|
| 1 | +/* |
| 2 | + * Licensed to the Apache Software Foundation (ASF) under one or more |
| 3 | + * contributor license agreements. See the NOTICE file distributed with |
| 4 | + * this work for additional information regarding copyright ownership. |
| 5 | + * The ASF licenses this file to You under the Apache License, Version 2.0 |
| 6 | + * (the "License"); you may not use this file except in compliance with |
| 7 | + * the License. You may obtain a copy of the License at |
| 8 | + * |
| 9 | + * http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | + * |
| 11 | + * Unless required by applicable law or agreed to in writing, software |
| 12 | + * distributed under the License is distributed on an "AS IS" BASIS, |
| 13 | + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 14 | + * See the License for the specific language governing permissions and |
| 15 | + * limitations under the License. |
| 16 | + */ |
| 17 | + |
| 18 | +package org.apache.spark.sql.catalyst.optimizer |
| 19 | + |
| 20 | +import org.apache.spark.sql.catalyst.analysis.EliminateSubqueryAliases |
| 21 | +import org.apache.spark.sql.catalyst.dsl.expressions._ |
| 22 | +import org.apache.spark.sql.catalyst.dsl.plans._ |
| 23 | +import org.apache.spark.sql.catalyst.plans._ |
| 24 | +import org.apache.spark.sql.catalyst.plans.logical._ |
| 25 | +import org.apache.spark.sql.catalyst.rules._ |
| 26 | + |
| 27 | +/** |
| 28 | + * This test suite ensures that the [[PushDownPredicates]] actually does predicate pushdown in |
| 29 | + * an efficient manner. This is enforced by asserting that a single predicate pushdown can push |
| 30 | + * all predicate to bottom as much as possible. |
| 31 | + */ |
| 32 | +class FilterPushdownOnePassSuite extends PlanTest { |
| 33 | + |
| 34 | + object Optimize extends RuleExecutor[LogicalPlan] { |
| 35 | + val batches = |
| 36 | + Batch("Subqueries", Once, |
| 37 | + EliminateSubqueryAliases) :: |
| 38 | + // this batch must reach expected state in one pass |
| 39 | + Batch("Filter Pushdown One Pass", Once, |
| 40 | + ReorderJoin, |
| 41 | + PushDownPredicates |
| 42 | + ) :: Nil |
| 43 | + } |
| 44 | + |
| 45 | + val testRelation1 = LocalRelation('a.int, 'b.int, 'c.int) |
| 46 | + val testRelation2 = LocalRelation('a.int, 'd.int, 'e.int) |
| 47 | + |
| 48 | + test("really simple predicate push down") { |
| 49 | + val x = testRelation1.subquery('x) |
| 50 | + val y = testRelation2.subquery('y) |
| 51 | + |
| 52 | + val originalQuery = x.join(y).where("x.a".attr === 1) |
| 53 | + |
| 54 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 55 | + val correctAnswer = x.where("x.a".attr === 1).join(y).analyze |
| 56 | + |
| 57 | + comparePlans(optimized, correctAnswer) |
| 58 | + } |
| 59 | + |
| 60 | + test("push down conjunctive predicates") { |
| 61 | + val x = testRelation1.subquery('x) |
| 62 | + val y = testRelation2.subquery('y) |
| 63 | + |
| 64 | + val originalQuery = x.join(y).where("x.a".attr === 1 && "y.d".attr < 1) |
| 65 | + |
| 66 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 67 | + val correctAnswer = x.where("x.a".attr === 1).join(y.where("y.d".attr < 1)).analyze |
| 68 | + |
| 69 | + comparePlans(optimized, correctAnswer) |
| 70 | + } |
| 71 | + |
| 72 | + test("push down predicates for simple joins") { |
| 73 | + val x = testRelation1.subquery('x) |
| 74 | + val y = testRelation2.subquery('y) |
| 75 | + |
| 76 | + val originalQuery = |
| 77 | + x.where("x.c".attr < 0) |
| 78 | + .join(y.where("y.d".attr > 1)) |
| 79 | + .where("x.a".attr === 1 && "y.d".attr < 2) |
| 80 | + |
| 81 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 82 | + val correctAnswer = |
| 83 | + x.where("x.c".attr < 0 && "x.a".attr === 1) |
| 84 | + .join(y.where("y.d".attr > 1 && "y.d".attr < 2)).analyze |
| 85 | + |
| 86 | + comparePlans(optimized, correctAnswer) |
| 87 | + } |
| 88 | + |
| 89 | + test("push down top-level filters for cascading joins") { |
| 90 | + val x = testRelation1.subquery('x) |
| 91 | + val y = testRelation2.subquery('y) |
| 92 | + |
| 93 | + val originalQuery = |
| 94 | + y.join(x).join(x).join(x).join(x).join(x).where("y.d".attr === 0) |
| 95 | + |
| 96 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 97 | + val correctAnswer = y.where("y.d".attr === 0).join(x).join(x).join(x).join(x).join(x).analyze |
| 98 | + |
| 99 | + comparePlans(optimized, correctAnswer) |
| 100 | + } |
| 101 | + |
| 102 | + test("push down predicates for tree-like joins") { |
| 103 | + val x = testRelation1.subquery('x) |
| 104 | + val y1 = testRelation2.subquery('y1) |
| 105 | + val y2 = testRelation2.subquery('y2) |
| 106 | + |
| 107 | + val originalQuery = |
| 108 | + y1.join(x).join(x) |
| 109 | + .join(y2.join(x).join(x)) |
| 110 | + .where("y1.d".attr === 0 && "y2.d".attr === 3) |
| 111 | + |
| 112 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 113 | + val correctAnswer = |
| 114 | + y1.where("y1.d".attr === 0).join(x).join(x) |
| 115 | + .join(y2.where("y2.d".attr === 3).join(x).join(x)).analyze |
| 116 | + |
| 117 | + comparePlans(optimized, correctAnswer) |
| 118 | + } |
| 119 | + |
| 120 | + test("push down through join and project") { |
| 121 | + val x = testRelation1.subquery('x) |
| 122 | + val y = testRelation2.subquery('y) |
| 123 | + |
| 124 | + val originalQuery = |
| 125 | + x.where('a > 0).select('a, 'b) |
| 126 | + .join(y.where('d < 100).select('e)) |
| 127 | + .where("x.a".attr < 100) |
| 128 | + |
| 129 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 130 | + val correctAnswer = |
| 131 | + x.where('a > 0 && 'a < 100).select('a, 'b) |
| 132 | + .join(y.where('d < 100).select('e)).analyze |
| 133 | + |
| 134 | + comparePlans(optimized, correctAnswer) |
| 135 | + } |
| 136 | + |
| 137 | + test("push down through deep projects") { |
| 138 | + val x = testRelation1.subquery('x) |
| 139 | + |
| 140 | + val originalQuery = |
| 141 | + x.select(('a + 1) as 'a1, 'b) |
| 142 | + .select(('a1 + 1) as 'a2, 'b) |
| 143 | + .select(('a2 + 1) as 'a3, 'b) |
| 144 | + .select(('a3 + 1) as 'a4, 'b) |
| 145 | + .select('b) |
| 146 | + .where('b > 0) |
| 147 | + |
| 148 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 149 | + val correctAnswer = |
| 150 | + x.where('b > 0) |
| 151 | + .select(('a + 1) as 'a1, 'b) |
| 152 | + .select(('a1 + 1) as 'a2, 'b) |
| 153 | + .select(('a2 + 1) as 'a3, 'b) |
| 154 | + .select(('a3 + 1) as 'a4, 'b) |
| 155 | + .select('b).analyze |
| 156 | + |
| 157 | + comparePlans(optimized, correctAnswer) |
| 158 | + } |
| 159 | + |
| 160 | + test("push down through aggregate and join") { |
| 161 | + val x = testRelation1.subquery('x) |
| 162 | + val y = testRelation2.subquery('y) |
| 163 | + |
| 164 | + val left = x |
| 165 | + .where('c > 0) |
| 166 | + .groupBy('a)('a, count('b)) |
| 167 | + .subquery('left) |
| 168 | + val right = y |
| 169 | + .where('d < 0) |
| 170 | + .groupBy('a)('a, count('d)) |
| 171 | + .subquery('right) |
| 172 | + val originalQuery = left |
| 173 | + .join(right).where("left.a".attr < 100 && "right.a".attr < 100) |
| 174 | + |
| 175 | + val optimized = Optimize.execute(originalQuery.analyze) |
| 176 | + val correctAnswer = |
| 177 | + x.where('c > 0 && 'a < 100).groupBy('a)('a, count('b)) |
| 178 | + .join(y.where('d < 0 && 'a < 100).groupBy('a)('a, count('d))) |
| 179 | + .analyze |
| 180 | + |
| 181 | + comparePlans(optimized, correctAnswer) |
| 182 | + } |
| 183 | +} |
0 commit comments