Github user hvanhovell commented on a diff in the pull request:
https://github.com/apache/spark/pull/14842#discussion_r76875904
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/SortOrder.scala
---
@@ -28,14 +28,24 @@ abstract sealed class SortDirection {
def sql: String
}
+// default null order is firt for asc
case object Ascending extends SortDirection {
override def sql: String = "ASC"
}
+case object AscendingNullLast extends SortDirection {
--- End diff --
Combining `SortDirection` with `NullOrdering` seems to make life more
complicated then needed, why not keep null ordering a separate concept? And add
that to `SortOrder`, for example:
```scala
case class SortOrder(child: Expression, direction: SortDirection,
nullOrdering: NullOrdering) {
def this(child: Expression, direction: SortDirection): SortOrder =
this(child, direction, direction.DefaultNullOrdering)
...
}
abstract sealed class SortDirection {
def sql: String
def defaultNullOrdering: NullOrdering
}
case object Ascending extends SortDirection {
override def sql: String = "ASC"
override def defaultNullOrdering: NullOrdering = NullsFirst
}
case object Descending extends SortDirection {
override def sql: String = "DESC"
override def defaultNullOrdering: NullOrdering = NullsLast
}
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
