[SPARK-45242][SQL][FOLLOWUP]Canonicalize DataFrame ID in CollectMetrics

apache · Oct 31, 2023 · 521805b · 521805b
1 parent 2fd1f10
commit 521805b
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 0 deletions.
diff --git a/...st/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala b/...st/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/basicLogicalOperators.scala
@@ -1983,6 +1983,10 @@ case class CollectMetrics(
 
   override protected def withNewChildInternal(newChild: LogicalPlan): CollectMetrics =
     copy(child = newChild)
+
+  override def doCanonicalize(): LogicalPlan = {
+    super.doCanonicalize().asInstanceOf[CollectMetrics].copy(dataframeId = 0L)
+  }
 }
 
 /**

diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/analysis/AnalysisSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/analysis/AnalysisSuite.scala
@@ -903,6 +903,11 @@ class AnalysisSuite extends AnalysisTest with Matchers {
     )
   }
 
+  test("Canonicalize CollectMetrics") {
+    assert(CollectMetrics("", Seq(Rand(10).as("rnd")), testRelation, 1).canonicalized ==
+      CollectMetrics("", Seq(Rand(10).as("rnd")), testRelation, 2).canonicalized)
+  }
+
   test("Analysis exceed max iterations") {
     // RuleExecutor only throw exception or log warning when the rule is supposed to run
     // more than once.