maropu commented on a change in pull request #31955:
URL: https://github.com/apache/spark/pull/31955#discussion_r601445222
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/ScalaUDF.scala
##########
@@ -124,10 +124,10 @@ case class ScalaUDF(
val toRow = enc.createSerializer().asInstanceOf[Any => Any]
if (enc.isSerializedAsStructForTopLevel) {
value: Any =>
- if (value == null) null else toRow(value).asInstanceOf[InternalRow]
+ if (value == null) null else
toRow(value).asInstanceOf[InternalRow].copy()
} else {
value: Any =>
- if (value == null) null else
toRow(value).asInstanceOf[InternalRow].get(0, dataType)
+ if (value == null) null else
toRow(value).asInstanceOf[InternalRow].copy().get(0, dataType)
Review comment:
If so, it looks like this is a bug of the higher order function. Why
don't we copy it in the higher order function side?
For example,
```
---
a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/higherOrderFunctions.scala
+++
b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/higherOrderFunctions.scala
@@ -279,7 +279,8 @@ case class ArrayTransform(
if (indexVar.isDefined) {
indexVar.get.value.set(i)
}
- result.update(i, f.eval(inputRow))
+ val v = InternalRow.copyValue(f.eval(inputRow))
+ result.update(i, v)
i += 1
}
result
```
I think the current fix can cause performance regression in case of
`ScalaUDF + non higher order functions`.
--
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]
