kazuyukitanimura commented on code in PR #407:
URL: https://github.com/apache/datafusion-comet/pull/407#discussion_r1602251565
##########
spark/src/main/scala/org/apache/spark/sql/comet/execution/shuffle/CometShuffleExchangeExec.scala:
##########
@@ -482,8 +483,10 @@ class CometShuffleWriteProcessor(
"elapsed_compute" -> metrics("shuffleReadElapsedCompute"))
val nativeMetrics = CometMetricNode(nativeSQLMetrics)
- val rawIter = cometRDD.iterator(partition, context)
- val cometIter = CometExec.getCometIterator(Seq(rawIter), nativePlan,
nativeMetrics)
+ val cometIter = CometExec.getCometIterator(
+ Seq(inputs.asInstanceOf[Iterator[ColumnarBatch]]),
Review Comment:
We have `ShimCometShuffleWriteProcessor.write()` for Spark 3.x
Yes, currently this may not be correct for Spark 4.0. This will be followup
with test fixes. I will ask for your help about how to work around the
interface change (or change Spark in the upstream).
--
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: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
