srowen commented on a change in pull request #24779: [SPARK-27929][SQL] Make
percentile function receive frq of double
URL: https://github.com/apache/spark/pull/24779#discussion_r313561101
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/aggregate/Percentile.scala
##########
@@ -67,16 +67,23 @@ case class Percentile(
child: Expression,
percentageExpression: Expression,
frequencyExpression : Expression,
+ isIntFreqExpression: Expression,
Review comment:
Does Hive support this? I didn't see it supports weighted percentile.
You're also changing the implementation here, right... is it possible that
is causing the test to fail without two branches of the logic? I don't think it
makes sense to have the caller tell the type of the column; you already know it
in the implementation. But I'm wondering how the same calculation can come out
differently if it just allows decimal types too. Is it a rounding thing?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
