Github user ravipesala commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/2209#discussion_r184918157
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/util/SparkTypeConverter.scala
---
@@ -97,16 +97,30 @@ private[spark] object SparkTypeConverter {
def getStructChildren(table: CarbonTable, dimName: String): String = {
table.getChildren(dimName).asScala.map(childDim => {
childDim.getDataType.getName.toLowerCase match {
- case "array" => s"${
+ case "array" => if (table.isTransactionalTable) {s"${
--- End diff --
Could explain in comments why this many checks are required for non
transactional table?
Better write another method rather than keeping so many if checks
---
