hvanhovell commented on code in PR #40997:
URL: https://github.com/apache/spark/pull/40997#discussion_r1244508167
##########
connector/connect/client/jvm/src/main/scala/org/apache/spark/sql/connect/client/SparkResult.scala:
##########
@@ -50,15 +50,33 @@ private[sql] class SparkResult[T](
private val idxToBatches = mutable.Map.empty[Int, ColumnarBatch]
private def createEncoder(schema: StructType): ExpressionEncoder[T] = {
- val agnosticEncoder = if (encoder == UnboundRowEncoder) {
- // Create a row encoder based on the schema.
- RowEncoder.encoderFor(schema).asInstanceOf[AgnosticEncoder[T]]
- } else {
- encoder
- }
+ val agnosticEncoder = createEncoder(encoder,
schema).asInstanceOf[AgnosticEncoder[T]]
ExpressionEncoder(agnosticEncoder)
}
+ /**
+ * Update RowEncoder and recursively update the fields of the ProductEncoder
if found.
+ */
+ private def createEncoder[_](
+ enc: AgnosticEncoder[_],
+ dataType: DataType): AgnosticEncoder[_] = {
+ enc match {
+ case UnboundRowEncoder =>
+ // Replace the row encoder with the encoder inferred from the schema.
+ RowEncoder.encoderFor(dataType.asInstanceOf[StructType])
+ case ProductEncoder(clsTag, fields) if ProductEncoder.isTuple(clsTag) =>
+ // Recursively continue updating the tuple product encoder
+ val schema = dataType.asInstanceOf[StructType]
+ assert(fields.length == schema.fields.length)
Review Comment:
I don't think this needs to hold. I think `fields.length <=
schema.fields.length` needs to hold.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
