
}
}
)
a = a.build_agg_result(
{
"min_price": {
"value" : 350
}
}
)
self.assertEqual(a.get_aggregation('min_price').value, 350)
a = agg.Nested(
f.resellers,
aggs={"resellers": agg.Terms(
f.resellers.id,
aggs={"reverse": agg.ReverseNested(
f.resellers,
aggs={"reseller_volume": agg.Sum(f.price)}
)})})
self.assert_expression(
a,
{
"nested": {"path": "resellers"},
"aggregations": {
"resellers": {
"terms": {"field": "resellers.id"},
"aggregations": {
"reverse": {
"reverse_nested": {"path": "resellers"},
"aggregations": {
"reseller_volume": {
"sum": {"field": "price"}
}
}
}
}
}
}
}
)
a = a.build_agg_result(
{
"doc_count": 100,
"resellers": {
"buckets": [
{
"key": 1122,
"doc_count": 48,
"reverse": {
"reseller_volume": {"value": 500100}
}
},
{
"key": 2233,
"doc_count": 52,
"reverse": {
"reseller_volume": {"value": 100500}
}
}
]
}
}
)
self.assertEqual(
a
.get_aggregation("resellers")
.buckets[1]
.get_aggregation("reverse")
.get_aggregation("reseller_volume")
.value,
100500
)
a = agg.Sampler(shard_size=1000, aggs={'avg_price': agg.Avg(f.price)})
self.assert_expression(
a,
{
"sampler": {"shard_size": 1000},
"aggregations": {
"avg_price": {"avg": {"field": "price"}}
}
}
)
a = a.build_agg_result(
{
"doc_count": 1000,
"avg_price": {
"value" : 750
}
}
)
self.assertEqual(a.doc_count, 1000)
self.assertEqual(a.get_aggregation('avg_price').value, 750)
# complex aggregation with sub aggregations
a = agg.Global()
a = a.aggs({
'selling_type': agg.Terms(
f.selling_type,
aggs={
'price_avg': agg.Avg(f.price),
'price_min': agg.Min(f.price),
'price_max': agg.Max(f.price),
'price_hist': agg.Histogram(f.price, interval=50),
}
),
'price_avg': agg.Avg(f.price),
}
)
self.assert_expression(
a,
{
"global": {},
"aggregations": {
"selling_type": {
"terms": {"field": "selling_type"},
"aggregations": {
"price_avg": {"avg": {"field": "price"}},
"price_min": {"min": {"field": "price"}},
"price_max": {"max": {"field": "price"}},
"price_hist": {
"histogram": {
"field": "price",
"interval": 50,
'min_doc_count': 1
}
},
}
},
"price_avg": {"avg": {"field": "price"}}
}
}
)
a = a.build_agg_result(
{
'doc_count': 100,
'selling_type': {
'buckets': [
{
'key': 'retail',
'doc_count': 70,
'price_avg': {'value': 60.5},
'price_min': {'value': 1.1},
'price_max': {'value': 83.4},
'price_hist': {
'buckets': [
{'key': 50, 'doc_count': 60},
{'key': 100, 'doc_count': 7},
{'key': 150, 'doc_count': 3},
]
},
},
{
'key': 'wholesale',
'doc_count': 30,
'price_avg': {'value': 47.9},
'price_min': {'value': 20.1},
'price_max': {'value': 64.8},
'price_hist': {
'buckets': [
{'key': 0, 'doc_count': 17},
{'key': 50, 'doc_count': 5},
{'key': 100, 'doc_count': 6},
{'key': 150, 'doc_count': 2},
]
},
},
],
},
'price_avg': {'value': 56.3},
}
)
self.assertEqual(a.doc_count, 100)
type_agg = a.get_aggregation('selling_type')
self.assertEqual(len(type_agg.buckets), 2)
self.assertEqual(type_agg.buckets[0].key, 'retail')
self.assertEqual(type_agg.buckets[0].doc_count, 70)
self.assertIs(type_agg.buckets[0], type_agg.get_bucket('retail'))
self.assertAlmostEqual(type_agg.buckets[0].get_aggregation('price_avg').value, 60.5)
self.assertAlmostEqual(type_agg.buckets[0].get_aggregation('price_min').value, 1.1)
self.assertAlmostEqual(type_agg.buckets[0].get_aggregation('price_max').value, 83.4)
price_hist_agg = type_agg.buckets[0].get_aggregation('price_hist')
self.assertEqual(price_hist_agg.buckets[0].key, 50)
self.assertEqual(price_hist_agg.buckets[0].doc_count, 60)
self.assertIs(price_hist_agg.buckets[0], price_hist_agg.get_bucket(50))
self.assertEqual(price_hist_agg.buckets[1].key, 100)
self.assertEqual(price_hist_agg.buckets[1].doc_count, 7)
self.assertIs(price_hist_agg.buckets[1], price_hist_agg.get_bucket(100))
self.assertEqual(price_hist_agg.buckets[2].key, 150)
self.assertEqual(price_hist_agg.buckets[2].doc_count, 3)
self.assertIs(price_hist_agg.buckets[2], price_hist_agg.get_bucket(150))
self.assertEqual(len(price_hist_agg.buckets), 3)
self.assertEqual(type_agg.buckets[1].key, 'wholesale')
self.assertEqual(type_agg.buckets[1].doc_count, 30)
self.assertIs(type_agg.buckets[1], type_agg.get_bucket('wholesale'))
self.assertAlmostEqual(type_agg.buckets[1].get_aggregation('price_avg').value, 47.9)
self.assertAlmostEqual(type_agg.buckets[1].get_aggregation('price_min').value, 20.1)
self.assertAlmostEqual(type_agg.buckets[1].get_aggregation('price_max').value, 64.8)
price_hist_agg = type_agg.buckets[1].get_aggregation('price_hist')
self.assertEqual(len(price_hist_agg.buckets), 4)
self.assertEqual(price_hist_agg.buckets[0].key, 0)
self.assertEqual(price_hist_agg.buckets[0].doc_count, 17)
self.assertIs(price_hist_agg.buckets[0], price_hist_agg.get_bucket(0))
self.assertEqual(price_hist_agg.buckets[1].key, 50)
self.assertEqual(price_hist_agg.buckets[1].doc_count, 5)
self.assertIs(price_hist_agg.buckets[1], price_hist_agg.get_bucket(50))
self.assertEqual(price_hist_agg.buckets[2].key, 100)
self.assertEqual(price_hist_agg.buckets[2].doc_count, 6)
self.assertIs(price_hist_agg.buckets[2], price_hist_agg.get_bucket(100))
self.assertEqual(price_hist_agg.buckets[3].key, 150)
self.assertEqual(price_hist_agg.buckets[3].doc_count, 2)
self.assertIs(price_hist_agg.buckets[3], price_hist_agg.get_bucket(150))
self.assertEqual(a.get_aggregation('price_avg').value, 56.3)
class QuestionDocument(DynamicDocument):
pass
class PaperDocument(DynamicDocument):
pass
question_mapper = Mock(
return_value={
'602679': Mock(id=602679, type='question'),
'602678': Mock(id=602678, type='question'),
}
)
paper_mapper = Mock(return_value={'602672': Mock(id=602672, type='paper')})
top_hits_agg = agg.Terms(
f.tags,
size=3,
aggs={
'top_tags_hits': agg.TopHits(
size=1,
sort=f.last_activity_date.desc(),
_source={'include': f.title},
instance_mapper={
QuestionDocument: question_mapper, PaperDocument: paper_mapper
},
)
}
)
self.assert_expression(
top_hits_agg,
{
"terms": {
"field": "tags",
"size": 3
},
"aggregations": {
"top_tags_hits": {
"top_hits": {
"sort": {
"last_activity_date": "desc"
},
"_source": {
"include": "title"
},
"size" : 1
}
}
}
}
)
a = top_hits_agg.build_agg_result(
{
"buckets": [
{
"key": "windows-7",
"doc_count": 25365,
"top_tags_hits": {
"hits": {
"total": 25365,
"max_score": 1,
"hits": [
{
"_index": "stack",
"_type": "question",
"_id": "602679",
"_score": 1,
"_source": {
"title": "Windows port opening"
},
"sort": [
1370143231177
]
}
]
}
}
},
{
"key": "linux",
"doc_count": 18342,
"top_tags_hits": {
"hits": {
"total": 18342,
"max_score": 1,
"hits": [
{
"_index": "stack",
"_type": "paper",
"_id": "602672",
"_score": 1,
"_source": {
"title": "Ubuntu RFID Screensaver lock-unlock"
},
"sort": [
1370143379747
]
}
]
}
}
},
{
"key": "windows",
"doc_count": 18119,
"top_tags_hits": {
"hits": {
"total": 18119,
"max_score": 1,
"hits": [
{
"_index": "stack",
"_type": "question",
"_id": "602678",
"_score": 1,
"_source": {
"title": "If I change my computers date / time, what could be affected?"
},
"sort": [
1370142868283
]
}
]
}
}
}
]
},
doc_cls_map={'question': QuestionDocument, 'paper': PaperDocument},
mapper_registry={},
)
self.assertEqual(len(a.buckets), 3)
self.assertEqual(a.buckets[0].doc_count, 25365)
self.assertEqual(a.buckets[0].key, 'windows-7')
top_tags_agg = a.buckets[0].get_aggregation('top_tags_hits')
self.assertEqual(top_tags_agg.total, 25365)
self.assertEqual(top_tags_agg.max_score, 1)
self.assertEqual(len(top_tags_agg.hits), 1)
self.assertIsInstance(top_tags_agg.hits[0], QuestionDocument)
self.assertEqual(top_tags_agg.hits[0]._index, 'stack')
self.assertEqual(top_tags_agg.hits[0]._type, 'question')
self.assertEqual(top_tags_agg.hits[0]._score, 1)
self.assertEqual(top_tags_agg.hits[0]._id, '602679')
self.assertEqual(top_tags_agg.hits[0].title, 'Windows port opening')
self.assertEqual(top_tags_agg.hits[0].instance.id, 602679)
self.assertEqual(top_tags_agg.hits[0].instance.type, 'question')
self.assertEqual(a.buckets[1].doc_count, 18342)
self.assertEqual(a.buckets[1].key, 'linux')
top_tags_agg = a.buckets[1].get_aggregation('top_tags_hits')
self.assertEqual(top_tags_agg.total, 18342)
self.assertEqual(top_tags_agg.max_score, 1)
self.assertEqual(len(top_tags_agg.hits), 1)
self.assertIsInstance(top_tags_agg.hits[0], PaperDocument)
self.assertEqual(top_tags_agg.hits[0]._index, 'stack')
self.assertEqual(top_tags_agg.hits[0]._type, 'paper')
self.assertEqual(top_tags_agg.hits[0]._score, 1)
self.assertEqual(top_tags_agg.hits[0]._id, '602672')
self.assertEqual(top_tags_agg.hits[0].title, 'Ubuntu RFID Screensaver lock-unlock')
self.assertEqual(top_tags_agg.hits[0].instance.id, 602672)
self.assertEqual(top_tags_agg.hits[0].instance.type, 'paper')
self.assertEqual(a.buckets[2].doc_count, 18119)
self.assertEqual(a.buckets[2].key, 'windows')
top_tags_agg = a.buckets[2].get_aggregation('top_tags_hits')
self.assertEqual(top_tags_agg.total, 18119)
self.assertEqual(top_tags_agg.max_score, 1)
self.assertEqual(len(top_tags_agg.hits), 1)
self.assertIsInstance(top_tags_agg.hits[0], DynamicDocument)
self.assertEqual(top_tags_agg.hits[0]._index, 'stack')
self.assertEqual(top_tags_agg.hits[0]._type, 'question')
self.assertEqual(top_tags_agg.hits[0]._score, 1)
self.assertEqual(top_tags_agg.hits[0]._id, '602678')
self.assertEqual(top_tags_agg.hits[0].title, 'If I change my computers date / time, what could be affected?')
self.assertEqual(top_tags_agg.hits[0].instance.id, 602678)
self.assertEqual(top_tags_agg.hits[0].instance.type, 'question')
self.assertEqual(question_mapper.call_count, 1)
self.assertEqual(paper_mapper.call_count, 1)
