[GitHub] [spark] singhpk234 commented on a change in pull request #35924: [SPARK-38578][SQL] Avoid unnecessary sort in FileFormatWriter if user has specified sort in AQE

[GitBox <gi...@apache.org>]

singhpk234 commented on a change in pull request #35924:
URL: https://github.com/apache/spark/pull/35924#discussion_r831779684



##########
File path: sql/core/src/main/scala/org/apache/spark/sql/execution/adaptive/AQEUtils.scala
##########
@@ -57,4 +58,24 @@ object AQEUtils {
       }
     case _ => Some(UnspecifiedDistribution)
   }
+
+  // Analyze the given plan and calculate the required ordering of this plan w.r.t. the
+  // user-specified sort.
+  def getRequiredOrdering(p: SparkPlan): Seq[SortOrder] = p match {
+    case f: FilterExec => getRequiredOrdering(f.child)
+    case c: CollectMetricsExec => getRequiredOrdering(c.child)
+    // We do not need to care whether the sort is global or not, since the output partitioning
+    // is ensured by requiredDistribution.
+    case s: SortExec => s.outputOrdering
+    case p: ProjectExec =>
+      val requiredOrdering = getRequiredOrdering(p.child)
+      // avoid case `df.sort(a, b).select(c)`
+      if (requiredOrdering.forall(e => p.projectList.exists(_.semanticEquals(e)))) {
+        requiredOrdering

Review comment:
       [question] can we handle aliasing as well. 
    For ex :  `df.sort(a, b).select(col('a').as('x'))`




-- 
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: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org