nsivabalan commented on code in PR #9666:
URL: https://github.com/apache/hudi/pull/9666#discussion_r1321801446
##########
hudi-common/src/main/java/org/apache/hudi/common/table/timeline/TimelineUtils.java:
##########
@@ -339,6 +343,31 @@ public static void
validateTimestampAsOf(HoodieTableMetaClient metaClient, Strin
timestampAsOf, incompleteCommitTime));
}
}
+
+ // also timestamp as of cannot query cleaned up data.
+ Option<HoodieInstant> latestCleanOpt =
metaClient.getActiveTimeline().getCleanerTimeline().filterCompletedInstants().lastInstant();
+ if (latestCleanOpt.isPresent()) {
+ // Ensure timestamp as of is > than the earliest commit to retain and
+ try {
+ HoodieCleanMetadata cleanMetadata =
CleanerUtils.getCleanerMetadata(metaClient, latestCleanOpt.get());
+ String earliestCommitToRetain =
cleanMetadata.getEarliestCommitToRetain();
+ if (!StringUtils.isNullOrEmpty(earliestCommitToRetain)) {
+
ValidationUtils.checkArgument(HoodieTimeline.compareTimestamps(earliestCommitToRetain,
LESSER_THAN_OR_EQUALS, timestampAsOf),
+ "Cleaner cleaned up the timestamp of interest. Please ensure
sufficient commits are retained with cleaner "
+ + "for Timestamp as of query to work");
+ } else {
+ // when cleaner is based on file versions, we may not find value for
earliestCommitToRetain. so, lets check for archival
+ Option<HoodieInstant> firstCompletedInstant =
metaClient.getActiveTimeline().getWriteTimeline().filterCompletedInstants().firstInstant();
+ if (firstCompletedInstant.isPresent()) {
+
ValidationUtils.checkArgument(HoodieTimeline.compareTimestamps(firstCompletedInstant.get().getTimestamp(),
LESSER_THAN_OR_EQUALS, timestampAsOf),
+ "Please ensure sufficient commits are retained (uncleaned and
un-archived) for timestamp as of query to work.");
Review Comment:
yes, we are ensuring that firstInstant <= timestampAsOf.
if not, we fail.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
