[GitHub] [spark] amaliujia commented on a diff in pull request #38468: [SPARK-41005][CONNECT][PYTHON] Arrow-based collect

[GitBox <gi...@apache.org>]

amaliujia commented on code in PR #38468:
URL: https://github.com/apache/spark/pull/38468#discussion_r1015819772


##########
connector/connect/src/main/scala/org/apache/spark/sql/connect/service/SparkConnectStreamHandler.scala:
##########
@@ -117,10 +129,91 @@ class SparkConnectStreamHandler(responseObserver: StreamObserver[Response]) exte
       responseObserver.onNext(response.build())
     }
 
-    responseObserver.onNext(sendMetricsToResponse(clientId, rows))
+    responseObserver.onNext(sendMetricsToResponse(clientId, dataframe))
     responseObserver.onCompleted()
   }
 
+  def processRowsAsArrowBatches(clientId: String, dataframe: DataFrame): Unit = {
+    val spark = dataframe.sparkSession
+    val schema = dataframe.schema
+    // TODO: control the batch size instead of max records
+    val maxRecordsPerBatch = spark.sessionState.conf.arrowMaxRecordsPerBatch
+    val timeZoneId = spark.sessionState.conf.sessionLocalTimeZone
+
+    SQLExecution.withNewExecutionId(dataframe.queryExecution, Some("collectArrow")) {
+      val pool = ThreadUtils.newDaemonSingleThreadExecutor("connect-collect-arrow")
+      val tasks = collection.mutable.ArrayBuffer.empty[Future[_]]
+      val rows = dataframe.queryExecution.executedPlan.execute()
+
+      if (rows.getNumPartitions > 0) {
+        val batches = rows.mapPartitionsInternal { iter =>
+          ArrowConverters
+            .toArrowBatchIterator(iter, schema, maxRecordsPerBatch, timeZoneId)
+        }
+
+        val processPartition = (iter: Iterator[(Array[Byte], Long, Long)]) => iter.toArray
+
+        val resultHandler = (partitionId: Int, taskResult: Array[(Array[Byte], Long, Long)]) => {

Review Comment:
   @cloud-fan and I had a discussion on this. I remember our initial thought was to maintain the partition ordering on the server side. For example, there is `def foreachPartition` API already. The question was whether we can foreach partitions in the partitioning order.



-- 
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-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org