Fixes 14217: database retrieve api and chat-messages api response doc…

…_metadata (#14219)
langgenius · Feb 27, 2025 · 1e3197a · 1e3197a
1 parent 5f692df
commit 1e3197a
Show file tree

Hide file tree

Showing 6 changed files with 6 additions and 0 deletions.
diff --git a/api/core/rag/retrieval/dataset_retrieval.py b/api/core/rag/retrieval/dataset_retrieval.py
@@ -203,6 +203,7 @@ def retrieve(
                                 "segment_id": segment.id,
                                 "retriever_from": invoke_from.to_source(),
                                 "score": record.score or 0.0,
+                                "doc_metadata": document.doc_metadata,
                             }
 
                             if invoke_from.to_source() == "dev":

diff --git a/api/core/tools/utils/dataset_retriever/dataset_multi_retriever_tool.py b/api/core/tools/utils/dataset_retriever/dataset_multi_retriever_tool.py
@@ -123,6 +123,7 @@ def _run(self, query: str) -> str:
                             "segment_id": segment.id,
                             "retriever_from": self.retriever_from,
                             "score": document_score_list.get(segment.index_node_id, None),
+                            "doc_metadata": document.doc_metadata,
                         }
 
                         if self.retriever_from == "dev":

diff --git a/api/core/tools/utils/dataset_retriever/dataset_retriever_tool.py b/api/core/tools/utils/dataset_retriever/dataset_retriever_tool.py
@@ -172,6 +172,7 @@ def _run(self, query: str) -> str:
                                     "segment_id": segment.id,
                                     "retriever_from": self.retriever_from,
                                     "score": record.score or 0.0,
+                                    "doc_metadata": document.doc_metadat,  # type: ignore
                                 }
 
                                 if self.retriever_from == "dev":

diff --git a/api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py b/api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py
@@ -240,6 +240,7 @@ def _fetch_dataset_retriever(self, node_data: KnowledgeRetrievalNodeData, query:
                                 "segment_word_count": segment.word_count,
                                 "segment_position": segment.position,
                                 "segment_index_node_hash": segment.index_node_hash,
+                                "doc_metadata": document.doc_metadata,
                             },
                             "title": document.name,
                         }

diff --git a/api/core/workflow/nodes/llm/node.py b/api/core/workflow/nodes/llm/node.py
@@ -459,6 +459,7 @@ def _convert_to_original_retriever_resource(self, context_dict: dict) -> Optiona
                 "index_node_hash": metadata.get("segment_index_node_hash"),
                 "content": context_dict.get("content"),
                 "page": metadata.get("page"),
+                "doc_metadata": metadata.get("doc_metadata"),
             }
 
             return source

diff --git a/api/fields/hit_testing_fields.py b/api/fields/hit_testing_fields.py
@@ -7,6 +7,7 @@
     "data_source_type": fields.String,
     "name": fields.String,
     "doc_type": fields.String,
+    "doc_metadata": fields.Raw,
 }
 
 segment_fields = {