refactor: Code refactoring & simplification for Google Cloud Enterprise Search retriever (#8369)

Followup to https://github.com/langchain-ai/langchain/pull/7857

- Changes `_convert_search_response()` to use object attributes instead
of converting to dictionary
- Simplifies logic for readability

libs/langchain/langchain/retrievers/google_cloud_enterprise_search.py

@@ -106,34 +106,23 @@ class GoogleCloudEnterpriseSearchRetriever(BaseRetriever):
         self, results: Sequence[SearchResult]
     ) -> List[Document]:
         """Converts a sequence of search results to a list of LangChain documents."""
-        from google.protobuf.json_format import MessageToDict
+        documents: List[Document] = []
 
-        documents = []
         for result in results:
-            document_dict = MessageToDict(result.document._pb)
+            derived_struct_data = result.document.derived_struct_data
-            derived_struct_data = document_dict.get("derivedStructData", None)
+            doc_metadata = result.document.struct_data
-            if derived_struct_data:
+            doc_metadata.source = derived_struct_data.link or ""
-                doc_metadata = document_dict.get("structData", {})
+            doc_metadata.id = result.document.id
-                chunk_type = (
+
-                    "extractive_answers"
+            for chunk in (
-                    if self.get_extractive_answers
+                derived_struct_data.extractive_answers
-                    else "extractive_segments"
+                or derived_struct_data.extractive_segments
+            ):
+                if hasattr(chunk, "page_number"):
+                    doc_metadata.source += f":{chunk.page_number}"
+                documents.append(
+                    Document(page_content=chunk.content, metadata=doc_metadata)
                 )
-                for chunk in derived_struct_data.get(chunk_type, []):
-                    if chunk_type == "extractive_answers":
-                        doc_metadata["source"] = (
-                            f"{derived_struct_data.get('link', '')}"
-                            f":{chunk.get('pageNumber', '')}"
-                        )
-                    else:
-                        doc_metadata[
-                            "source"
-                        ] = f"{derived_struct_data.get('link', '')}"
-                    doc_metadata["id"] = document_dict["id"]
-                    document = Document(
-                        page_content=chunk.get("content", ""), metadata=doc_metadata
-                    )
-                    documents.append(document)
 
         return documents
 
@@ -162,7 +151,7 @@ class GoogleCloudEnterpriseSearchRetriever(BaseRetriever):
             extractive_content_spec=extractive_content_spec,
         )
 
-        request = SearchRequest(
+        return SearchRequest(
             query=query,
             filter=self.filter,
             serving_config=self._serving_config,
@@ -171,8 +160,6 @@ class GoogleCloudEnterpriseSearchRetriever(BaseRetriever):
             query_expansion_spec=query_expansion_spec,
         )
 
-        return request
-
     def _get_relevant_documents(
         self, query: str, *, run_manager: CallbackManagerForRetrieverRun
     ) -> List[Document]: