-
Notifications
You must be signed in to change notification settings - Fork 2.3k
/
ExpressionSelectivityCalculatorTest.scala
157 lines (126 loc) · 7.51 KB
/
ExpressionSelectivityCalculatorTest.scala
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
/*
* Copyright (c) 2002-2016 "Neo Technology,"
* Network Engine for Objects in Lund AB [http://neotechnology.com]
*
* This file is part of Neo4j.
*
* Neo4j is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
package org.neo4j.cypher.internal.compiler.v3_0.planner.logical.cardinality
import org.mockito.Mockito.when
import org.neo4j.cypher.internal.compiler.v3_0.planner.logical.plans.IdName
import org.neo4j.cypher.internal.compiler.v3_0.planner.logical.{Cardinality, Selectivity}
import org.neo4j.cypher.internal.compiler.v3_0.planner.{Predicate, Selections}
import org.neo4j.cypher.internal.compiler.v3_0.spi.GraphStatistics
import org.neo4j.cypher.internal.frontend.v3_0.ast._
import org.neo4j.cypher.internal.frontend.v3_0.helpers.NonEmptyList
import org.neo4j.cypher.internal.frontend.v3_0.symbols._
import org.neo4j.cypher.internal.frontend.v3_0.test_helpers.CypherFunSuite
import org.neo4j.cypher.internal.frontend.v3_0.{InputPosition, LabelId, PropertyKeyId, SemanticTable}
class ExpressionSelectivityCalculatorTest extends CypherFunSuite with AstConstructionTestSupport {
test("Should consider parameter expressions when calculating index selectivity") {
implicit val semanticTable = SemanticTable()
semanticTable.resolvedLabelIds.put("Page", LabelId(0))
semanticTable.resolvedPropertyKeyNames.put("title", PropertyKeyId(0))
implicit val selections = Selections(Set(Predicate(Set(IdName("n")), HasLabels(varFor("n"), Seq(LabelName("Page")_))_)))
val stats = mock[GraphStatistics]
when(stats.nodesWithLabelCardinality(None)).thenReturn(1000.0)
when(stats.indexSelectivity(LabelId(0), PropertyKeyId(0))).thenReturn(Some(Selectivity.of(0.1d).get))
val calculator = ExpressionSelectivityCalculator(stats, IndependenceCombiner)
val result = calculator(In(Property(varFor("n"), PropertyKeyName("title")_)_, Parameter("titles", CTAny)_)_)
result.factor should equal (0.92 +- 0.01)
}
test("Should peek inside sub predicates") {
implicit val semanticTable = SemanticTable()
semanticTable.resolvedLabelIds.put("Page", LabelId(0))
implicit val selections = Selections(Set(Predicate(Set(IdName("n")), HasLabels(varFor("n"), Seq(LabelName("Page")_))_)))
val stats = mock[GraphStatistics]
when(stats.nodesWithLabelCardinality(None)).thenReturn(2000.0)
when(stats.nodesWithLabelCardinality(Some(LabelId(0)))).thenReturn(1000.0)
val calculator = ExpressionSelectivityCalculator(stats, IndependenceCombiner)
val result = calculator(PartialPredicate[HasLabels](HasLabels(varFor("n"), Seq(LabelName("Page")_))_, mock[HasLabels]))
result.factor should equal(0.5)
}
test("Should look at range predicates that could benefit from using an index") {
implicit val semanticTable = SemanticTable()
semanticTable.resolvedLabelIds.put("Person", LabelId(0))
val n_is_Person = Predicate(Set(IdName("n")), HasLabels(varFor("n"), Seq(LabelName("Person") _)) _)
val n_prop: Property = Property(varFor("n"), PropertyKeyName("prop")_)_
val n_gt_3_and_lt_4 = Predicate(Set(IdName("n")), AndedPropertyInequalities(varFor("n"), n_prop, NonEmptyList(
GreaterThan(n_prop, SignedDecimalIntegerLiteral("3")_)_,
LessThan(n_prop, SignedDecimalIntegerLiteral("4")_)_
)))
implicit val selections = Selections(Set(n_is_Person, n_gt_3_and_lt_4))
val stats = mock[GraphStatistics]
when(stats.nodesWithLabelCardinality(None)).thenReturn(2000.0)
when(stats.nodesWithLabelCardinality(Some(LabelId(0)))).thenReturn(1000.0)
val calculator = ExpressionSelectivityCalculator(stats, IndependenceCombiner)
val result = calculator(n_gt_3_and_lt_4.expr)
result.factor should equal(0.06)
}
test("Should optimize selectivity with respect to prefix length for STARTS WITH predicates") {
implicit val semanticTable = SemanticTable()
semanticTable.resolvedLabelIds.put("A", LabelId(0))
semanticTable.resolvedPropertyKeyNames.put("prop", PropertyKeyId(0))
implicit val selections = mock[Selections]
val label = LabelName("A")(InputPosition.NONE)
val propKey = PropertyKeyName("prop")(InputPosition.NONE)
when(selections.labelsOnNode(IdName("a"))).thenReturn(Set(label))
val stats = mock[GraphStatistics]
when(stats.indexSelectivity(LabelId(0), PropertyKeyId(0))).thenReturn(Some(Selectivity.of(.01).get))
when(stats.indexPropertyExistsSelectivity(LabelId(0), PropertyKeyId(0))).thenReturn(Some(Selectivity.ONE))
val calculator = ExpressionSelectivityCalculator(stats, IndependenceCombiner)
val prefixes = Map("p" -> 0.23384596099184043,
"p2" -> 0.2299568541948447,
"p33" -> 0.22801230079634685,
"p5555" -> 0.22606774739784896,
"reallylong" -> 0.22429997158103274)
prefixes.foreach { case (prefix, selectivity) =>
val actual = calculator(StartsWith(Property(Variable("a") _, propKey) _, StringLiteral(prefix)(InputPosition.NONE)) _)
assert( actual.factor === selectivity +- selectivity * 0.000000000000001)
}
}
test("Selectivity should never be worse than corresponding existence selectivity") {
implicit val semanticTable = SemanticTable()
semanticTable.resolvedLabelIds.put("A", LabelId(0))
semanticTable.resolvedPropertyKeyNames.put("prop", PropertyKeyId(0))
implicit val selections = mock[Selections]
val label = LabelName("A")(InputPosition.NONE)
val propKey = PropertyKeyName("prop")(InputPosition.NONE)
when(selections.labelsOnNode(IdName("a"))).thenReturn(Set(label))
val stats = mock[GraphStatistics]
when(stats.indexSelectivity(LabelId(0), PropertyKeyId(0))).thenReturn(Some(Selectivity.of(0.01).get))
val existenceSelectivity = .2285
when(stats.indexPropertyExistsSelectivity(LabelId(0), PropertyKeyId(0))).thenReturn(Some(Selectivity.of(existenceSelectivity).get))
val calculator = ExpressionSelectivityCalculator(stats, IndependenceCombiner)
val prefixes = Map("p" -> existenceSelectivity,
"p2" -> existenceSelectivity,
"p33" -> 0.22801230079634685,
"p5555" -> 0.22606774739784896,
"reallylong" -> 0.22429997158103274)
prefixes.foreach { case (prefix, selectivity) =>
val actual = calculator(StartsWith(Property(Variable("a") _, propKey) _, StringLiteral(prefix)(InputPosition.NONE)) _)
assert( actual.factor === selectivity +- selectivity * 0.000000000000001)
}
}
test("should default to single cardinality for HasLabels with previously unknown label") {
val stats = mock[GraphStatistics]
when(stats.nodesWithLabelCardinality(None)).thenReturn(Cardinality(10))
val calculator = ExpressionSelectivityCalculator(stats, IndependenceCombiner)
implicit val semanticTable = SemanticTable()
implicit val selections = mock[Selections]
val expr = HasLabels(null, Seq(LabelName("Foo")(pos)))(pos)
calculator(expr) should equal(Selectivity(1.0 / 10.0))
}
}