the-other-tim-brown commented on code in PR #13444:
URL: https://github.com/apache/hudi/pull/13444#discussion_r2188566283
##########
hudi-common/src/main/java/org/apache/hudi/common/table/read/FileGroupRecordBuffer.java:
##########
@@ -565,27 +570,62 @@ protected boolean hasNextBaseRecord(T baseRecord,
BufferedRecord<T> logRecordInf
Pair<Boolean, T> isDeleteAndRecord = merge(baseRecordInfo,
logRecordInfo);
if (!isDeleteAndRecord.getLeft()) {
// Updates
- nextRecord = readerContext.seal(isDeleteAndRecord.getRight());
+ nextRecord =
readerContext.seal(applyOutputSchemaConversion(isDeleteAndRecord.getRight()));
+ // If the record is not the same as the base record, we can emit an
update
+ if (isDeleteAndRecord.getRight() != baseRecord) {
+ callbackOption.ifPresent(callback -> {
+ BufferedRecord<T> mergeResult =
BufferedRecord.forRecordWithContext(nextRecord,
readerContext.getSchemaHandler().getRequestedSchema(), readerContext,
orderingFieldName, false);
+
callback.onUpdate(readerContext.constructHoodieRecord(applyOutputSchemaConversion(baseRecordInfo)),
readerContext.constructHoodieRecord(applyOutputSchemaConversion(logRecordInfo)),
+
readerContext.constructHoodieRecord(applyOutputSchemaConversion(mergeResult)));
+ });
+ }
readStats.incrementNumUpdates();
return true;
- } else if (emitDelete) {
- // emit Deletes
- nextRecord = readerContext.getDeleteRow(isDeleteAndRecord.getRight(),
baseRecordInfo.getRecordKey());
- readStats.incrementNumDeletes();
- return nextRecord != null;
} else {
- // not emit Deletes
+ // emit Deletes
+ callbackOption.ifPresent(callback ->
callback.onDelete(readerContext.constructHoodieRecord(applyOutputSchemaConversion(baseRecordInfo)),
+
readerContext.constructHoodieRecord(applyOutputSchemaConversion(logRecordInfo))));
readStats.incrementNumDeletes();
- return false;
+ if (emitDelete) {
+ nextRecord =
applyOutputSchemaConversion(readerContext.getDeleteRow(isDeleteAndRecord.getRight(),
baseRecordInfo.getRecordKey()));
+ return nextRecord != null;
+ } else {
+ return false;
+ }
}
}
// Inserts
- nextRecord = readerContext.seal(baseRecord);
+ nextRecord = readerContext.seal(applyOutputSchemaConversion(baseRecord));
readStats.incrementNumInserts();
return true;
}
+ /**
+ * Applies the final output schema conversion to the buffered record if
required. This ensures the records match the requested schema.
+ * @param bufferedRecord the buffered record to convert
+ * @return a new buffered record with the converted record and the proper
schema ID set
+ */
+ protected BufferedRecord<T> applyOutputSchemaConversion(BufferedRecord<T>
bufferedRecord) {
+ if (bufferedRecord.getRecord() != null && outputConverter.isPresent()) {
+ return new BufferedRecord<>(bufferedRecord.getRecordKey(),
bufferedRecord.getOrderingValue(),
+ outputConverter.get().apply(bufferedRecord.getRecord()),
readerContext.encodeAvroSchema(readerContext.getSchemaHandler().getRequestedSchema()),
bufferedRecord.isDelete());
Review Comment:
Yes, will update
--
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]
