[GitHub] BryanCutler commented on a change in pull request #23787: [SPARK-26830][SQL][R] Vectorized R dapply() implementation

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

BryanCutler commented on a change in pull request #23787: [SPARK-26830][SQL][R] Vectorized R dapply() implementation
URL: https://github.com/apache/spark/pull/23787#discussion_r259465179
 
 

 ##########
 File path: sql/core/src/main/scala/org/apache/spark/sql/execution/r/ArrowRRunner.scala
 ##########
 @@ -55,13 +56,32 @@ class ArrowRRunner(
     numPartitions = -1,
     isDataFrame = true,
     schema.fieldNames,
-    RRunnerModes.DATAFRAME_GAPPLY) {
+    mode) {
+
+  // TODO: it needs to refactor to share the same code with RRunner, and have separate
+  // ArrowRRunners.
+  private val getNextBatch = {
+    if (mode == RRunnerModes.DATAFRAME_GAPPLY) {
+      // gapply
+      (inputIterator: Iterator[_], keys: collection.mutable.ArrayBuffer[Array[Byte]]) => {
+        val (key, nextBatch) = inputIterator
+          .asInstanceOf[Iterator[(Array[Byte], Iterator[InternalRow])]].next()
+        keys.append(key)
+        nextBatch
+      }
+    } else {
+      // dapply
+      (inputIterator: Iterator[_], keys: collection.mutable.ArrayBuffer[Array[Byte]]) => {
+        inputIterator
+          .asInstanceOf[Iterator[Iterator[InternalRow]]].next()
+      }
+    }
+  }
 
   protected override def writeData(
       dataOut: DataOutputStream,
       printOut: PrintStream,
-      iter: Iterator[_]): Unit = if (iter.hasNext) {
-    val inputIterator = iter.asInstanceOf[Iterator[(Array[Byte], Iterator[InternalRow])]]
+      inputIterator: Iterator[_]): Unit = if (inputIterator.hasNext) {
 
 Review comment:
   Is it possible for `inputIterator.hasNext` to be false, like from an empty partition, and is there a test for this? It might be better to send an empty Arrow stream in this case, but I'm not entirely sure what the R worker is expecting.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

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