Github user ravipesala commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/1728#discussion_r159056348
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/hive/CarbonPreAggregateRules.scala
---
@@ -330,6 +207,264 @@ case class CarbonPreAggregateQueryRules(sparkSession:
SparkSession) extends Rule
}
}
+ /**
+ * Below method will be used to validate the logical plan
+ * and get all the details from to select proper aggregate table
+ * @param logicalPlan
+ * actual query logical plan
+ * @param list
+ * list of projection column present in plan
+ * @param qAggExprs
+ * list of aggregate expression
+ * @return if plan is valid for tranformation, parent table, parent
logical relaion
+ */
+ def validatePlanAndGetFields(logicalPlan: LogicalPlan,
+ list: scala.collection.mutable.HashSet[QueryColumn],
+ qAggExprs: scala.collection.mutable.HashSet[AggregateExpression]):
(Boolean,
+ CarbonTable, LogicalRelation) = {
+ var isValidPlan = false
+ var pTable: CarbonTable = null
+ var qLRelation: LogicalRelation = null
+ logicalPlan.transform {
--- End diff --
It seems Join condition of two plans is not handled here. Please handle
Join, Union cases also using recursion.
---
