[GLUTEN-7220][CH]Fix expand bug in grouping sets query (apache#7221)

* fix grouping set * remove useless code
sharkdtu · Oct 14, 2024 · cb926ad · cb926ad
1 parent 6f9cf39
commit cb926ad
Show file tree

Hide file tree

Showing 6 changed files with 18 additions and 5 deletions.
diff --git a/...ouse/src/test/scala/org/apache/gluten/execution/hive/GlutenClickHouseHiveTableSuite.scala b/...ouse/src/test/scala/org/apache/gluten/execution/hive/GlutenClickHouseHiveTableSuite.scala
@@ -1415,4 +1415,5 @@ class GlutenClickHouseHiveTableSuite
     runQueryAndCompare(selectSql)(df => checkOperatorCount[ProjectExecTransformer](3)(df))
     spark.sql("DROP TABLE test_tbl_7054")
   }
+
 }
diff --git a/...est/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala b/...est/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
@@ -2991,5 +2991,17 @@ class GlutenClickHouseTPCHSaltNullParquetSuite extends GlutenClickHouseTPCHAbstr
       checkGlutenOperatorMatch[ProjectExecTransformer]
     }
   }
+
+  test("GLUTEN-7220: Fix bug of grouping sets") {
+    val table_create_sql = "create table test_tbl_7220(id bigint, name string) using parquet"
+    val insert_data_sql = "insert into test_tbl_7220 values(1, 'a123'), (2, 'a124'), (3, 'a125')"
+    val query_sql = "select '2024-08-26' as day, id,name from" +
+      " (select id, name from test_tbl_7220 group by id, name grouping sets((id),(id,name))) " +
+      " where name  = 'a124'"
+    spark.sql(table_create_sql)
+    spark.sql(insert_data_sql)
+    compareResultsAgainstVanillaSpark(query_sql, true, { _ => })
+    spark.sql("drop table test_tbl_7220")
+  }
 }
 // scalastyle:on line.size.limit
diff --git a/cpp-ch/local-engine/Operator/DefaultHashAggregateResult.cpp b/cpp-ch/local-engine/Operator/DefaultHashAggregateResult.cpp
@@ -22,7 +22,7 @@
 #include <Core/ColumnsWithTypeAndName.h>
 #include <DataTypes/DataTypeNullable.h>
 #include <DataTypes/DataTypesNumber.h>
-#include <Operator/ExpandTransorm.h>
+#include <Operator/ExpandTransform.h>
 #include <Processors/Chunk.h>
 #include <Processors/IProcessor.h>
 #include <Processors/Transforms/AggregatingTransform.h>

diff --git a/cpp-ch/local-engine/Operator/ExpandStep.cpp b/cpp-ch/local-engine/Operator/ExpandStep.cpp
@@ -22,7 +22,7 @@
 #include <Core/ColumnsWithTypeAndName.h>
 #include <DataTypes/DataTypeNullable.h>
 #include <DataTypes/DataTypesNumber.h>
-#include <Operator/ExpandTransorm.h>
+#include <Operator/ExpandTransform.h>
 #include <Processors/IProcessor.h>
 #include <QueryPipeline/Pipe.h>
 #include <QueryPipeline/QueryPipelineBuilder.h>

diff --git a/cpp-ch/local-engine/Operator/ExpandTransform.cpp b/cpp-ch/local-engine/Operator/ExpandTransform.cpp
@@ -27,7 +27,7 @@
 #include <Common/Exception.h>
 #include <Common/logger_useful.h>
 
-#include "ExpandTransorm.h"
+#include "ExpandTransform.h"
 
 namespace DB
 {
@@ -122,8 +122,8 @@ void ExpandTransform::work()
         else if (kind == EXPAND_FIELD_KIND_LITERAL)
         {
             /// Add const column with field value
-            auto column = type->createColumnConst(rows, field);
-            columns[col_i] = column;
+            auto column = type->createColumnConst(rows, field)->convertToFullColumnIfConst();
+            columns[col_i] = std::move(column);
         }
         else
             throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Unknown ExpandFieldKind {}", magic_enum::enum_name(kind));

diff --git a/...ch/local-engine/Operator/ExpandTransorm.h → ...h/local-engine/Operator/ExpandTransform.h b/...ch/local-engine/Operator/ExpandTransorm.h → ...h/local-engine/Operator/ExpandTransform.h
-Original file line number
+Diff line change
@@ Expand Up / @@ -1415,4 +1415,5 @@ class GlutenClickHouseHiveTableSuite @@
         runQueryAndCompare(selectSql)(df => checkOperatorCount[ProjectExecTransformer](3)(df))
         spark.sql("DROP TABLE test_tbl_7054")
       }
     }