Add exists query to semantic_text

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/mapper/SemanticTextFieldMapper.java

@@ -7,12 +7,15 @@
 
 package org.elasticsearch.xpack.inference.mapper;
 
+import org.apache.lucene.search.MatchNoDocsQuery;
 import org.apache.lucene.search.Query;
 import org.apache.lucene.search.join.BitSetProducer;
+import org.apache.lucene.search.join.QueryBitSetProducer;
 import org.apache.lucene.search.join.ScoreMode;
 import org.elasticsearch.cluster.metadata.InferenceFieldMetadata;
 import org.elasticsearch.common.Explicit;
 import org.elasticsearch.common.Strings;
+import org.elasticsearch.common.lucene.search.Queries;
 import org.elasticsearch.common.xcontent.XContentHelper;
 import org.elasticsearch.common.xcontent.support.XContentMapValues;
 import org.elasticsearch.core.Nullable;
@@ -42,6 +45,7 @@
 import org.elasticsearch.index.query.QueryBuilder;
 import org.elasticsearch.index.query.QueryBuilders;
 import org.elasticsearch.index.query.SearchExecutionContext;
+import org.elasticsearch.index.search.ESToParentBlockJoinQuery;
 import org.elasticsearch.inference.InferenceResults;
 import org.elasticsearch.inference.SimilarityMeasure;
 import org.elasticsearch.search.vectors.KnnVectorQueryBuilder;
@@ -352,6 +356,23 @@ public Query termQuery(Object value, SearchExecutionContext context) {
             throw new IllegalArgumentException(CONTENT_TYPE + " fields do not support term query");
         }
 
+        @Override
+        public Query existsQuery(SearchExecutionContext context) {
+            if (getEmbeddingsField() == null) {
+                return new MatchNoDocsQuery();
+            }
+
+            // Do the equivalent of a nested query with an exists query for the embeddings field
+            String nestedFieldPath = getChunksFieldName(name());
+            BitSetProducer parentFilter = new QueryBitSetProducer(Queries.newNonNestedFilter(IndexVersion.current()));
+            return new ESToParentBlockJoinQuery(
+                getEmbeddingsField().fieldType().existsQuery(context),
+                parentFilter,
+                ScoreMode.None,
+                nestedFieldPath
+            );
+        }
+
         @Override
         public ValueFetcher valueFetcher(SearchExecutionContext context, String format) {
             // Redirect the fetcher to load the original values of the field

x-pack/plugin/inference/src/yamlRestTest/resources/rest-api-spec/test/inference/70_semantic_text_exists_query.yml

@@ -0,0 +1,144 @@
+setup:
+  - requires:
+      cluster_features: "gte_v8.15.0"
+      reason: semantic_text introduced in 8.15.0
+
+  - do:
+      inference.put:
+        task_type: sparse_embedding
+        inference_id: sparse-inference-id
+        body: >
+          {
+            "service": "test_service",
+            "service_settings": {
+              "model": "my_model",
+              "api_key": "abc64"
+            },
+            "task_settings": {
+            }
+          }
+
+  - do:
+      inference.put:
+        task_type: text_embedding
+        inference_id: dense-inference-id
+        body: >
+          {
+            "service": "text_embedding_test_service",
+            "service_settings": {
+              "model": "my_model",
+              "dimensions": 10,
+              "api_key": "abc64",
+              "similarity": "COSINE"
+            },
+            "task_settings": {
+            }
+          }
+
+  - do:
+      indices.create:
+        index: test-sparse-index
+        body:
+          mappings:
+            properties:
+              inference_field:
+                type: semantic_text
+                inference_id: sparse-inference-id
+
+  - do:
+      indices.create:
+        index: test-dense-index
+        body:
+          mappings:
+            properties:
+              inference_field:
+                type: semantic_text
+                inference_id: dense-inference-id
+
+---
+"Exists query with no indexed documents":
+  - do:
+      search:
+        index: test-sparse-index
+        body:
+          query:
+            exists:
+              field: "inference_field"
+
+  - match: { hits.total.value: 0 }
+
+---
+"Exists query with null indexed documents":
+  - do:
+      index:
+        index: test-sparse-index
+        id: doc
+        body:
+          inference_field: null
+        refresh: true
+
+  - do:
+      search:
+        index: test-sparse-index
+        body:
+          query:
+            exists:
+              field: "inference_field"
+
+  - match: { hits.total.value: 0 }
+
+  - do:
+      index:
+        index: test-dense-index
+        id: doc
+        body:
+          inference_field: null
+        refresh: true
+
+  - do:
+      search:
+        index: test-dense-index
+        body:
+          query:
+            exists:
+              field: "inference_field"
+
+  - match: { hits.total.value: 0 }
+
+---
+"Exists query with indexed documents":
+  - do:
+      index:
+        index: test-sparse-index
+        id: doc
+        body:
+          inference_field: "hello world"
+        refresh: true
+
+  - do:
+      search:
+        index: test-sparse-index
+        body:
+          query:
+            exists:
+              field: "inference_field"
+
+  - match: { hits.total.value: 1 }
+
+  - do:
+      index:
+        index: test-dense-index
+        id: doc
+        body:
+          inference_field: "hello world"
+        refresh: true
+
+  - do:
+      search:
+        index: test-dense-index
+        body:
+          query:
+            exists:
+              field: "inference_field"
+
+  - match: { hits.total.value: 1 }