Posted by
GitBox on
Mar 09, 2021; 10:23am
URL: http://apache-carbondata-dev-mailing-list-archive.168.s1.nabble.com/GitHub-carbondata-vikramahuja1001-opened-a-new-pull-request-4072-WIP-Clean-files-phase2-tp105322p106604.html
vikramahuja1001 commented on a change in pull request #4072:
URL:
https://github.com/apache/carbondata/pull/4072#discussion_r590185386##########
File path: integration/spark/src/main/scala/org/apache/carbondata/trash/DataTrashManager.scala
##########
@@ -112,13 +141,91 @@ object DataTrashManager {
carbonTable: CarbonTable,
isForceDelete: Boolean,
cleanStaleInProgress: Boolean,
- partitionSpecsOption: Option[Seq[PartitionSpec]]): Unit = {
+ partitionSpecsOption: Option[Seq[PartitionSpec]]): Seq[Long] = {
val partitionSpecs = partitionSpecsOption.map(_.asJava).orNull
- SegmentStatusManager.deleteLoadsAndUpdateMetadata(carbonTable,
+ val sizeStatistics = SegmentStatusManager.deleteLoadsAndUpdateMetadata(carbonTable,
isForceDelete, partitionSpecs, cleanStaleInProgress, true)
if (carbonTable.isHivePartitionTable && partitionSpecsOption.isDefined) {
SegmentFileStore.cleanSegments(carbonTable, partitionSpecs, isForceDelete)
}
+ sizeStatistics
+ }
+
+ private def dryRunOnExpiredSegments(
+ carbonTable: CarbonTable,
+ isForceDelete: Boolean,
+ cleanStaleInProgress: Boolean,
+ partitionSpecsOption: Option[Seq[PartitionSpec]]): Seq[Long] = {
+ var sizeFreed: Long = 0
+ var trashSizeRemaining: Long = 0
+ val loadMetadataDetails = SegmentStatusManager.readLoadMetadata(carbonTable.getMetadataPath)
+ if (SegmentStatusManager.isLoadDeletionRequired(loadMetadataDetails)) {
+ loadMetadataDetails.foreach { oneLoad =>
+ val segmentFilePath = CarbonTablePath.getSegmentFilePath(carbonTable.getTablePath,
+ oneLoad.getSegmentFile)
+ if (DeleteLoadFolders.canDeleteThisLoad(oneLoad, isForceDelete, cleanStaleInProgress)) {
+ // No need to consider physical data for external segments, only consider metadata.
+ if (oneLoad.getPath() == null || oneLoad.getPath().equalsIgnoreCase("NA")) {
+ if (!carbonTable.isHivePartitionTable) {
+ sizeFreed += FileFactory.getDirectorySize(CarbonTablePath.getSegmentPath(carbonTable
+ .getTablePath, oneLoad.getLoadName))
+ } else {
+ sizeFreed += partitionTableSegmentSize(carbonTable, oneLoad, loadMetadataDetails,
+ partitionSpecsOption)
+ }
+ }
+ sizeFreed += FileFactory.getCarbonFile(segmentFilePath).getSize
+ } else {
+ if (SegmentStatusManager.isExpiredSegment(oneLoad, carbonTable
+ .getAbsoluteTableIdentifier)) {
+ if (!carbonTable.isHivePartitionTable) {
+ trashSizeRemaining += FileFactory.getDirectorySize(CarbonTablePath.getSegmentPath(
+ carbonTable.getTablePath, oneLoad.getLoadName))
+ } else {
+ trashSizeRemaining += partitionTableSegmentSize(carbonTable, oneLoad,
+ loadMetadataDetails, partitionSpecsOption)
+ }
+ trashSizeRemaining += FileFactory.getCarbonFile(segmentFilePath).getSize
+ }
+ }
+ }
+ }
+ Seq(sizeFreed, trashSizeRemaining)
+ }
+
+ def partitionTableSegmentSize( carbonTable: CarbonTable, oneLoad: LoadMetadataDetails,
+ loadMetadataDetails: Array[LoadMetadataDetails], partitionSpecsOption:
+ Option[Seq[PartitionSpec]]) : Long = {
+ var segmentSize: Long = 0
Review comment:
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.
For queries about this service, please contact Infrastructure at:
[hidden email]