Github user gatorsmile commented on a diff in the pull request:
https://github.com/apache/spark/pull/21608#discussion_r205539125
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/command/CommandUtils.scala
---
@@ -47,15 +48,23 @@ object CommandUtils extends Logging {
}
}
- def calculateTotalSize(sessionState: SessionState, catalogTable:
CatalogTable): BigInt = {
+ def calculateTotalSize(spark: SparkSession, catalogTable: CatalogTable):
BigInt = {
+ val sessionState = spark.sessionState
if (catalogTable.partitionColumnNames.isEmpty) {
calculateLocationSize(sessionState, catalogTable.identifier,
catalogTable.storage.locationUri)
} else {
// Calculate table size as a sum of the visible partitions. See
SPARK-21079
val partitions =
sessionState.catalog.listPartitions(catalogTable.identifier)
- partitions.map { p =>
- calculateLocationSize(sessionState, catalogTable.identifier,
p.storage.locationUri)
- }.sum
+ val paths = partitions.map(x => new Path(x.storage.locationUri.get))
+ val stagingDir =
sessionState.conf.getConfString("hive.exec.stagingdir", ".hive-staging")
+ val pathFilter = new PathFilter with Serializable {
+ override def accept(path: Path): Boolean = {
+ DataSourceUtils.isDataPath(path) &&
!path.getName.startsWith(stagingDir)
--- End diff --
This is a behavior change. Could you document the changes in the migration
guide?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
