def test_aggregations_sort_by_and_limit(client):
client.ft().create_index((
TextField("t1"),
TextField("t2"),
))
client.ft().client.hset("doc1", mapping={"t1": "a", "t2": "b"})
client.ft().client.hset("doc2", mapping={"t1": "b", "t2": "a"})
# test sort_by using SortDirection
req = aggregations.AggregateRequest("*").sort_by(aggregations.Asc("@t2"),
aggregations.Desc("@t1"))
res = client.ft().aggregate(req)
assert res.rows[0] == ["t2", "a", "t1", "b"]
assert res.rows[1] == ["t2", "b", "t1", "a"]
# test sort_by without SortDirection
req = aggregations.AggregateRequest("*").sort_by("@t1")
res = client.ft().aggregate(req)
assert res.rows[0] == ["t1", "a"]
assert res.rows[1] == ["t1", "b"]
# test sort_by with max
req = aggregations.AggregateRequest("*").sort_by("@t1", max=1)
res = client.ft().aggregate(req)
assert len(res.rows) == 1
# test limit
req = aggregations.AggregateRequest("*").sort_by("@t1").limit(1, 1)
res = client.ft().aggregate(req)
assert len(res.rows) == 1
assert res.rows[0] == ["t1", "b"]
def test_aggregations_apply(client):
client.ft().create_index((
TextField("PrimaryKey", sortable=True),
NumericField("CreatedDateTimeUTC", sortable=True),
))
client.ft().client.hset(
"doc1",
mapping={
"PrimaryKey": "9::362330",
"CreatedDateTimeUTC": "637387878524969984"
},
)
client.ft().client.hset(
"doc2",
mapping={
"PrimaryKey": "9::362329",
"CreatedDateTimeUTC": "637387875859270016"
},
)
req = aggregations.AggregateRequest("*").apply(
CreatedDateTimeUTC="@CreatedDateTimeUTC * 10")
res = client.ft().aggregate(req)
assert res.rows[0] == ["CreatedDateTimeUTC", "6373878785249699840"]
assert res.rows[1] == ["CreatedDateTimeUTC", "6373878758592700416"]
def test_aggregations_filter(client):
client.ft().create_index((
TextField("name", sortable=True),
NumericField("age", sortable=True),
))
client.ft().client.hset("doc1", mapping={"name": "bar", "age": "25"})
client.ft().client.hset("doc2", mapping={"name": "foo", "age": "19"})
req = aggregations.AggregateRequest("*").filter(
"@name=='foo' && @age < 20")
res = client.ft().aggregate(req)
assert len(res.rows) == 1
assert res.rows[0] == ["name", "foo", "age", "19"]
req = aggregations.AggregateRequest("*").filter("@age > 15").sort_by(
"@age")
res = client.ft().aggregate(req)
assert len(res.rows) == 2
assert res.rows[0] == ["age", "19"]
assert res.rows[1] == ["age", "25"]
def test_aggregations_load(client):
client.ft().create_index((
TextField("t1"),
TextField("t2"),
))
client.ft().client.hset("doc1", mapping={"t1": "hello", "t2": "world"})
# load t1
req = aggregations.AggregateRequest("*").load("t1")
res = client.ft().aggregate(req)
assert res.rows[0] == ["t1", "hello"]
# load t2
req = aggregations.AggregateRequest("*").load("t2")
res = client.ft().aggregate(req)
assert res.rows[0] == ["t2", "world"]
# load all
req = aggregations.AggregateRequest("*").load()
res = client.ft().aggregate(req)
assert res.rows[0] == ["t1", "hello", "t2", "world"]
def test_profile(client):
client.ft().create_index((TextField("t"), ))
client.ft().client.hset("1", "t", "hello")
client.ft().client.hset("2", "t", "world")
# check using Query
q = Query("hello|world").no_content()
res, det = client.ft().profile(q)
assert det["Iterators profile"]["Counter"] == 2.0
assert len(det["Iterators profile"]["Child iterators"]) == 2
assert det["Iterators profile"]["Type"] == "UNION"
assert det["Parsing time"] < 0.5
assert len(res.docs) == 2 # check also the search result
# check using AggregateRequest
req = (aggregations.AggregateRequest("*").load("t").apply(
prefix="startswith(@t, 'hel')"))
res, det = client.ft().profile(req)
assert det["Iterators profile"]["Counter"] == 2.0
assert det["Iterators profile"]["Type"] == "WILDCARD"
assert det["Parsing time"] < 0.5
assert len(res.rows) == 2 # check also the search result
def test_aggregations_groupby(client):
# Creating the index definition and schema
client.ft().create_index((
NumericField("random_num"),
TextField("title"),
TextField("body"),
TextField("parent"),
))
# Indexing a document
client.ft().add_document(
"search",
title="RediSearch",
body="Redisearch impements a search engine on top of redis",
parent="redis",
random_num=10,
)
client.ft().add_document(
"ai",
title="RedisAI",
body=
"RedisAI executes Deep Learning/Machine Learning models and managing their data.", # noqa
parent="redis",
random_num=3,
)
client.ft().add_document(
"json",
title="RedisJson",
body=
"RedisJSON implements ECMA-404 The JSON Data Interchange Standard as a native data type.", # noqa
parent="redis",
random_num=8,
)
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.count(),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "3"
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.count_distinct("@title"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "3"
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.count_distinctish("@title"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "3"
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.sum("@random_num"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "21" # 10+8+3
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.min("@random_num"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "3" # min(10,8,3)
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.max("@random_num"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "10" # max(10,8,3)
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.avg("@random_num"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "7" # (10+3+8)/3
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.stddev("random_num"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "3.60555127546"
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.quantile("@random_num", 0.5),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == "8" # median of 3,8,10
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.tolist("@title"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[3] == ["RediSearch", "RedisAI", "RedisJson"]
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.first_value("@title").alias("first"),
)
res = client.ft().aggregate(req).rows[0]
assert res == ["parent", "redis", "first", "RediSearch"]
req = aggregations.AggregateRequest("redis").group_by(
"@parent",
reducers.random_sample("@title", 2).alias("random"),
)
res = client.ft().aggregate(req).rows[0]
assert res[1] == "redis"
assert res[2] == "random"
assert len(res[3]) == 2
assert res[3][0] in ["RediSearch", "RedisAI", "RedisJson"]