Skip to content

[SPARK-40429][SQL] Only set KeyGroupedPartitioning when the referenced column is in the output #37886

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
huaxingao wants to merge 3 commits into from
Closed

[SPARK-40429][SQL] Only set KeyGroupedPartitioning when the referenced column is in the output #37886

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
51c6160
[SPARK-40429][SQL] Only set KeyGroupedPartitioning when the reference…
huaxingao Sep 14, 2022
ba9f4ca
fix style
huaxingao Sep 14, 2022
ab5ea9b
address comments
huaxingao Sep 15, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 14 additions & 3 deletions ...n/scala/org/apache/spark/sql/execution/datasources/v2/V2ScanPartitioningAndOrdering.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@
package org.apache.spark.sql.execution.datasources.v2

import org.apache.spark.sql.catalyst.SQLConfHelper
import org.apache.spark.sql.catalyst.expressions.V2ExpressionUtils
import org.apache.spark.sql.catalyst.expressions.{AttributeSet, V2ExpressionUtils}
import org.apache.spark.sql.catalyst.plans.logical.LogicalPlan
import org.apache.spark.sql.catalyst.rules.Rule
import org.apache.spark.sql.connector.read.{SupportsReportOrdering, SupportsReportPartitioning}
Expand All @@ -41,8 +41,19 @@ object V2ScanPartitioningAndOrdering extends Rule[LogicalPlan] with SQLConfHelpe
private def partitioning(plan: LogicalPlan) = plan.transformDown {
case d @ DataSourceV2ScanRelation(relation, scan: SupportsReportPartitioning, _, None, _) =>
val catalystPartitioning = scan.outputPartitioning() match {
case kgp: KeyGroupedPartitioning => sequenceToOption(kgp.keys().map(
V2ExpressionUtils.toCatalystOpt(_, relation, relation.funCatalog)))
case kgp: KeyGroupedPartitioning =>
val partitioning = sequenceToOption(
kgp.keys().map(V2ExpressionUtils.toCatalystOpt(_, relation, relation.funCatalog)))
if (partitioning.isEmpty) {
None
} else {
val ref = AttributeSet.fromAttributeSets(partitioning.get.map(_.references))
Copy link
Contributor

@cloud-fan cloud-fan Sep 15, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

how about partitioning.get.forall(p => p.references.subsetOf(...))

Copy link
Contributor Author

@huaxingao huaxingao Sep 15, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sounds good! Changed.

if (ref.subsetOf(AttributeSet(d.output))) {
Copy link
Contributor

@cloud-fan cloud-fan Sep 15, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
if (ref.subsetOf(AttributeSet(d.output))) {
if (ref.subsetOf(d.outputSet)) {

partitioning
} else {
None
}
}
case _: UnknownPartitioning => None
case p => throw new IllegalArgumentException("Unsupported data source V2 partitioning " +
"type: " + p.getClass.getSimpleName)
Expand Down
16 changes: 16 additions & 0 deletions sql/core/src/test/scala/org/apache/spark/sql/connector/MetadataColumnSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -216,4 +216,20 @@ class MetadataColumnSuite extends DatasourceV2SQLBase {
.withColumn("right_all", struct($"right.*"))
checkAnswer(dfQuery, Row(1, "a", "b", Row(1, "a"), Row(1, "b")))
}

test("SPARK-40429: Only set KeyGroupedPartitioning when the referenced column is in the output") {
withTable(tbl) {
sql(s"CREATE TABLE $tbl (id bigint, data string) PARTITIONED BY (id)")
sql(s"INSERT INTO $tbl VALUES (1, 'a'), (2, 'b'), (3, 'c')")
checkAnswer(
spark.table(tbl).select("index", "_partition"),
Seq(Row(0, "3"), Row(0, "2"), Row(0, "1"))
)

checkAnswer(
spark.table(tbl).select("id", "index", "_partition"),
Seq(Row(3, 0, "3"), Row(2, 0, "2"), Row(1, 0, "1"))
)
}
}
}