diff --git a/backends-velox/src/main/scala/org/apache/gluten/extension/EmptySchemaWorkaround.scala b/backends-velox/src/main/scala/org/apache/gluten/extension/EmptySchemaWorkaround.scala index eae4dd063984..3f34e7fc262d 100644 --- a/backends-velox/src/main/scala/org/apache/gluten/extension/EmptySchemaWorkaround.scala +++ b/backends-velox/src/main/scala/org/apache/gluten/extension/EmptySchemaWorkaround.scala @@ -31,6 +31,7 @@ import org.apache.spark.sql.types.StringType /** Rules to make Velox backend work correctly with query plans that have empty output schemas. */ object EmptySchemaWorkaround { + /** * This rule plans [[RDDScanExec]] with a fake schema to make gluten work, because gluten does not * support empty output relation, see [[FallbackEmptySchemaRelation]].