Github user ravipesala commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/1521#discussion_r154637843
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/hive/CarbonPreAggregateRules.scala
---
@@ -197,8 +197,17 @@ case class CarbonPreAggregateQueryRules(sparkSession:
SparkSession) extends Rule
.asInstanceOf[LogicalRelation]
(selectedDataMapSchema, carbonRelation)
}.minBy(f =>
f._2.relation.asInstanceOf[CarbonDatasourceHadoopRelation].sizeInBytes)
- // transform the query plan based on selected child schema
- transformPreAggQueryPlan(plan, aggDataMapSchema, carbonRelation)
+ if
(carbonRelation.relation.asInstanceOf[CarbonDatasourceHadoopRelation].sizeInBytes
==
--- End diff --
While calculating `sizeInBytes` in CarbonRelation we can first check the
size of the valid segments and then calculate the store size of that table.
---
