Github user ericl commented on a diff in the pull request:
https://github.com/apache/spark/pull/15814#discussion_r87113460
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/DataSourceStrategy.scala
---
@@ -182,41 +182,53 @@ case class DataSourceAnalysis(conf: CatalystConf)
extends Rule[LogicalPlan] {
"Cannot overwrite a path that is also being read from.")
}
- val overwritingSinglePartition =
- overwrite.specificPartition.isDefined &&
+ val partitionSchema = query.resolve(
+ t.partitionSchema, t.sparkSession.sessionState.analyzer.resolver)
+ val partitionsTrackedByCatalog =
t.sparkSession.sessionState.conf.manageFilesourcePartitions &&
+ l.catalogTable.isDefined &&
l.catalogTable.get.partitionColumnNames.nonEmpty &&
l.catalogTable.get.tracksPartitionsInCatalog
- val effectiveOutputPath = if (overwritingSinglePartition) {
- val partition = t.sparkSession.sessionState.catalog.getPartition(
- l.catalogTable.get.identifier, overwrite.specificPartition.get)
- new Path(partition.storage.locationUri.get)
- } else {
- outputPath
- }
-
- val effectivePartitionSchema = if (overwritingSinglePartition) {
- Nil
- } else {
- query.resolve(t.partitionSchema,
t.sparkSession.sessionState.analyzer.resolver)
+ var initialMatchingPartitions: Seq[TablePartitionSpec] = Nil
+ var customPartitionLocations: Map[TablePartitionSpec, String] =
Map.empty
+
+ // When partitions are tracked by the catalog, compute all custom
partition locations that
+ // may be relevant to the insertion job.
+ if (partitionsTrackedByCatalog) {
+ val matchingPartitions =
t.sparkSession.sessionState.catalog.listPartitions(
--- End diff --
You also need the set of matching partitions (including those with default
locations) in order to determine which ones to delete at the end of an
overwrite call.
This makes the optimization quite messy, so I'd rather not push it to the
catalog for now.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
