lilei1128 commented on code in PR #10415:
URL: https://github.com/apache/paimon/pull/10415#discussion_r4215398474
##########
paimon-core/src/main/java/org/apache/paimon/operation/DataEvolutionSplitRead.java:
##########
@@ -767,6 +755,21 @@ private FileRecordReader<InternalRow> createFileReader(
}
}
+ FileReadTarget readTarget =
+ readTarget(file, dataFilePathFactory, rowRanges,
fileIndexResult);
+ String formatIdentifier = readTarget.formatIdentifier;
+ FormatReaderMapping formatReaderMapping =
+ singleFileReaderMappings.computeIfAbsent(
+ new SingleFileKey(
+ schemaId,
+ formatIdentifier,
+ file.writeCols(),
+ readRowType,
+ nestedFieldEnabled),
+ key ->
+ formatBuilder(readRowType, fileFilters,
nestedFieldEnabled)
+ .build(formatIdentifier, schema,
dataSchema));
Review Comment:
Thanks for identifying this regression. For this PR, I narrowed the scope so
row-sidecar selection requires non-empty `rowRanges`, preserving the previous
behavior for ordinary File Index/DV reads. The final File Index/DV-aware
selection is still used for row-sidecar decisions when row ranges are
available. I added regression coverage and both now return the correct results
through `executeFilter()`.
I will handle `rowRanges == null` in a follow-up PR by fixing the physical
row-sidecar schema mapping first, then re-enabling bitmap-driven row-sidecar
selection for that 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]
