You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2020/06/10 22:57:53 UTC

[GitHub] [spark] viirya commented on a change in pull request #28764: [SPARK-21117][SQL] Built-in SQL Function Support - WIDTH_BUCKET

viirya commented on a change in pull request #28764:
URL: https://github.com/apache/spark/pull/28764#discussion_r438451085



##########
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 = {
+    // Checks error cases below:
+    //  - `numBucket` must be greater than zero and be less than Long.MaxValue
+    //  - `value`, `min`, and `max` cannot be NaN
+    //  - `min` bound cannot equal `max
+    //  - `min` and `max` must be finite
+    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
+      if (value > upper) {
+        0L
+      } else if (value <= lower) {
+        numBucket + 1L
+      } else {
+        (numBucket.toDouble * (upper - value) / (upper - lower)).toLong + 1L
+      }
+    }
+  }
+}
+
+/**
+ * Returns the bucket number into which the value of this expression would fall
+ * after being evaluated.
+ *

Review comment:
       Should we also list the error cases in the expression description?




----------------------------------------------------------------
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:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org