Github user kumarvishal09 commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/1704#discussion_r165369872
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/hive/CarbonAnalysisRules.scala
---
@@ -206,6 +208,30 @@ case class CarbonIUDAnalysisRule(sparkSession:
SparkSession) extends Rule[Logica
table)
}
}
+
+
+ /**
+ * This method will check whether the data type of new value is
supported or not
+ * For example, arrayType is not supported
+ */
+ def checkForUnsupportedType(selectStmt: String): Unit = {
+ // selectStmt will have "select 'newValue' from table"
+ // we can validate the new value by executing the select query
+ val sel = selectStmt.substring(0, selectStmt.lastIndexOf("from")).trim;
--- End diff --
Instead of getting from the data set , can we get a logical plan and check
the data type
---
