RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.
Go agent-harnessagent-native-databaseagentic-aiagentic-retrievalagentic-searchaiai-agentscontext-enginecontext-engineeringcontext-managementharness-engineeringknowledge-compilationragretrieval-augmented-generationretrieval-harnesssearch-harness
Fix: add doc_type_kwd back (#17658)
Wang Qia1a79ab98cf4
diff --git a/api/apps/restful_apis/chunk_api.py b/api/apps/restful_apis/chunk_api.pyindex 049ed60e04..e5a78cdf96 100644--- a/api/apps/restful_apis/chunk_api.py+++ b/api/apps/restful_apis/chunk_api.py@@ -105,6 +105,7 @@ class Chunk(BaseModel):questions: list = Field(default_factory=list)question_tks: str = ""image_id: str = ""+ doc_type_kwd: str = ""available: bool = Truepositions: list[list[int]] = Field(default_factory=list)@@ -489,6 +490,7 @@ async def list_chunks(tenant_id, dataset_id, document_id):"questions": chunk.get("question_kwd", []),"dataset_id": chunk.get("kb_id", chunk.get("dataset_id")),"image_id": chunk.get("img_id", ""),+ "doc_type_kwd": chunk.get("doc_type_kwd") if isinstance(chunk.get("doc_type_kwd"), str) else "text","available": bool(chunk.get("available_int", 1)),"positions": chunk.get("position_int", []),"tag_kwd": chunk.get("tag_kwd", []),@@ -516,6 +518,7 @@ async def list_chunks(tenant_id, dataset_id, document_id):"questions": sres.field[chunk_id].get("question_kwd", []),"dataset_id": sres.field[chunk_id].get("kb_id", sres.field[chunk_id].get("dataset_id")),"image_id": sres.field[chunk_id].get("img_id", ""),+ "doc_type_kwd": sres.field[chunk_id].get("doc_type_kwd") if isinstance(sres.field[chunk_id].get("doc_type_kwd"), str) else "text","available": bool(int(sres.field[chunk_id].get("available_int", "1"))),"positions": sres.field[chunk_id].get("position_int", []),}@@ -969,6 +972,7 @@ async def add_chunk(tenant_id, dataset_id, document_id):"id": chunk_id,"content_ltks": rag_tokenizer.tokenize(req["content"]),"content_with_weight": req["content"],+ "doc_type_kwd": "text",}d["content_sm_ltks"] = rag_tokenizer.fine_grained_tokenize(d["content_ltks"])d["important_kwd"] = req.get("important_keywords", [])@@ -1024,6 +1028,7 @@ async def add_chunk(tenant_id, dataset_id, document_id):"create_time": "create_time","document_keyword": "document","img_id": "image_id",+ "doc_type_kwd": "doc_type_kwd",}renamed_chunk = {new_key: d[key] for key, new_key in key_mapping.items() if key in d}_ = Chunk(**renamed_chunk)diff --git a/test/testcases/test_web_api/test_chunk_app/test_chunk_routes_unit.py b/test/testcases/test_web_api/test_chunk_app/test_chunk_routes_unit.pyindex 5a2f2aee4b..3708c76d43 100644--- a/test/testcases/test_web_api/test_chunk_app/test_chunk_routes_unit.py+++ b/test/testcases/test_web_api/test_chunk_app/test_chunk_routes_unit.py@@ -521,6 +521,13 @@ def test_restful_chunk_list_get_and_delete_unit(monkeypatch):assert res["data"]["total"] == 1, resassert res["data"]["chunks"][0]["id"] == "chunk-1", resassert res["data"]["chunks"][0]["available"] is True, res+ assert res["data"]["chunks"][0]["doc_type_kwd"] == "text", res++ module.request = SimpleNamespace(args={"id": "chunk-1"}, headers={})+ module.settings.docStoreConn.chunk["doc_type_kwd"] = "image"+ res = _run(_route_core(module.list_chunks)("tenant-1", "kb-1", "doc-1"))+ assert res["code"] == 0, res+ assert res["data"]["chunks"][0]["doc_type_kwd"] == "image", resres = _run(_route_core(module.get_chunk)("tenant-1", "kb-1", "doc-1", "chunk-1"))assert res["code"] == 0, res@@ -715,3 +722,4 @@ def test_restful_add_chunk_valid_image_base64_stores_before_insert(monkeypatch):inserted = module.settings.docStoreConn.inserted[-1]assert inserted.get("img_id"), insertedassert inserted.get("doc_type_kwd") == "image", inserted+ assert res["data"]["chunk"]["doc_type_kwd"] == "image", res