beliefer commented on code in PR #43629:
URL: https://github.com/apache/spark/pull/43629#discussion_r1387522474
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/command/AnalyzePartitionCommand.scala:
##########
@@ -101,56 +98,19 @@ case class AnalyzePartitionCommand(
if (noscan) {
Map.empty
} else {
- calculateRowCountsPerPartition(sparkSession, tableMeta,
partitionValueSpec)
+ CommandUtils.calculateRowCountsPerPartition(sparkSession, tableMeta,
partitionValueSpec)
}
// Update the metastore if newly computed statistics are different from
those
// recorded in the metastore.
-
- val sizes = CommandUtils.calculateMultipleLocationSizes(sparkSession,
tableMeta.identifier,
- partitions.map(_.storage.locationUri))
- val newPartitions = partitions.zipWithIndex.flatMap { case (p, idx) =>
- val newRowCount = rowCounts.get(p.spec)
- val newStats = CommandUtils.compareAndGetNewStats(p.stats, sizes(idx),
newRowCount)
- newStats.map(_ => p.copy(stats = newStats))
- }
-
+ val (_, newPartitions) = CommandUtils.calculatePartitionStats(
+ sparkSession, tableMeta, partitions, Some(rowCounts))
if (newPartitions.nonEmpty) {
sessionState.catalog.alterPartitions(tableMeta.identifier, newPartitions)
}
Seq.empty[Row]
}
- private def calculateRowCountsPerPartition(
Review Comment:
I think we shouldn't move `calculateRowCountsPerPartition` if it is used
only once.
--
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]
