the-other-tim-brown commented on code in PR #13830:
URL: https://github.com/apache/hudi/pull/13830#discussion_r2322692955


##########
hudi-client/hudi-client-common/src/main/java/org/apache/hudi/index/HoodieIndexUtils.java:
##########
@@ -619,9 +621,10 @@ public static <R> HoodieData<HoodieRecord<R>> 
tagGlobalLocationBackToRecords(
           Option<HoodieRecordGlobalLocation> currentLocOpt = 
Option.ofNullable(v.getRight().orElse(null));
           if (currentLocOpt.isPresent()) {
             HoodieRecordGlobalLocation currentLoc = currentLocOpt.get();
-            boolean shouldDoMergedLookUpThenTag = mayContainDuplicateLookup
-                || !Objects.equals(incomingRecord.getPartitionPath(), 
currentLoc.getPartitionPath());
-            if (shouldUpdatePartitionPath && shouldDoMergedLookUpThenTag) {
+            boolean shouldDoMergedLookUpThenTag = mayContainDuplicateLookup // 
handle event time ordering updates
+                || shouldUpdatePartitionPath && 
!Objects.equals(incomingRecord.getPartitionPath(), 
currentLoc.getPartitionPath())  // handle partition updates
+                || (!isCommitTimeOrdered && 
incomingRecord.isDelete(writerSchema.get(), properties)); // handle event time 
ordering deletes

Review Comment:
   @nsivabalan I had this special handling for CoW tables but @danny0405 
removed it. I agree that this is not necessary but Danny seems to disagree.
   
   Previously there were no tests for this and likely the deletes were not 
properly detected. This is the second time we've found regressions in this path.



-- 
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]

Reply via email to