zhenlineo commented on code in PR #41129:
URL: https://github.com/apache/spark/pull/41129#discussion_r1198057053
##########
connector/connect/server/src/main/scala/org/apache/spark/sql/connect/planner/SparkConnectPlanner.scala:
##########
@@ -2386,10 +2393,26 @@ class SparkConnectPlanner(val session: SparkSession) {
}
if (writeOp.hasForeachWriter) {
- val foreach = writeOp.getForeachWriter.getPythonWriter
- val pythonFcn = transformPythonFunction(foreach)
- writer.foreachImplementation(
- new PythonForeachWriter(pythonFcn,
dataset.schema).asInstanceOf[ForeachWriter[Any]])
+ if (writeOp.getForeachWriter.hasPythonWriter) {
+ val foreach = writeOp.getForeachWriter.getPythonWriter
+ val pythonFcn = transformPythonFunction(foreach)
+ writer.foreachImplementation(
+ new PythonForeachWriter(pythonFcn,
dataset.schema).asInstanceOf[ForeachWriter[Any]])
+ } else {
+ val foreachWriterPkt =
unpackForeachWriter(writeOp.getForeachWriter.getScalaWriter)
+ val clientWriter = foreachWriterPkt.foreachWriter
+ if (foreachWriterPkt.rowEncoder == null) {
+ // rowEncoder is null means the client-side writer has type
parameter Row,
+ // Since server-side dataset is always dataframe, here just use
foreach directly.
+ writer.foreach(clientWriter.asInstanceOf[ForeachWriter[Row]])
+ } else {
+ val encoder = ExpressionEncoder(
+ foreachWriterPkt.rowEncoder.asInstanceOf[AgnosticEncoder[Any]])
Review Comment:
It is okay for now.
--
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]
