sunchao commented on a change in pull request #35657:
URL: https://github.com/apache/spark/pull/35657#discussion_r840815754
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala
##########
@@ -68,16 +70,36 @@ case class BatchScanExec(
val newPartitions = scan.toBatch.planInputPartitions()
originalPartitioning match {
- case p: DataSourcePartitioning if p.numPartitions !=
newPartitions.size =>
- throw new SparkException(
- "Data source must have preserved the original partitioning during
runtime filtering; " +
- s"reported num partitions: ${p.numPartitions}, " +
- s"num partitions after runtime filtering: ${newPartitions.size}")
+ case p: KeyGroupedPartitioning =>
+ if (newPartitions.exists(!_.isInstanceOf[HasPartitionKey])) {
+ throw new SparkException("Data source must have preserved the
original partitioning " +
+ "during runtime filtering: not all partitions implement
HasPartitionKey after " +
+ "filtering")
+ }
+
+ val newRows = new InternalRowSet(p.expressions.map(_.dataType))
+ newRows ++=
newPartitions.map(_.asInstanceOf[HasPartitionKey].partitionKey())
+ val oldRows = p.partitionValuesOpt.get
+
+ if (oldRows.size != newRows.size) {
Review comment:
yea we can - I think my original intention here was to provide different
error messages based on whether a) the number of partitions changed, or b) the
number of partitions are the same but partition values have changed.
--
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]
