couchbase: Add document id to vector search results

libs/partners/couchbase/langchain_couchbase/vectorstores.py

@@ -559,12 +559,13 @@ def similarity_search_with_score_by_vector(
             # Parse the results
             for row in search_iter.rows():
                 text = row.fields.pop(self._text_key, "")
+                id = row.id
 
                 # Format the metadata from Couchbase
                 metadata = self._format_metadata(row.fields)
 
                 score = row.score
-                doc = Document(page_content=text, metadata=metadata)
+                doc = Document(id=id, page_content=text, metadata=metadata)
                 docs_with_score.append((doc, score))
 
         except Exception as e:

libs/partners/couchbase/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "langchain-couchbase"
-version = "0.2.0"
+version = "0.2.1"
 description = "An integration package connecting Couchbase and LangChain"
 authors = []
 readme = "README.md"

libs/partners/couchbase/tests/integration_tests/test_vector_store.py

@@ -193,6 +193,7 @@ def test_add_texts_with_ids_and_metadatas(self, cluster: Any) -> None:
         time.sleep(SLEEP_DURATION)
 
         output = vectorstore.similarity_search("foo", k=1)
+        assert output[0].id == "a"
         assert output[0].page_content == "foo"
         assert output[0].metadata["a"] == 1
 
@@ -364,3 +365,32 @@ def test_hybrid_search(self, cluster: Any) -> None:
 
         assert result == hybrid_result
         assert score <= hybrid_score
+
+    def test_id_in_results(self, cluster: Any) -> None:
+        """Test that the id is returned in the result documents."""
+
+        texts = [
+            "foo",
+            "bar",
+            "baz",
+        ]
+
+        metadatas = [{"a": 1}, {"b": 2}, {"c": 3}]
+
+        vectorstore = CouchbaseVectorStore(
+            cluster=cluster,
+            embedding=ConsistentFakeEmbeddings(),
+            index_name=INDEX_NAME,
+            bucket_name=BUCKET_NAME,
+            scope_name=SCOPE_NAME,
+            collection_name=COLLECTION_NAME,
+        )
+
+        ids = vectorstore.add_texts(texts, metadatas=metadatas)
+        assert len(ids) == len(texts)
+
+        # Wait for the documents to be indexed
+        time.sleep(SLEEP_DURATION)
+
+        output = vectorstore.similarity_search("foo", k=1)
+        assert output[0].id == ids[0]