uros-b commented on code in PR #57629:
URL: https://github.com/apache/spark/pull/57629#discussion_r3683325662
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/InsertMapSortExpression.scala:
##########
@@ -42,42 +44,73 @@ object InsertMapSortInGroupingExpressions extends
Rule[LogicalPlan] {
if (!plan.containsPattern(AGGREGATE)) {
return plan
}
- val shouldRewrite = plan.exists {
- case agg: Aggregate if
agg.groupingExpressions.exists(mapTypeExistsRecursively) => true
- case _ => false
- }
- if (!shouldRewrite) {
- return plan
- }
plan transformUpWithNewOutput {
case agg @ Aggregate(groupingExprs, aggregateExpressions, child, hint) =>
- val exprToMapSort = new mutable.HashMap[Expression, NamedExpression]
- val newGroupingKeys = groupingExprs.map { expr =>
- val inserted = insertMapSortRecursively(expr)
- if (expr.ne(inserted)) {
- exprToMapSort.getOrElseUpdate(
- expr.canonicalized,
- Alias(inserted, "_groupingmapsort")()
- ).toAttribute
+ val distinctExpressions =
+ if
(conf.getConf(SQLConf.INSERT_MAP_SORT_IN_DISTINCT_AGGREGATES_ENABLED)) {
+ distinctAggregateChildren(aggregateExpressions)
} else {
- expr
+ Seq.empty
}
+ val expressionsToNormalize = groupingExprs ++ distinctExpressions
+ if (!expressionsToNormalize.exists(mapTypeExistsRecursively)) {
+ agg -> Nil
+ } else {
+ val groupingMapSortAliases = mapSortAliases(groupingExprs)
+ val distinctMapSortAliases = mapSortAliases(distinctExpressions,
groupingMapSortAliases)
+ val newGroupingKeys = groupingExprs.map { expr =>
+
groupingMapSortAliases.get(expr.canonicalized).map(_.toAttribute).getOrElse(expr)
+ }
+ val newAggregateExprs = aggregateExpressions.map {
+ case named if groupingMapSortAliases.contains(named.canonicalized)
=>
+ // If we replace the top-level named expr, then should add back
the original name
+
groupingMapSortAliases(named.canonicalized).toAttribute.withName(named.name)
+ case other =>
+ other.transformDown {
+ case ae: AggregateExpression if ae.isDistinct =>
+ ae.copy(aggregateFunction =
ae.aggregateFunction.withNewChildren(
+ ae.aggregateFunction.children.map { child =>
+ distinctMapSortAliases.get(child.canonicalized)
+ .map(_.toAttribute).getOrElse(child)
+ }).asInstanceOf[AggregateFunction])
+ case e =>
+
groupingMapSortAliases.get(e.canonicalized).map(_.toAttribute).getOrElse(e)
+ }.asInstanceOf[NamedExpression]
+ }
+ val newChild = Project(
+ child.output ++ (groupingMapSortAliases ++
distinctMapSortAliases).values,
+ child)
+ val newAgg = Aggregate(newGroupingKeys, newAggregateExprs, newChild,
hint)
+ newAgg -> agg.output.zip(newAgg.output)
}
- val newAggregateExprs = aggregateExpressions.map {
- case named if exprToMapSort.contains(named.canonicalized) =>
- // If we replace the top-level named expr, then should add back
the original name
- exprToMapSort(named.canonicalized).toAttribute.withName(named.name)
- case other =>
- other.transformUp {
- case e =>
exprToMapSort.get(e.canonicalized).map(_.toAttribute).getOrElse(e)
- }.asInstanceOf[NamedExpression]
- }
- val newChild = Project(child.output ++ exprToMapSort.values, child)
- val newAgg = Aggregate(newGroupingKeys, newAggregateExprs, newChild,
hint)
- newAgg -> agg.output.zip(newAgg.output)
}
}
+
+ private def distinctAggregateChildren(
+ aggregateExpressions: Seq[NamedExpression]): Seq[Expression] = {
+ aggregateExpressions
+ .flatMap(_.collect {
+ case ae: AggregateExpression if ae.isDistinct => ae
+ })
+ .flatMap(_.aggregateFunction.children)
+ }
+
+ private def mapSortAliases(
+ expressions: Seq[Expression],
+ existingAliases: Map[Expression, NamedExpression] = Map.empty) = {
+ val aliases = new mutable.HashMap[Expression, NamedExpression]
+ expressions.foreach { expr =>
+ val inserted = insertMapSortRecursively(expr)
Review Comment:
insertMapSortRecursively is now applied to expressions that aren't
attributes, which breaks the invariant the rule's placement was chosen to
guarantee. The rule sits immediately after PullOutGroupingExpressions (in
Optimizer) for exactly this reason.
The cleanest fix is to alias the raw distinct argument into the new Project
and run insertMapSortRecursively on the resulting attribute (two stacked
projections, or teach PullOutGroupingExpressions to pull out distinct-aggregate
arguments too), which restores the attribute-only invariant. Either way, the
placement comment above InsertMapSortInGroupingExpressions needs updating,
since it no longer describes what the rule does.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
