ueshin commented on a change in pull request #25728: [SPARK-29020][WIP][SQL]
Improving array_sort behaviour
URL: https://github.com/apache/spark/pull/25728#discussion_r336976058
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/higherOrderFunctions.scala
##########
@@ -285,6 +287,110 @@ case class ArrayTransform(
override def prettyName: String = "transform"
}
+/**
+ * Sorts elements in an array using a comparator function.
+ */
+@ExpressionDescription(
+ usage = "_FUNC_(expr, func) - Sorts and returns the array based on the given
" +
+ "comparator function. The comparator will take two nullable arguments " +
+ "representing two nullable elements of the array." +
+ "It returns -1, 0, or 1 as the first nullable element is less than, equal
to, or greater" +
+ "than the second nullable element. Null elements will be placed at the end
of the returned" +
+ "array. If the comparator function returns other values (including NULL),"
+
+ "the query will fail and raise an error. By the default it will sort the
array in " +
+ "ascending mode",
+ examples = """
+ Examples:
+ > SELECT _FUNC_(array(5, 6, 1), (x,y) -> f(x,y));
+ [1,5,6]
+ > SELECT _FUNC_(array('bc', 'ab', 'dc'), (x, y) -> f(x,y));
+ ['dc', 'bc', 'ab']
+ """,
+ since = "3.0.0")
+case class ArraySort(
+ argument: Expression,
+ function: Expression)
+ extends ArrayBasedSimpleHigherOrderFunction with ArraySortLike with
CodegenFallback {
+
+ def this(argument: Expression) = this(argument, Literal(true))
+
+ @transient lazy val argumentsType: DataType =
+ argument.dataType.asInstanceOf[ArrayType].elementType
+
+ override protected def arrayExpression: Expression = argument
+
+ override protected def nullOrder: NullOrder = NullOrder.Greatest
+
+ override def dataType: ArrayType = ArrayType(argumentsType,
argument.nullable)
+
+ override def checkInputDataTypes(): TypeCheckResult = {
+
+ if (function.dataType == BooleanType) {
Review comment:
Any reply?
I assume that we won't support `array_sort(column, order)` signature as per
discussion above?
----------------------------------------------------------------
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]
