def testGroup(self):
# Check the group class on its own
self.assertRaises(ValueError, a.Group, [], [])
self.assertRaises(ValueError, a.Group, ['foo'], [])
# Zero fields, single reducer
g = a.Group([], r.count())
ret = g.build_args()
self.assertEqual(['GROUPBY', '0', 'REDUCE', 'COUNT', '0'], ret)
# Single field, single reducer
g = a.Group('foo', r.count())
ret = g.build_args()
self.assertEqual(['GROUPBY', '1', 'foo', 'REDUCE', 'COUNT', '0'], ret)
# Multiple fields, single reducer
g = a.Group(['foo', 'bar'], r.count())
self.assertEqual(
['GROUPBY', '2', 'foo', 'bar', 'REDUCE', 'COUNT', '0'],
g.build_args())
# Multiple fields, multiple reducers
g = a.Group(['foo', 'bar'], [r.count(), r.count_distinct('@fld1')])
self.assertEqual([
'GROUPBY', '2', 'foo', 'bar', 'REDUCE', 'COUNT', '0', 'REDUCE',
'COUNT_DISTINCT', '1', '@fld1'
], g.build_args())
def test_reducers(self):
self.assertEqual((), r.count().args)
self.assertEqual(('f1', ), r.sum('f1').args)
self.assertEqual(('f1', ), r.min('f1').args)
self.assertEqual(('f1', ), r.max('f1').args)
self.assertEqual(('f1', ), r.avg('f1').args)
self.assertEqual(('f1', ), r.tolist('f1').args)
self.assertEqual(('f1', ), r.count_distinct('f1').args)
self.assertEqual(('f1', ), r.count_distinctish('f1').args)
self.assertEqual(('f1', '0.95'), r.quantile('f1', 0.95).args)
self.assertEqual(('f1', ), r.stddev('f1').args)
self.assertEqual(('f1', ), r.first_value('f1').args)
self.assertEqual(('f1', 'BY', 'f2', 'ASC'),
r.first_value('f1', a.Asc('f2')).args)
self.assertEqual(('f1', 'BY', 'f1', 'ASC'),
r.first_value('f1', a.Asc).args)
self.assertEqual(('f1', '50'), r.random_sample('f1', 50).args)
def testAggRequest(self):
req = a.AggregateRequest()
self.assertEqual(['*'], req.build_args())
# Test with group_by
req = a.AggregateRequest().group_by('@foo', r.count())
self.assertEqual(['*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0'],
req.build_args())
# Test with limit
req = a.AggregateRequest().\
group_by('@foo', r.count()).\
sort_by('@foo')
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '1',
'@foo'
], req.build_args())
# Test with sort_by
req = a.AggregateRequest().group_by('@foo', r.count()).sort_by('@date')
# print req.build_args()
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '1',
'@date'
], req.build_args())
req = a.AggregateRequest().group_by('@foo',
r.count()).sort_by(a.Desc('@date'))
# print req.build_args()
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '2',
'@date', 'DESC'
], req.build_args())
req = a.AggregateRequest().group_by('@foo', r.count()).sort_by(
a.Desc('@date'), a.Asc('@time'))
# print req.build_args()
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '4',
'@date', 'DESC', '@time', 'ASC'
], req.build_args())
req = a.AggregateRequest().group_by('@foo',
r.count()).sort_by(a.Desc('@date'),
a.Asc('@time'),
max=10)
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '4',
'@date', 'DESC', '@time', 'ASC', 'MAX', '10'
], req.build_args())
def get_movie_group_by(field):
req = aggregations.AggregateRequest("*").group_by(
"@" + field,
reducers.count().alias("nb_of_movies")).sort_by(
aggregations.Asc("@" + field)).limit(0, 1000)
res = g.movieIdx.aggregate(req)
reslist = []
for row in res.rows:
item = {
row[0].decode("utf-8"): row[1].decode("utf-8"),
row[2].decode("utf-8"): int(row[3].decode("utf-8"))
}
reslist.append(item)
dictResult = {
"totalResults": len(res.rows),
"rows": reslist,
}
return dictResult
def testAggregations(self):
conn = self.redis()
with conn as r:
client = Client('myIndex', port=conn.port)
client.redis.flushdb()
# Creating the index definition and schema
client.create_index(
(NumericField('random_num'), TextField('title'),
TextField('body'), TextField('parent')))
# Indexing a document
client.add_document(
'search',
title='RediSearch',
body='Redisearch impements a search engine on top of redis',
parent='redis',
random_num=10)
client.add_document(
'ai',
title='RedisAI',
body=
'RedisAI executes Deep Learning/Machine Learning models and managing their data.',
parent='redis',
random_num=3)
client.add_document(
'json',
title='RedisJson',
body=
'RedisJSON implements ECMA-404 The JSON Data Interchange Standard as a native data type.',
parent='redis',
random_num=8)
req = aggregations.AggregateRequest('redis').group_by(
"@parent",
reducers.count(),
reducers.count_distinct('@title'),
reducers.count_distinctish('@title'),
reducers.sum("@random_num"),
reducers.min("@random_num"),
reducers.max("@random_num"),
reducers.avg("@random_num"),
reducers.stddev("random_num"),
reducers.quantile("@random_num", 0.5),
reducers.tolist("@title"),
reducers.first_value("@title"),
reducers.random_sample("@title", 2),
)
res = client.aggregate(req)
res = res.rows[0]
self.assertEqual(len(res), 26)
self.assertEqual(b'redis', res[1])
self.assertEqual(b'3', res[3])
self.assertEqual(b'3', res[5])
self.assertEqual(b'3', res[7])
self.assertEqual(b'21', res[9])
self.assertEqual(b'3', res[11])
self.assertEqual(b'10', res[13])
self.assertEqual(b'7', res[15])
self.assertEqual(b'3.60555127546', res[17])
self.assertEqual(b'10', res[19])
self.assertEqual([b'RediSearch', b'RedisAI', b'RedisJson'],
res[21])
self.assertEqual(b'RediSearch', res[23])
self.assertEqual(2, len(res[25]))
def agg_by(field):
ar = aggregation.AggregateRequest().group_by(field, reducers.count().alias('my_count')).sort_by(aggregation.Desc('@my_count'))
return (client.aggregate(ar).rows)
def testAggRequest(self):
req = a.AggregateRequest()
self.assertEqual(['*'], req.build_args())
# Test with group_by
req = a.AggregateRequest().group_by('@foo', r.count())
self.assertEqual(['*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0'],
req.build_args())
# Test with group_by and alias on reducer
req = a.AggregateRequest().group_by('@foo',
r.count().alias('foo_count'))
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'AS',
'foo_count'
], req.build_args())
# Test with limit
req = a.AggregateRequest(). \
group_by('@foo', r.count()). \
sort_by('@foo')
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '1',
'@foo'
], req.build_args())
# Test with apply
req = a.AggregateRequest(). \
apply(foo="@bar / 2"). \
group_by('@foo', r.count())
self.assertEqual([
'*', 'APPLY', '@bar / 2', 'AS', 'foo', 'GROUPBY', '1', '@foo',
'REDUCE', 'COUNT', '0'
], req.build_args())
# Test with filter
req = a.AggregateRequest().group_by('@foo',
r.count()).filter("@foo=='bar'")
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'FILTER',
"@foo=='bar'"
], req.build_args())
# Test with filter on different state of the pipeline
req = a.AggregateRequest().filter("@foo=='bar'").group_by(
'@foo', r.count())
self.assertEqual([
'*', 'FILTER', "@foo=='bar'", 'GROUPBY', '1', '@foo', 'REDUCE',
'COUNT', '0'
], req.build_args())
# Test with filter on different state of the pipeline
req = a.AggregateRequest().filter(["@foo=='bar'", "@foo2=='bar2'"
]).group_by('@foo', r.count())
self.assertEqual([
'*', 'FILTER', "@foo=='bar'", 'FILTER', "@foo2=='bar2'", 'GROUPBY',
'1', '@foo', 'REDUCE', 'COUNT', '0'
], req.build_args())
# Test with sort_by
req = a.AggregateRequest().group_by('@foo', r.count()).sort_by('@date')
# print req.build_args()
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '1',
'@date'
], req.build_args())
req = a.AggregateRequest().group_by('@foo',
r.count()).sort_by(a.Desc('@date'))
# print req.build_args()
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '2',
'@date', 'DESC'
], req.build_args())
req = a.AggregateRequest().group_by('@foo', r.count()).sort_by(
a.Desc('@date'), a.Asc('@time'))
# print req.build_args()
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '4',
'@date', 'DESC', '@time', 'ASC'
], req.build_args())
req = a.AggregateRequest().group_by('@foo',
r.count()).sort_by(a.Desc('@date'),
a.Asc('@time'),
max=10)
self.assertEqual([
'*', 'GROUPBY', '1', '@foo', 'REDUCE', 'COUNT', '0', 'SORTBY', '4',
'@date', 'DESC', '@time', 'ASC', 'MAX', '10'
], req.build_args())
