LuciferYang commented on code in PR #38865:
URL: https://github.com/apache/spark/pull/38865#discussion_r1043270635
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/collectionOperations.scala:
##########
@@ -4600,3 +4600,92 @@ case class ArrayExcept(left: Expression, right:
Expression) extends ArrayBinaryL
override protected def withNewChildrenInternal(
newLeft: Expression, newRight: Expression): ArrayExcept = copy(left =
newLeft, right = newRight)
}
+
+/**
+ * Given an array, and another element append the element at the end of the
array.
+ */
+@ExpressionDescription(
+ usage = "_FUNC_(array, element) - Append the element",
+ examples =
+ """
+ Examples:
+ > SELECT _FUNC_(array('b', 'd', 'c', 'a'), 'd');
+ ["b","d","c","a","d"]
+
+ """,
+ since = "3.4.0",
+ group = "array_funcs")
+case class ArrayAppend(left: Expression, right: Expression)
+ extends BinaryExpression
+ with ImplicitCastInputTypes
+ with ComplexTypeMergingExpression
+ with QueryErrorsBase {
+ override def prettyName: String = "array_append"
+
+ override def inputTypes: Seq[AbstractDataType] = {
+ (left.dataType, right.dataType) match {
+ case (ArrayType(e1, hasNull), e2) =>
+ TypeCoercion.findTightestCommonType(e1, e2) match {
+ case Some(dt) => Seq(ArrayType(dt, hasNull), dt)
+ case _ => Seq.empty
+ }
+ case _ => Seq.empty
+ }
+ }
+
+ override def checkInputDataTypes(): TypeCheckResult = {
+ (left.dataType, right.dataType) match {
+ case (ArrayType(e1, _), e2) if e1.sameType(e2) =>
TypeCheckResult.TypeCheckSuccess
+ case (ArrayType(e1, _), e2) => DataTypeMismatch(
+ errorSubClass = "ARRAY_FUNCTION_DIFF_TYPES",
+ messageParameters = Map(
+ "functionName" -> toSQLId(prettyName),
+ "leftType" -> toSQLType(left.dataType),
+ "rightType" -> toSQLType(right.dataType),
+ "dataType" -> toSQLType(ArrayType)
+ ))
+ case _ =>
+ DataTypeMismatch(
+ errorSubClass = "UNEXPECTED_INPUT_TYPE",
+ messageParameters = Map(
+ "paramIndex" -> "0",
+ "requiredType" -> toSQLType(ArrayType),
+ "inputSql" -> toSQLExpr(left),
+ "inputType" -> toSQLType(left.dataType)
+ )
+ )
+ }
+ }
+
+ protected def withNewChildrenInternal(newLeft: Expression, newRight:
Expression): ArrayAppend =
+ copy(left = newLeft, right = newRight)
+
+ override protected def nullSafeEval(input1: Any, input2: Any): Any = {
+ val arrayData = input1.asInstanceOf[ArrayData]
+ val arrayElementType = dataType.asInstanceOf[ArrayType].elementType
+ val elementData = input2
+ val numberOfElements = arrayData.numElements() + 1
+ if (numberOfElements > ByteArrayMethods.MAX_ROUNDED_ARRAY_LENGTH) {
+ throw
QueryExecutionErrors.concatArraysWithElementsExceedLimitError(numberOfElements)
+ }
+ val finalData = new Array[Any](numberOfElements)
+ arrayData.foreach(arrayElementType, finalData.update)
+ finalData.update(numberOfElements - 1, elementData)
+ new GenericArrayData(finalData)
+ }
+
+ override def doGenCode(ctx: CodegenContext, ev: ExprCode): ExprCode = {
+ nullSafeCodeGen(
+ ctx, ev, (left: String, right: String) => {
+ val expr = ctx.addReferenceObj("arraysAppendExpr", this)
+ s"${ev.value} = (ArrayData)$expr.nullSafeEval($left, $right);"
Review Comment:
@beliefer I found that the prs of these 4 new functions all choice this
implementation way. What are the advantages of implement `doGenCode` by
codegen over the current implementation way? Thanks !!!
--
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.
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
