Github user mgaido91 commented on a diff in the pull request:
https://github.com/apache/spark/pull/21072#discussion_r183311275
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
---
@@ -736,12 +736,22 @@ object EliminateSorts extends Rule[LogicalPlan] {
}
/**
- * Removes Sort operation if the child is already sorted
+ * Removes redundant Sort operation. This can happen:
+ * 1) if the child is already sorted
+ * 2) if there is another Sort operator separated by 0...n Project/Filter
operators
*/
object RemoveRedundantSorts extends Rule[LogicalPlan] {
def apply(plan: LogicalPlan): LogicalPlan = plan transform {
case Sort(orders, true, child) if
SortOrder.orderingSatisfies(child.outputOrdering, orders) =>
child
+ case s @ Sort(_, _, child) => s.copy(child =
recursiveRemoveSort(child))
+ }
+
+ def recursiveRemoveSort(plan: LogicalPlan): LogicalPlan = plan match {
+ case Project(fields, child) => Project(fields,
recursiveRemoveSort(child))
+ case Filter(condition, child) => Filter(condition,
recursiveRemoveSort(child))
--- End diff --
why do you think we should check for the filter condition and the projected
items to be deterministic?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
