Create separate SourceLookup instance per segment slice in Significan…

…tTextAggregatorFactory

Signed-off-by: Jay Deng <[email protected]>

CHANGELOG.md

@@ -90,6 +90,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 - Make Span exporter configurable ([#8620](https://github.com/opensearch-project/OpenSearch/issues/8620))
 - Change InternalSignificantTerms to sum shard-level superset counts only in final reduce ([#8735](https://github.com/opensearch-project/OpenSearch/pull/8735))
 - Exclude 'benchmarks' from codecov report ([#8805](https://github.com/opensearch-project/OpenSearch/pull/8805))
+- Create separate SourceLookup instance per segment slice in SignificantTextAggregatorFactory ([#8807](https://github.com/opensearch-project/OpenSearch/pull/8807))
 
 ### Deprecated
 
@@ -100,4 +101,4 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 ### Security
 
 [Unreleased 3.0]: https://github.com/opensearch-project/OpenSearch/compare/2.x...HEAD
-[Unreleased 2.x]: https://github.com/opensearch-project/OpenSearch/compare/2.8...2.x
+[Unreleased 2.x]: https://github.com/opensearch-project/OpenSearch/compare/2.8...2.x

server/src/main/java/org/opensearch/search/aggregations/bucket/terms/SignificantTextAggregatorFactory.java

@@ -148,7 +148,9 @@ protected Aggregator createInternal(
             : includeExclude.convertToStringFilter(DocValueFormat.RAW, maxRegexLength);
 
         MapStringTermsAggregator.CollectorSource collectorSource = new SignificantTextCollectorSource(
-            queryShardContext.lookup().source(),
+            // Create a new SourceLookup instance per aggregator instead of use the shared one from SearchLookup. This is fine because it
+            // will only be accessed by this Aggregator instance and not anywhere else.
+            new SourceLookup(),
             queryShardContext.bigArrays(),
             fieldType,
             sourceFieldNames,