Github user ueshin commented on a diff in the pull request:
https://github.com/apache/spark/pull/21965#discussion_r207485560
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/higherOrderFunctions.scala
---
@@ -210,3 +219,54 @@ case class ArrayTransform(
override def prettyName: String = "transform"
}
+
+/**
+ * Filters the input array using the given lambda function.
+ */
+@ExpressionDescription(
+ usage = "_FUNC_(expr, func) - Filters the input array using the given
predicate.",
+ examples = """
+ Examples:
+ > SELECT _FUNC_(array(1, 2, 3), x -> x % 2 == 1);
+ array(1, 3)
+ """,
+ since = "2.4.0")
+case class ArrayFilter(
+ input: Expression,
+ function: Expression)
+ extends ArrayBasedHigherOrderFunction with CodegenFallback {
+
+ override def nullable: Boolean = input.nullable
+
+ override def dataType: DataType = input.dataType
+
+ override def expectingFunctionType: AbstractDataType = BooleanType
+
+ override def bind(f: (Expression, Seq[(DataType, Boolean)]) =>
LambdaFunction): ArrayFilter = {
+ val elem =
ArrayBasedHigherOrderFunction.elementArgumentType(input.dataType)
+ copy(function = f(function, elem :: Nil))
+ }
+
+ @transient lazy val LambdaFunction(_, Seq(elementVar:
NamedLambdaVariable), _) = function
+
+ override def eval(input: InternalRow): Any = {
+ val arr = this.input.eval(input).asInstanceOf[ArrayData]
+ if (arr == null) {
+ null
+ } else {
+ val f = functionForEval
+ val buffer = new mutable.ArrayBuffer[Any]
--- End diff --
Sounds good. I'll use the original length as a size hint.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
