From 4a9d538dd08350e7c6e2e24b30cecffef027cc5d Mon Sep 17 00:00:00 2001
From: Mariam-Almesfer <mariam.almesfer@ibm.com>
Date: Sun, 10 May 2026 13:37:12 +0300
Subject: [PATCH] Enable hour(timestamp_ntz) native execution in Velox backend

---
 .../DateFunctionsValidateSuite.scala          |  7 ++-----
 .../columnar/validator/Validators.scala       | 19 +++++++++++++++++--
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/backends-velox/src/test/scala/org/apache/gluten/functions/DateFunctionsValidateSuite.scala b/backends-velox/src/test/scala/org/apache/gluten/functions/DateFunctionsValidateSuite.scala
index e5e25c88b6f..47b3e5b692e 100644
--- a/backends-velox/src/test/scala/org/apache/gluten/functions/DateFunctionsValidateSuite.scala
+++ b/backends-velox/src/test/scala/org/apache/gluten/functions/DateFunctionsValidateSuite.scala
@@ -599,12 +599,9 @@ class DateFunctionsValidateSuite extends FunctionsValidateSuite {
           checkGlutenPlan[BatchScanExecTransformer]
         }
 
-        // Ensures the fallback of unsupported function works.
+        // hour(timestamp_ntz) runs natively; output is int (no NTZ propagation).
         runQueryAndCompare("select hour(ts) from view") {
-          df =>
-            assert(collect(df.queryExecution.executedPlan) {
-              case p if p.isInstanceOf[ProjectExec] => p
-            }.nonEmpty)
+          checkGlutenPlan[ProjectExecTransformer]
         }
     }
   }
diff --git a/gluten-substrait/src/main/scala/org/apache/gluten/extension/columnar/validator/Validators.scala b/gluten-substrait/src/main/scala/org/apache/gluten/extension/columnar/validator/Validators.scala
index 3218e045f54..1290527a5b3 100644
--- a/gluten-substrait/src/main/scala/org/apache/gluten/extension/columnar/validator/Validators.scala
+++ b/gluten-substrait/src/main/scala/org/apache/gluten/extension/columnar/validator/Validators.scala
@@ -262,9 +262,24 @@ object Validators {
           case p if HiveTableScanExecTransformer.isHiveTableScan(p) => true
           case _ => false
         }
-        val hasNTZ = plan.output.exists(a => containsNTZ(a.dataType)) ||
+        def ntzOnlyFromFileScans(p: SparkPlan): Boolean = {
+          if (!p.output.exists(a => containsNTZ(a.dataType))) {
+            true
+          } else {
+            p match {
+              case _: BatchScanExec => true
+              case _: FileSourceScanExec => true
+              case q if HiveTableScanExecTransformer.isHiveTableScan(q) => true
+              case _ if p.children.isEmpty => false
+              case _ => p.children.forall(ntzOnlyFromFileScans)
+            }
+          }
+        }
+        val outputHasNTZ = plan.output.exists(a => containsNTZ(a.dataType))
+        val writeChildHasNTZ = plan.isInstanceOf[WriteFilesExec] &&
           plan.children.exists(_.output.exists(a => containsNTZ(a.dataType)))
-        if (isScan || !hasNTZ) {
+        val childNTZFromFileScans = plan.children.forall(ntzOnlyFromFileScans)
+        if (isScan || (!outputHasNTZ && !writeChildHasNTZ && childNTZFromFileScans)) {
           return pass()
         }
       }