Merge remote-tracking branch 'elastic/8.19' into backport/8.19/pr-130849

Author: julian-elastic

docs/changelog/131032.yaml

@@ -0,0 +1,5 @@
+pr: 131032
+summary: "Fix: `GET _synonyms` returns synonyms with empty rules"
+area: Relevance
+type: bug
+issues: []

docs/changelog/131081.yaml

@@ -0,0 +1,6 @@
+pr: 131081
+summary: Fix knn search error when dimensions are not set
+area: Vector Search
+type: bug
+issues:
+ - 129550

muted-tests.yml

@@ -441,3 +441,6 @@ tests:
   - class: org.elasticsearch.search.CCSDuelIT
     method: testTerminateAfter
     issue: https://github.com/elastic/elasticsearch/issues/126085
+  - class: org.elasticsearch.smoketest.DocsClientYamlTestSuiteIT
+    method: test {yaml=reference/search/retriever/line_906}
+    issue: https://github.com/elastic/elasticsearch/issues/131041

rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/40_knn_search.yml

@@ -621,3 +621,36 @@ setup:
           properties:
             embedding:
               type: dense_vector
+
+
+---
+"Searching with no data dimensions specified":
+  - requires:
+      cluster_features: "search.vectors.no_dimensions_bugfix"
+      reason: "Search with no dimensions bugfix"
+
+  - do:
+      indices.create:
+        index: empty-test
+        body:
+          mappings:
+            properties:
+              vector:
+                type: dense_vector
+                index: true
+
+  - do:
+      search:
+        index: empty-test
+        body:
+          fields: [ "name" ]
+          knn:
+            field: vector
+            query_vector: [ -0.5, 90.0, -10, 14.8, -156.0 ]
+            k: 3
+            num_candidates: 3
+            rescore_vector:
+              oversample: 1.5
+            similarity: 0.1
+
+  - match: { hits.total.value: 0 }

rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/synonyms/40_synonyms_sets_get.yml

@@ -178,3 +178,33 @@ teardown:
 
   - match:
       count: 12
+
+---
+"Return empty rule set":
+  - requires:
+      cluster_features: [ synonyms_set.get.return_empty_synonym_sets ]
+      reason: "synonyms_set get api return empty synonym sets"
+
+  - do:
+      synonyms.put_synonym:
+        id: empty-synonyms
+        body:
+          synonyms_set: []
+
+  - do:
+      synonyms.get_synonyms_sets: {}
+
+  - match:
+      count: 4
+
+  - match:
+      results:
+        - synonyms_set: "empty-synonyms"
+          count: 0
+        - synonyms_set: "test-synonyms-1"
+          count: 3
+        - synonyms_set: "test-synonyms-2"
+          count: 1
+        - synonyms_set: "test-synonyms-3"
+          count: 2
+

server/src/main/java/module-info.java

@@ -432,6 +432,7 @@
             org.elasticsearch.index.IndexFeatures,
             org.elasticsearch.ingest.IngestGeoIpFeatures,
             org.elasticsearch.search.SearchFeatures,
+            org.elasticsearch.synonyms.SynonymFeatures,
             org.elasticsearch.script.ScriptFeatures,
             org.elasticsearch.search.retriever.RetrieversFeatures,
             org.elasticsearch.reservedstate.service.FileSettingsFeatures,

server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java

@@ -30,6 +30,7 @@
 import org.apache.lucene.index.VectorEncoding;
 import org.apache.lucene.index.VectorSimilarityFunction;
 import org.apache.lucene.search.FieldExistsQuery;
+import org.apache.lucene.search.MatchNoDocsQuery;
 import org.apache.lucene.search.Query;
 import org.apache.lucene.search.join.BitSetProducer;
 import org.apache.lucene.util.BitUtil;
@@ -2216,6 +2217,9 @@ public Query createKnnQuery(
                     "to perform knn search on field [" + name() + "], its mapping must have [index] set to [true]"
                 );
             }
+            if (dims == null) {
+                return new MatchNoDocsQuery("No data has been indexed for field [" + name() + "]");
+            }
             return switch (getElementType()) {
                 case BYTE -> createKnnByteQuery(queryVector.asByteVector(), k, numCands, filter, similarityThreshold, parentFilter);
                 case FLOAT -> createKnnFloatQuery(

server/src/main/java/org/elasticsearch/search/SearchFeatures.java

@@ -25,9 +25,11 @@ public Set<NodeFeature> getFeatures() {
     public static final NodeFeature COMPLETION_FIELD_SUPPORTS_DUPLICATE_SUGGESTIONS = new NodeFeature(
         "search.completion_field.duplicate.support"
     );
+    public static final NodeFeature RESCORER_MISSING_FIELD_BAD_REQUEST = new NodeFeature("search.rescorer.missing.field.bad.request");
     public static final NodeFeature INT_SORT_FOR_INT_SHORT_BYTE_FIELDS = new NodeFeature("search.sort.int_sort_for_int_short_byte_fields");
     static final NodeFeature MULTI_MATCH_CHECKS_POSITIONS = new NodeFeature("search.multi.match.checks.positions");
     private static final NodeFeature KNN_QUERY_BUGFIX_130254 = new NodeFeature("search.knn.query.bugfix.130254", true);
+    public static final NodeFeature SEARCH_WITH_NO_DIMENSIONS_BUGFIX = new NodeFeature("search.vectors.no_dimensions_bugfix");
 
     @Override
     public Set<NodeFeature> getTestFeatures() {
@@ -36,7 +38,8 @@ public Set<NodeFeature> getTestFeatures() {
             COMPLETION_FIELD_SUPPORTS_DUPLICATE_SUGGESTIONS,
             INT_SORT_FOR_INT_SHORT_BYTE_FIELDS,
             MULTI_MATCH_CHECKS_POSITIONS,
-            KNN_QUERY_BUGFIX_130254
+            KNN_QUERY_BUGFIX_130254,
+            SEARCH_WITH_NO_DIMENSIONS_BUGFIX
         );
     }
 }

server/src/main/java/org/elasticsearch/synonyms/SynonymFeatures.java

@@ -0,0 +1,24 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the "Elastic License
+ * 2.0", the "GNU Affero General Public License v3.0 only", and the "Server Side
+ * Public License v 1"; you may not use this file except in compliance with, at
+ * your election, the "Elastic License 2.0", the "GNU Affero General Public
+ * License v3.0 only", or the "Server Side Public License, v 1".
+ */
+
+package org.elasticsearch.synonyms;
+
+import org.elasticsearch.features.FeatureSpecification;
+import org.elasticsearch.features.NodeFeature;
+
+import java.util.Set;
+
+public class SynonymFeatures implements FeatureSpecification {
+    private static final NodeFeature RETURN_EMPTY_SYNONYM_SETS = new NodeFeature("synonyms_set.get.return_empty_synonym_sets");
+
+    @Override
+    public Set<NodeFeature> getTestFeatures() {
+        return Set.of(RETURN_EMPTY_SYNONYM_SETS);
+    }
+}

server/src/main/java/org/elasticsearch/synonyms/SynonymsManagementAPIService.java

@@ -45,6 +45,9 @@
 import org.elasticsearch.indices.SystemIndexDescriptor;
 import org.elasticsearch.rest.RestStatus;
 import org.elasticsearch.search.aggregations.BucketOrder;
+import org.elasticsearch.search.aggregations.bucket.filter.Filters;
+import org.elasticsearch.search.aggregations.bucket.filter.FiltersAggregationBuilder;
+import org.elasticsearch.search.aggregations.bucket.filter.FiltersAggregator;
 import org.elasticsearch.search.aggregations.bucket.terms.Terms;
 import org.elasticsearch.search.aggregations.bucket.terms.TermsAggregationBuilder;
 import org.elasticsearch.search.builder.SearchSourceBuilder;
@@ -89,6 +92,8 @@ public class SynonymsManagementAPIService {
     private static final int MAX_SYNONYMS_SETS = 10_000;
     private static final String SYNONYM_RULE_ID_FIELD = SynonymRule.ID_FIELD.getPreferredName();
     private static final String SYNONYM_SETS_AGG_NAME = "synonym_sets_aggr";
+    private static final String RULE_COUNT_AGG_NAME = "rule_count";
+    private static final String RULE_COUNT_FILTER_KEY = "synonym_rules";
     private static final int SYNONYMS_INDEX_MAPPINGS_VERSION = 1;
     private final int maxSynonymsSets;
 
@@ -180,27 +185,45 @@ private static XContentBuilder mappings() {
         }
     }
 
+    /**
+     * Returns all synonym sets with their rule counts, including empty synonym sets.
+     * @param from The index of the first synonym set to return
+     * @param size The number of synonym sets to return
+     * @param listener The listener to return the synonym sets to
+     */
     public void getSynonymsSets(int from, int size, ActionListener<PagedResult<SynonymSetSummary>> listener) {
+        BoolQueryBuilder synonymSetQuery = QueryBuilders.boolQuery()
+            .should(QueryBuilders.termQuery(OBJECT_TYPE_FIELD, SYNONYM_SET_OBJECT_TYPE))
+            .should(QueryBuilders.termQuery(OBJECT_TYPE_FIELD, SYNONYM_RULE_OBJECT_TYPE))
+            .minimumShouldMatch(1);
+
+        // Aggregation query to count only synonym rules (excluding synonym set objects)
+        FiltersAggregationBuilder ruleCountAggregation = new FiltersAggregationBuilder(
+            RULE_COUNT_AGG_NAME,
+            new FiltersAggregator.KeyedFilter(RULE_COUNT_FILTER_KEY, QueryBuilders.termQuery(OBJECT_TYPE_FIELD, SYNONYM_RULE_OBJECT_TYPE))
+        );
+
         client.prepareSearch(SYNONYMS_ALIAS_NAME)
             .setSize(0)
             // Retrieves aggregated synonym rules for each synonym set, excluding the synonym set object type
-            .setQuery(QueryBuilders.termQuery(OBJECT_TYPE_FIELD, SYNONYM_RULE_OBJECT_TYPE))
+            .setQuery(synonymSetQuery)
             .addAggregation(
                 new TermsAggregationBuilder(SYNONYM_SETS_AGG_NAME).field(SYNONYMS_SET_FIELD)
                     .order(BucketOrder.key(true))
                     .size(maxSynonymsSets)
+                    .subAggregation(ruleCountAggregation)
             )
             .setPreference(Preference.LOCAL.type())
             .execute(new ActionListener<>() {
                 @Override
                 public void onResponse(SearchResponse searchResponse) {
                     Terms termsAggregation = searchResponse.getAggregations().get(SYNONYM_SETS_AGG_NAME);
                     List<? extends Terms.Bucket> buckets = termsAggregation.getBuckets();
-                    SynonymSetSummary[] synonymSetSummaries = buckets.stream()
-                        .skip(from)
-                        .limit(size)
-                        .map(bucket -> new SynonymSetSummary(bucket.getDocCount(), bucket.getKeyAsString()))
-                        .toArray(SynonymSetSummary[]::new);
+                    SynonymSetSummary[] synonymSetSummaries = buckets.stream().skip(from).limit(size).map(bucket -> {
+                        Filters ruleCountFilters = bucket.getAggregations().get(RULE_COUNT_AGG_NAME);
+                        Filters.Bucket ruleCountBucket = ruleCountFilters.getBucketByKey(RULE_COUNT_FILTER_KEY);
+                        return new SynonymSetSummary(ruleCountBucket.getDocCount(), bucket.getKeyAsString());
+                    }).toArray(SynonymSetSummary[]::new);
 
                     listener.onResponse(new PagedResult<>(buckets.size(), synonymSetSummaries));
                 }