From 1c8b42489d38f7ff12ca26376178444e105cb343 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 6 Jun 2025 13:06:22 +0100
Subject: [PATCH] Fix NPE in semantic highlighter (#128989)

This PR fixes an NPE in the semantic highlighter when the document being highlighted doesn't contain value for the semantic text field.

Closes #128975
---
 docs/changelog/128989.yaml                    |  6 +++
 .../highlight/SemanticTextHighlighter.java    |  3 ++
 .../SemanticTextHighlighterTests.java         | 40 +++++++++++++++++--
 3 files changed, 46 insertions(+), 3 deletions(-)
 create mode 100644 docs/changelog/128989.yaml

diff --git a/docs/changelog/128989.yaml b/docs/changelog/128989.yaml
new file mode 100644
index 0000000000000..94ee4c6923d7b
--- /dev/null
+++ b/docs/changelog/128989.yaml
@@ -0,0 +1,6 @@
+pr: 128989
+summary: Fix NPE in semantic highlighter
+area: Search
+type: bug
+issues:
+ - 128975
diff --git a/x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighter.java b/x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighter.java
index 1ad205cd6aad6..844267244263e 100644
--- a/x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighter.java
+++ b/x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighter.java
@@ -209,6 +209,9 @@ private List<OffsetAndScore> extractOffsetAndScores(
         leafQueries.stream().forEach(q -> bq.add(q, BooleanClause.Occur.SHOULD));
         Weight weight = new IndexSearcher(reader).createWeight(bq.build(), ScoreMode.COMPLETE, 1);
         Scorer scorer = weight.scorer(reader.getContext());
+        if (scorer == null) {
+            return List.of();
+        }
         if (previousParent != -1) {
             if (scorer.iterator().advance(previousParent) == DocIdSetIterator.NO_MORE_DOCS) {
                 return List.of();
diff --git a/x-pack/plugin/inference/src/test/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighterTests.java b/x-pack/plugin/inference/src/test/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighterTests.java
index 4b459c25556e2..9612719e0da65 100644
--- a/x-pack/plugin/inference/src/test/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighterTests.java
+++ b/x-pack/plugin/inference/src/test/java/org/elasticsearch/xpack/inference/highlight/SemanticTextHighlighterTests.java
@@ -32,6 +32,8 @@
 import org.elasticsearch.index.mapper.MapperService;
 import org.elasticsearch.index.mapper.MapperServiceTestCase;
 import org.elasticsearch.index.mapper.SourceToParse;
+import org.elasticsearch.index.query.BoolQueryBuilder;
+import org.elasticsearch.index.query.MatchAllQueryBuilder;
 import org.elasticsearch.index.query.NestedQueryBuilder;
 import org.elasticsearch.index.query.QueryBuilder;
 import org.elasticsearch.index.shard.ShardId;
@@ -170,6 +172,34 @@ public void testSparseVector() throws Exception {
         );
     }
 
+    @SuppressWarnings("unchecked")
+    public void testNoSemanticField() throws Exception {
+        var mapperService = createDefaultMapperService(useLegacyFormat);
+        Map<String, Object> queryMap = (Map<String, Object>) queries.get("sparse_vector_1");
+        List<WeightedToken> tokens = readSparseVector(queryMap.get("embeddings"));
+        var fieldType = (SemanticTextFieldMapper.SemanticTextFieldType) mapperService.mappingLookup().getFieldType(SEMANTIC_FIELD_ELSER);
+        SparseVectorQueryBuilder sparseQuery = new SparseVectorQueryBuilder(
+            fieldType.getEmbeddingsField().fullPath(),
+            tokens,
+            null,
+            null,
+            null,
+            null
+        );
+        var query = new BoolQueryBuilder().should(sparseQuery).should(new MatchAllQueryBuilder());
+        var shardRequest = createShardSearchRequest(query);
+        var sourceToParse = new SourceToParse("0", new BytesArray("{}"), XContentType.JSON);
+        assertHighlightOneDoc(
+            mapperService,
+            shardRequest,
+            sourceToParse,
+            SEMANTIC_FIELD_ELSER,
+            10,
+            HighlightBuilder.Order.SCORE,
+            new String[0]
+        );
+    }
+
     private MapperService createDefaultMapperService(boolean useLegacyFormat) throws IOException {
         var mappings = Streams.readFully(SemanticTextHighlighterTests.class.getResourceAsStream("mappings.json"));
         var settings = Settings.builder()
@@ -264,9 +294,13 @@ private void assertHighlightOneDoc(
                             new HashMap<>()
                         );
                         var result = highlighter.highlight(context);
-                        assertThat(result.fragments().length, equalTo(expectedPassages.length));
-                        for (int i = 0; i < result.fragments().length; i++) {
-                            assertThat(result.fragments()[i].string(), equalTo(expectedPassages[i]));
+                        if (result == null) {
+                            assertThat(expectedPassages.length, equalTo(0));
+                        } else {
+                            assertThat(result.fragments().length, equalTo(expectedPassages.length));
+                            for (int i = 0; i < result.fragments().length; i++) {
+                                assertThat(result.fragments()[i].string(), equalTo(expectedPassages[i]));
+                            }
                         }
                     }
                 } finally {