wwj6591812 commented on code in PR #9181:
URL: https://github.com/apache/paimon/pull/9181#discussion_r3766461761
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/sink/FlinkSinkBuilder.java:
##########
@@ -225,7 +225,9 @@ public DataStreamSink<?> build() {
table.rowType(),
readerFactoryForDescriptor,
table.coreOptions().blobWriteNullOnMissingFile(),
- table.coreOptions().blobWriteNullOnFetchFailure());
+ table.coreOptions().blobWriteNullOnFetchFailure(),
+ materializedBlobFieldIndexes(
Review Comment:
Thanks, fixed in `2c76ffa90`. Primary-key tables now retain the existing
descriptor preflight by passing an empty materialized-BLOB set to
`FlinkRowWrapper`; append-only tables keep the single-GET optimization. I also
added `testPrimaryKeyWriteHttpNotFoundWithMissingFileRetainsPreflight`, which
covers HTTP 404 -> NULL and asserts the retained HEAD + range-GET sequence.
The focused unit/append-only HTTP suite passes 19/19 with Checkstyle and
Spotless. On this local JDK 8 environment the new primary-key integration path
is blocked before reaching BLOB handling by the existing
CodeGenerator/ServiceLoader failure; I reproduced the identical failure with
the unchanged `testPrimaryKeyArrayBlobField` baseline test.
--
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]
