the-other-tim-brown commented on code in PR #13679:
URL: https://github.com/apache/hudi/pull/13679#discussion_r2257178886
##########
hudi-client/hudi-client-common/src/main/java/org/apache/hudi/io/FileGroupReaderBasedMergeHandle.java:
##########
@@ -289,4 +469,141 @@ private GenericRecord convertOutput(T record) {
return convertedRecord == null ? null :
readerContext.getRecordContext().convertToAvroRecord(convertedRecord,
requestedSchema.get());
}
}
+
+ private static class RecordLevelIndexCallback<T> implements
BaseFileUpdateCallback<T> {
+ private final Map<String, HoodieRecordUpdateState> recordUpdateStateMap =
new HashMap<>();
+
+ @Override
+ public void onUpdate(String recordKey, T previousRecord, T mergedRecord) {
+ this.recordUpdateStateMap.put(recordKey, HoodieRecordUpdateState.UPDATE);
Review Comment:
I spoke with Siva about this and we can get rid of the map by directly
updating the write status here similar to how the secondary index updates are
handled. We can also further optimize to avoid converting to HoodieRecord for
the secondary index update since that wrapper is just used for field access and
we can use the RecordContext for fetching the fields.
--
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]
