Github user jackylk commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/1759#discussion_r159589011
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/execution/command/management/RefreshCarbonTableCommand.scala
---
@@ -91,6 +95,11 @@ case class RefreshCarbonTableCommand(
registerAggregates(databaseName, dataMapSchemaList)(sparkSession)
}
registerTableWithHive(databaseName, tableName,
tableInfo)(sparkSession)
+ // Register partitions to hive metastore in case of hive
partitioning carbon table
+ if (tableInfo.getFactTable.getPartitionInfo != null &&
+ tableInfo.getFactTable.getPartitionInfo.getPartitionType ==
PartitionType.NATIVE_HIVE) {
+ registerAllPartitionsToHive(tablePath, databaseName, tableName,
sparkSession)
--- End diff --
I think passing AbsolutionTableIdentifier is better, it has all
information. It is created in line 65
---
