chaoqin-li1123 commented on code in PR #45977:
URL: https://github.com/apache/spark/pull/45977#discussion_r1583663736
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/python/PythonMicroBatchStream.scala:
##########
@@ -35,6 +38,14 @@ class PythonMicroBatchStream(
ds.source.createPythonFunction(
ds.getOrCreateDataSourceInPython(shortName, options,
Some(outputSchema)).dataSource)
+ private val streamId = nextStreamId
+ private var nextBlockId = 0L
+
+ // planInputPartitions() maybe be called multiple times for the current
microbatch.
+ // Cache the result of planInputPartitions() because it may involve sending
data
+ // from python to JVM.
+ private var cachedInputPartition: Option[(String, String,
PythonStreamingInputPartition)] = None
Review Comment:
I added the overwrite that columnar read is not supported.
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/python/PythonMicroBatchStream.scala:
##########
@@ -44,9 +55,29 @@ class PythonMicroBatchStream(
override def latestOffset(): Offset =
PythonStreamingSourceOffset(runner.latestOffset())
override def planInputPartitions(start: Offset, end: Offset):
Array[InputPartition] = {
- runner.partitions(start.asInstanceOf[PythonStreamingSourceOffset].json,
- end.asInstanceOf[PythonStreamingSourceOffset].json)
- .zipWithIndex.map(p => PythonInputPartition(p._2, p._1))
+ val start_offset_json =
start.asInstanceOf[PythonStreamingSourceOffset].json
Review Comment:
Fixed.
--
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]
