Github user manishgupta88 commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/2109#discussion_r178761202
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/execution/command/partition/CarbonAlterTableDropHivePartitionCommand.scala
---
@@ -147,6 +152,18 @@ case class CarbonAlterTableDropHivePartitionCommand(
table.getDatabaseName,
table.getTableName,
locksToBeAcquired)(sparkSession)
+ // If flow is for child table then get the uuid from operation
context.
+ // If flow is for parent table then generate uuid for child flows
and set the uuid to ""
+ // for parent table
+ // If normal table then set uuid to "".
+ val uuid = if (table.isChildDataMap) {
+ Option(operationContext.getProperty("uuid")).getOrElse("").toString
--- End diff --
In case of isChildDataMap condition it is mandatory to get UUID set by
parent table. If not set then log the message here or better to throw the
exception from here
---
