[GitHub] [spark] cloud-fan commented on a diff in pull request #38497: [SPARK-40999] Hint propagation to subqueries

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

cloud-fan commented on code in PR #38497:
URL: https://github.com/apache/spark/pull/38497#discussion_r1018755179


##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/EliminateResolvedHint.scala:
##########
@@ -31,20 +34,35 @@ object EliminateResolvedHint extends Rule[LogicalPlan] {
   // This is also called in the beginning of the optimization phase, and as a result
   // is using transformUp rather than resolveOperators.
   def apply(plan: LogicalPlan): LogicalPlan = {
-    val pulledUp = plan transformUp {
+    val joinsWithHints = plan transformUp {
       case j: Join if j.hint == JoinHint.NONE =>
         val (newLeft, leftHints) = extractHintsFromPlan(j.left)
         val (newRight, rightHints) = extractHintsFromPlan(j.right)
         val newJoinHint = JoinHint(mergeHints(leftHints), mergeHints(rightHints))
         j.copy(left = newLeft, right = newRight, hint = newJoinHint)
     }
-    pulledUp.transformUp {
+    val shouldPullHintsIntoSubqueries = SQLConf.get.getConf(SQLConf.PULL_HINTS_INTO_SUBQUERIES)
+    val joinsAndSubqueriesWithHints = if (shouldPullHintsIntoSubqueries) {
+      pullHintsIntoSubqueries(joinsWithHints)
+    } else {
+      joinsWithHints
+    }
+    joinsAndSubqueriesWithHints.transformUp {
       case h: ResolvedHint =>
         hintErrorHandler.joinNotFoundForJoinHint(h.hints)
         h.child
     }
   }
 
+  def pullHintsIntoSubqueries(plan: LogicalPlan): LogicalPlan = {
+    plan.transformAllExpressionsUpWithPruning(_.containsPattern(PLAN_EXPRESSION)) {
+      case s: SubqueryExpression if s.hint.isEmpty =>
+        val (newPlan, subqueryHints) = extractHintsFromPlan(s.plan)

Review Comment:
   question: what if the subquery is complicated and has joins, the hint is for joins in subquery but not subquery itself?



-- 
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