[VL] Use collect_list aggregate function in velox (apache#5285)

guixiaowen · Apr 9, 2024 · c54b559 · c54b559
1 parent 8ec0e30
commit c54b559
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 4 deletions.
diff --git a/cpp/velox/substrait/SubstraitParser.cc b/cpp/velox/substrait/SubstraitParser.cc
@@ -401,8 +401,7 @@ std::unordered_map<std::string, std::string> SubstraitParser::substraitVeloxFunc
     {"murmur3hash", "hash_with_seed"},
     {"modulus", "remainder"},
     {"date_format", "format_datetime"},
-    {"collect_set", "set_agg"},
-    {"collect_list", "array_agg"}};
+    {"collect_set", "set_agg"}};
 
 const std::unordered_map<std::string, std::string> SubstraitParser::typeMap_ = {
     {"bool", "BOOLEAN"},

diff --git a/gluten-core/src/main/scala/org/apache/gluten/extension/RewriteCollect.scala b/gluten-core/src/main/scala/org/apache/gluten/extension/RewriteCollect.scala
@@ -20,7 +20,7 @@ import org.apache.gluten.backendsapi.BackendsApiManager
 import org.apache.gluten.utils.PullOutProjectHelper
 
 import org.apache.spark.sql.catalyst.expressions.{And, Attribute, AttributeSet, If, IsNotNull, IsNull, Literal, NamedExpression}
-import org.apache.spark.sql.catalyst.expressions.aggregate.{AggregateExpression, CollectList, CollectSet, Complete, Final, Partial}
+import org.apache.spark.sql.catalyst.expressions.aggregate.{AggregateExpression, CollectSet, Complete, Final, Partial}
 import org.apache.spark.sql.catalyst.rules.Rule
 import org.apache.spark.sql.execution.SparkPlan
 import org.apache.spark.sql.execution.aggregate.BaseAggregateExec
@@ -53,7 +53,7 @@ object RewriteCollect extends Rule[SparkPlan] with PullOutProjectHelper {
 
   private def shouldReplaceNullToEmptyArray(ae: AggregateExpression): Boolean = {
     ae.aggregateFunction match {
-      case _: CollectSet | _: CollectList =>
+      case _: CollectSet =>
         ae.mode match {
           case Final | Complete => true
           case _ => false