Github user hvanhovell commented on a diff in the pull request:
https://github.com/apache/spark/pull/17028#discussion_r102673129
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/aggregate/Percentile.scala
---
@@ -130,20 +130,30 @@ case class Percentile(
}
}
- override def createAggregationBuffer(): OpenHashMap[Number, Long] = {
+ private def toLongValue(d: Any): Long = d match {
+ case d: Decimal => d.toLong
+ case n: Number => n.longValue
+ }
+
+ private def toDoubleValue(d: Any): Double = d match {
+ case d: Decimal => d.toDouble
+ case n: Number => n.doubleValue
+ }
+
+ override def createAggregationBuffer(): OpenHashMap[AnyRef, Long] = {
// Initialize new counts map instance here.
- new OpenHashMap[Number, Long]()
+ new OpenHashMap[AnyRef, Long]()
}
override def update(
- buffer: OpenHashMap[Number, Long],
- input: InternalRow): OpenHashMap[Number, Long] = {
- val key = child.eval(input).asInstanceOf[Number]
+ buffer: OpenHashMap[AnyRef, Long],
+ input: InternalRow): OpenHashMap[AnyRef, Long] = {
+ val key = child.eval(input).asInstanceOf[AnyRef]
val frqValue = frequencyExpression.eval(input)
// Null values are ignored in counts map.
if (key != null && frqValue != null) {
- val frqLong = frqValue.asInstanceOf[Number].longValue()
+ val frqLong = toLongValue(frqValue)
--- End diff --
`frqValue` is guaranteed to return a integral value. So this is not needed.
We could also force it to be a Long, that would make this even simpler.
---
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]
