debug

apache · Feb 25, 2025 · fe580b2 · fe580b2
1 parent 172638e
commit fe580b2
Show file tree

Hide file tree

Showing 4 changed files with 21 additions and 0 deletions.
diff --git a/...ouse/src/test/scala/org/apache/gluten/execution/GlutenCoalesceAggregationUnionSuite.scala b/...ouse/src/test/scala/org/apache/gluten/execution/GlutenCoalesceAggregationUnionSuite.scala
@@ -453,6 +453,23 @@ class GlutenCoalesceAggregationUnionSuite extends GlutenClickHouseWholeStageTran
     compareResultsAgainstVanillaSpark(sql, true, checkNoUnion, true)
   }
 
+  test("coalesce project union. case 1") {
+
+    val sql =
+      """
+        |select a, x, y, t from (
+        | select a, x, y, 1 as t from coalesce_union_t1 where b % 2 = 0
+        | union all
+        | select a, x, y, 2 as t from coalesce_union_t1 where b % 3 = 1
+        | union all
+        | select a, x, y, 3 as t from coalesce_union_t1 where b % 4 = 1
+        | union all
+        | select a, x, y, 4 as t from coalesce_union_t1 where b % 5 = 1
+        |) order by a, x, y, t
+        |""".stripMargin
+    compareResultsAgainstVanillaSpark(sql, true, checkNoUnion, true)
+  }
+
   test("no coalesce project union. case 1") {
     val sql =
       """

diff --git a/...en-ut/spark33/src/test/scala/org/apache/spark/sql/GlutenDataFrameSetOperationsSuite.scala b/...en-ut/spark33/src/test/scala/org/apache/spark/sql/GlutenDataFrameSetOperationsSuite.scala
@@ -24,5 +24,6 @@ class GlutenDataFrameSetOperationsSuite
   override def sparkConf: SparkConf =
     super.sparkConf
       .set("spark.gluten.sql.columnar.backend.ch.enable.coalesce.project.union", "false")
+      .set("spark.gluten.sql.columnar.backend.ch.enable.coalesce.aggregation.union", "false")
 
 }
diff --git a/...ut/spark35/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala b/...ut/spark35/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
@@ -622,6 +622,8 @@ class ClickHouseTestSettings extends BackendTestSettings {
     .exclude("ordering and partitioning reporting")
   enableSuite[GlutenDatasetAggregatorSuite]
   enableSuite[GlutenDatasetCacheSuite]
+    // Disable this since coalesece union clauses rule will rewrite the query.
+    .exclude("SPARK-44653: non-trivial DataFrame unions should not break caching")
   enableSuite[GlutenDatasetOptimizationSuite]
   enableSuite[GlutenDatasetPrimitiveSuite]
   enableSuite[GlutenDatasetSerializerRegistratorSuite]

diff --git a/...en-ut/spark35/src/test/scala/org/apache/spark/sql/GlutenDataFrameSetOperationsSuite.scala b/...en-ut/spark35/src/test/scala/org/apache/spark/sql/GlutenDataFrameSetOperationsSuite.scala
@@ -24,5 +24,6 @@ class GlutenDataFrameSetOperationsSuite
   override def sparkConf: SparkConf =
     super.sparkConf
       .set("spark.gluten.sql.columnar.backend.ch.enable.coalesce.project.union", "false")
+      .set("spark.gluten.sql.columnar.backend.ch.enable.coalesce.aggregation.union", "false")
 
 }