Github user setjet commented on a diff in the pull request:
https://github.com/apache/spark/pull/18113#discussion_r119175369
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/aggregate/typedaggregators.scala
---
@@ -95,7 +93,123 @@ class TypedAverage[IN](val f: IN => Double) extends
Aggregator[IN, (Double, Long
// Java api support
def this(f: MapFunction[IN, java.lang.Double]) = this(x =>
f.call(x).asInstanceOf[Double])
+
def toColumnJava: TypedColumn[IN, java.lang.Double] = {
toColumn.asInstanceOf[TypedColumn[IN, java.lang.Double]]
}
}
+
+class TypedMinDouble[IN](val f: IN => Double)
+ extends Aggregator[IN, java.lang.Double, java.lang.Double] {
+
+ override def zero: java.lang.Double = null
--- End diff --
`TypedSum` is actually correct because it will return 0 in case of an empty
set ['In mathematics, an empty sum, or nullary sum, is a summation where the
number of terms is zero. By convention,[1] the value of any empty sum of
numbers is the additive identity,
zero.'](https://en.wikipedia.org/wiki/Empty_sum). One could therefore argue
that `Sum.scala` is actually wrong because it returns null:
`emptyTestData.agg(sum('key))`. We could either fix Sum.scala, although that
might affect existing applications, or align both to return null, even though
that is not technically correct.
The same does go for `TypedAvg`, which returns Double.Nan instead of null.
---
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]
