advancedxy commented on a change in pull request #25863:
[WIP][SPARK-29037][CORE][SQL] For static partition overwrite, spark may give
duplicate result.
URL: https://github.com/apache/spark/pull/25863#discussion_r327992313
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/InsertIntoHadoopFsRelationCommand.scala
##########
@@ -263,4 +281,108 @@ case class InsertIntoHadoopFsRelationCommand(
}
}.toMap
}
+
+ /**
+ * Detect the conflict when there are several InsertIntoHadoopFsRelation
operations
+ * write concurrently.
+ */
+ private def detectConflict(
+ fs: FileSystem,
+ path: Path,
+ isOverwrite: Boolean,
+ staticPartitionKVs: Seq[(String, String)]): Unit = {
+ for (i <- 0 until partitionColumns.size) {
+ val checkedStagingPath = ListBuffer(s".spark-staging-overwrite-$i")
+ if (isOverwrite) {
+ checkedStagingPath += s".spark-staging-append-$i"
Review comment:
what about two concurrent job both appending data to the same partitions?
In current approach, we didn't detect the conflict? I think we have to
disable concurrent append to the same partition also.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
