This is an automated email from the ASF dual-hosted git repository.
gurwls223 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/master by this push:
new 8bca01e30ae [SPARK-44523][SQL] Filter's maxRows/maxRowsPerPartition is
0 if condition is FalseLiteral
8bca01e30ae is described below
commit 8bca01e30ae2dd0df06ec96926ad481a1a7050a2
Author: Yuming Wang <[email protected]>
AuthorDate: Tue Jul 25 12:06:06 2023 +0900
[SPARK-44523][SQL] Filter's maxRows/maxRowsPerPartition is 0 if condition
is FalseLiteral
### What changes were proposed in this pull request?
This PR updates Filter's maxRows/maxRowsPerPartition to 0 if condition is
`FalseLiteral`.
### Why are the changes needed?
Provide an accurate value, as some optimization rules use this value to
optimize queries.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Unit test.
Closes #42126 from wangyum/SPARK-44523.
Lead-authored-by: Yuming Wang <[email protected]>
Co-authored-by: Yuming Wang <[email protected]>
Signed-off-by: Hyukjin Kwon <[email protected]>
---
.../sql/catalyst/plans/logical/basicLogicalOperators.scala | 11 +++++++++--
.../apache/spark/sql/catalyst/plans/LogicalPlanSuite.scala | 6 ++++++
2 files changed, 15 insertions(+), 2 deletions(-)
diff --git
a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala
b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala
index 4bb830662a3..57e4b31dbe0 100644
---
a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala
+++
b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala
@@ -303,8 +303,15 @@ case class Filter(condition: Expression, child:
LogicalPlan)
extends OrderPreservingUnaryNode with PredicateHelper {
override def output: Seq[Attribute] = child.output
- override def maxRows: Option[Long] = child.maxRows
- override def maxRowsPerPartition: Option[Long] = child.maxRowsPerPartition
+ override def maxRows: Option[Long] = condition match {
+ case Literal.FalseLiteral => Some(0L)
+ case _ => child.maxRows
+ }
+
+ override def maxRowsPerPartition: Option[Long] = condition match {
+ case Literal.FalseLiteral => Some(0L)
+ case _ => child.maxRowsPerPartition
+ }
final override val nodePatterns: Seq[TreePattern] = Seq(FILTER)
diff --git
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/plans/LogicalPlanSuite.scala
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/plans/LogicalPlanSuite.scala
index 1d533e9d0d4..a3bdfd07aee 100644
---
a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/plans/LogicalPlanSuite.scala
+++
b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/plans/LogicalPlanSuite.scala
@@ -139,4 +139,10 @@ class LogicalPlanSuite extends SparkFunSuite {
assert(sample.maxRows === Some(200))
assert(sample.maxRowsPerPartition === Some(68))
}
+
+ test("SPARK-44523: Filter's maxRows/maxRowsPerPartition is 0 if condition is
FalseLiteral") {
+ val query = Range(0, 100, 1, 10)
+ assert(query.where(Literal.FalseLiteral).maxRows.contains(0))
+ assert(query.where(Literal.FalseLiteral).maxRowsPerPartition.contains(0))
+ }
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
