thecoop commented on code in PR #16696:
URL: https://github.com/apache/lucene/pull/16696#discussion_r4061798972
##########
lucene/core/src/java/org/apache/lucene/codecs/lucene104/Lucene104ScalarQuantizedVectorsWriter.java:
##########
@@ -432,21 +459,172 @@ private void mergeOneFlatVectorFieldCentered(FieldInfo
fieldInfo, MergeState mer
segmentWriteState.infoStream.message(
QUANTIZED_VECTOR_COMPONENT, "Vectors' count:" + vectorCount);
}
- QuantizedByteVectorValues quantizedVectorValues =
- mergedQuantizedVectorValues(fieldInfo, mergeState, mergedCentroid);
+ // Only asymmetric encodings have query-side records, and only FLOAT32
fields score graphs with
+ // them. FLOAT16 fields use fp16 vectors. The predicate sees vectorCount
before deletions, so it
+ // can request data even when the merged field is too small to build a
graph.
+ boolean prepareQueryData =
+ needsMergeScorer != null
Review Comment:
Rather than nullable, the non-scorer case should pass `i -> false` for the
lambda here, then it doesn't need to check for null at all
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
