[GitHub] [carbondata] vikramahuja1001 commented on a change in pull request #4072: [CARBONDATA-4110] Support clean files dry run operation and show statistics after clean files operation

[GitBox]

vikramahuja1001 commented on a change in pull request #4072:
URL: https://github.com/apache/carbondata/pull/4072#discussion_r590192598



##########
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(

Review comment:
       changed logic




----------------------------------------------------------------
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:
us...@infra.apache.org