viirya commented on a change in pull request #28764:
URL: https://github.com/apache/spark/pull/28764#discussion_r438470608
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/mathExpressions.scala
##########
@@ -1325,3 +1325,94 @@ case class BRound(child: Expression, scale: Expression)
with Serializable with ImplicitCastInputTypes {
def this(child: Expression) = this(child, Literal(0))
}
+
+object WidthBucket {
+
+ def computeBucketNumber(value: Double, min: Double, max: Double, numBucket:
Long): jl.Long = {
+ if (numBucket <= 0 || numBucket == Long.MaxValue || jl.Double.isNaN(value)
|| min == max ||
+ jl.Double.isNaN(min) || jl.Double.isInfinite(min) ||
+ jl.Double.isNaN(max) || jl.Double.isInfinite(max)) {
+ return null
+ }
+
+ val lower = Math.min(min, max)
+ val upper = Math.max(min, max)
+
+ if (min < max) {
+ if (value < lower) {
+ 0L
+ } else if (value >= upper) {
+ numBucket + 1L
+ } else {
+ (numBucket.toDouble * (value - lower) / (upper - lower)).toLong + 1L
+ }
+ } else { // `min > max` case
Review comment:
Is the behavior of min > max the same across all DBs implementing
`WidthBucket`?
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
