Fix(api): correct default value handling in dataset parser config (#7589)

### What problem does this PR solve?

Fix  HTTP API Create/Update dataset parser config default value error

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

api/apps/sdk/dataset.py

@@ -344,8 +344,11 @@ def update(tenant_id, dataset_id):
     if req.get("parser_config"):
         req["parser_config"] = deep_merge(kb.parser_config, req["parser_config"])
 
-    if (chunk_method := req.get("parser_id")) and chunk_method != kb.parser_id and req.get("parser_config") is None:
-        req["parser_config"] = get_parser_config(chunk_method, None)
+    if (chunk_method := req.get("parser_id")) and chunk_method != kb.parser_id:
+        if not req.get("parser_config"):
+            req["parser_config"] = get_parser_config(chunk_method, None)
+    elif "parser_config" in req and not req["parser_config"]:
+        del req["parser_config"]
 
     if "name" in req and req["name"].lower() != kb.name.lower():
         try:

api/utils/validation_utils.py

@@ -218,7 +218,7 @@ class CreateDatasetReq(Base):
     permission: Annotated[PermissionEnum, StringConstraints(strip_whitespace=True, min_length=1, max_length=16), Field(default=PermissionEnum.me)]
     chunk_method: Annotated[ChunkMethodnEnum, StringConstraints(strip_whitespace=True, min_length=1, max_length=32), Field(default=ChunkMethodnEnum.naive, serialization_alias="parser_id")]
     pagerank: int = Field(default=0, ge=0, le=100)
-    parser_config: ParserConfig = Field(default_factory=dict)
+    parser_config: ParserConfig | None = Field(default=None)
 
     @field_validator("avatar")
     @classmethod
@@ -330,15 +330,42 @@ class CreateDatasetReq(Base):
         """
         return v.lower() if isinstance(v, str) else v
 
+    @field_validator("parser_config", mode="before")
+    @classmethod
+    def normalize_empty_parser_config(cls, v: Any) -> Any:
+        """
+        Normalizes empty parser configuration by converting empty dictionaries to None.
+
+        This validator ensures consistent handling of empty parser configurations across
+        the application by converting empty dicts to None values.
+
+        Args:
+            v (Any): Raw input value for the parser config field
+
+        Returns:
+            Any: Returns None if input is an empty dict, otherwise returns the original value
+
+        Example:
+            >>> normalize_empty_parser_config({})
+            None
+
+            >>> normalize_empty_parser_config({"key": "value"})
+            {"key": "value"}
+        """
+        if v == {}:
+            return None
+        return v
+
     @field_validator("parser_config", mode="after")
     @classmethod
-    def validate_parser_config_json_length(cls, v: ParserConfig) -> ParserConfig:
+    def validate_parser_config_json_length(cls, v: ParserConfig | None) -> ParserConfig | None:
         """
         Validates serialized JSON length constraints for parser configuration.
 
         Implements a two-stage validation workflow:
-        1. Model serialization - convert Pydantic model to JSON string
-        2. Size verification - enforce maximum allowed payload size
+        1. Null check - bypass validation for empty configurations
+        2. Model serialization - convert Pydantic model to JSON string
+        3. Size verification - enforce maximum allowed payload size
 
         Args:
             v (ParserConfig | None): Raw parser configuration object
@@ -349,6 +376,9 @@ class CreateDatasetReq(Base):
         Raises:
             ValueError: When serialized JSON exceeds 65,535 characters
         """
+        if v is None:
+            return None
+
         if (json_str := v.model_dump_json()) and len(json_str) > 65535:
             raise ValueError(f"Parser config exceeds size limit (max 65,535 characters). Current size: {len(json_str):,}")
         return v

sdk/python/test/test_http_api/test_dataset_mangement/test_create_dataset.py

@@ -141,7 +141,7 @@ class TestDatasetCreate:
     def test_avatar(self, get_http_api_auth, tmp_path):
         fn = create_image_file(tmp_path / "ragflow_test.png")
         payload = {
-            "name": "avatar_test",
+            "name": "avatar",
             "avatar": f"data:image/png;base64,{encode_avatar(fn)}",
         }
         res = create_dataset(get_http_api_auth, payload)
@@ -149,14 +149,14 @@ class TestDatasetCreate:
 
     @pytest.mark.p2
     def test_avatar_exceeds_limit_length(self, get_http_api_auth):
-        payload = {"name": "exceeds_limit_length_avatar", "avatar": "a" * 65536}
+        payload = {"name": "avatar_exceeds_limit_length", "avatar": "a" * 65536}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 101, res
         assert "String should have at most 65535 characters" in res["message"], res
 
     @pytest.mark.p3
     @pytest.mark.parametrize(
-        "name, avatar_prefix, expected_message",
+        "name, prefix, expected_message",
         [
             ("empty_prefix", "", "Missing MIME prefix. Expected format: data:<mime>;base64,<data>"),
             ("missing_comma", "data:image/png;base64", "Missing MIME prefix. Expected format: data:<mime>;base64,<data>"),
@@ -165,11 +165,11 @@ class TestDatasetCreate:
         ],
         ids=["empty_prefix", "missing_comma", "unsupported_mine_type", "invalid_mine_type"],
     )
-    def test_avatar_invalid_prefix(self, get_http_api_auth, tmp_path, name, avatar_prefix, expected_message):
+    def test_avatar_invalid_prefix(self, get_http_api_auth, tmp_path, name, prefix, expected_message):
         fn = create_image_file(tmp_path / "ragflow_test.png")
         payload = {
             "name": name,
-            "avatar": f"{avatar_prefix}{encode_avatar(fn)}",
+            "avatar": f"{prefix}{encode_avatar(fn)}",
         }
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 101, res
@@ -177,42 +177,42 @@ class TestDatasetCreate:
 
     @pytest.mark.p3
     def test_avatar_unset(self, get_http_api_auth):
-        payload = {"name": "test_avatar_unset"}
+        payload = {"name": "avatar_unset"}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["avatar"] is None, res
 
     @pytest.mark.p3
     def test_avatar_none(self, get_http_api_auth):
-        payload = {"name": "test_avatar_none", "avatar": None}
+        payload = {"name": "avatar_none", "avatar": None}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["avatar"] is None, res
 
     @pytest.mark.p2
     def test_description(self, get_http_api_auth):
-        payload = {"name": "test_description", "description": "description"}
+        payload = {"name": "description", "description": "description"}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["description"] == "description", res
 
     @pytest.mark.p2
     def test_description_exceeds_limit_length(self, get_http_api_auth):
-        payload = {"name": "exceeds_limit_length_description", "description": "a" * 65536}
+        payload = {"name": "description_exceeds_limit_length", "description": "a" * 65536}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 101, res
         assert "String should have at most 65535 characters" in res["message"], res
 
     @pytest.mark.p3
     def test_description_unset(self, get_http_api_auth):
-        payload = {"name": "test_description_unset"}
+        payload = {"name": "description_unset"}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["description"] is None, res
 
     @pytest.mark.p3
     def test_description_none(self, get_http_api_auth):
-        payload = {"name": "test_description_none", "description": None}
+        payload = {"name": "description_none", "description": None}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["description"] is None, res
@@ -283,7 +283,7 @@ class TestDatasetCreate:
 
     @pytest.mark.p2
     def test_embedding_model_none(self, get_http_api_auth):
-        payload = {"name": "test_embedding_model_none", "embedding_model": None}
+        payload = {"name": "embedding_model_none", "embedding_model": None}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 101, res
         assert "Input should be a valid string" in res["message"], res
@@ -323,14 +323,14 @@ class TestDatasetCreate:
 
     @pytest.mark.p2
     def test_permission_unset(self, get_http_api_auth):
-        payload = {"name": "test_permission_unset"}
+        payload = {"name": "permission_unset"}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["permission"] == "me", res
 
     @pytest.mark.p3
     def test_permission_none(self, get_http_api_auth):
-        payload = {"name": "test_permission_none", "permission": None}
+        payload = {"name": "permission_none", "permission": None}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 101, res
         assert "Input should be 'me' or 'team'" in res["message"], res
@@ -378,7 +378,7 @@ class TestDatasetCreate:
 
     @pytest.mark.p2
     def test_chunk_method_unset(self, get_http_api_auth):
-        payload = {"name": "test_chunk_method_unset"}
+        payload = {"name": "chunk_method_unset"}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["chunk_method"] == "naive", res
@@ -674,28 +674,20 @@ class TestDatasetCreate:
 
     @pytest.mark.p2
     def test_parser_config_empty(self, get_http_api_auth):
-        payload = {"name": "default_empty", "parser_config": {}}
+        payload = {"name": "parser_config_empty", "parser_config": {}}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["parser_config"] == {
-            "auto_keywords": 0,
-            "auto_questions": 0,
             "chunk_token_num": 128,
             "delimiter": r"\n",
-            "filename_embd_weight": None,
-            "graphrag": None,
             "html4excel": False,
             "layout_recognize": "DeepDOC",
-            "pages": None,
-            "raptor": None,
-            "tag_kb_ids": [],
-            "task_page_size": None,
-            "topn_tags": 1,
-        }
+            "raptor": {"use_raptor": False},
+        }, res
 
     @pytest.mark.p2
     def test_parser_config_unset(self, get_http_api_auth):
-        payload = {"name": "default_unset"}
+        payload = {"name": "parser_config_unset"}
         res = create_dataset(get_http_api_auth, payload)
         assert res["code"] == 0, res
         assert res["data"]["parser_config"] == {
@@ -708,10 +700,16 @@ class TestDatasetCreate:
 
     @pytest.mark.p3
     def test_parser_config_none(self, get_http_api_auth):
-        payload = {"name": "default_none", "parser_config": None}
+        payload = {"name": "parser_config_none", "parser_config": None}
         res = create_dataset(get_http_api_auth, payload)
-        assert res["code"] == 101, res
-        assert "Input should be a valid dictionary or instance of ParserConfig" in res["message"], res
+        assert res["code"] == 0, res
+        assert res["data"]["parser_config"] == {
+            "chunk_token_num": 128,
+            "delimiter": "\\n",
+            "html4excel": False,
+            "layout_recognize": "DeepDOC",
+            "raptor": {"use_raptor": False},
+        }, res
 
     @pytest.mark.p2
     @pytest.mark.parametrize(

sdk/python/test/test_http_api/test_dataset_mangement/test_update_dataset.py

@@ -91,7 +91,7 @@ class TestCapability:
 class TestDatasetUpdate:
     @pytest.mark.p3
     def test_dataset_id_not_uuid(self, get_http_api_auth):
-        payload = {"name": "dataset_id_not_uuid"}
+        payload = {"name": "not_uuid"}
         res = update_dataset(get_http_api_auth, "not_uuid", payload)
         assert res["code"] == 101, res
         assert "Input should be a valid UUID" in res["message"], res
@@ -178,22 +178,19 @@ class TestDatasetUpdate:
 
     @pytest.mark.p3
     @pytest.mark.parametrize(
-        "name, avatar_prefix, expected_message",
+        "avatar_prefix, expected_message",
         [
-            ("empty_prefix", "", "Missing MIME prefix. Expected format: data:<mime>;base64,<data>"),
-            ("missing_comma", "data:image/png;base64", "Missing MIME prefix. Expected format: data:<mime>;base64,<data>"),
-            ("unsupported_mine_type", "invalid_mine_prefix:image/png;base64,", "Invalid MIME prefix format. Must start with 'data:'"),
-            ("invalid_mine_type", "data:unsupported_mine_type;base64,", "Unsupported MIME type. Allowed: ['image/jpeg', 'image/png']"),
+            ("", "Missing MIME prefix. Expected format: data:<mime>;base64,<data>"),
+            ("data:image/png;base64", "Missing MIME prefix. Expected format: data:<mime>;base64,<data>"),
+            ("invalid_mine_prefix:image/png;base64,", "Invalid MIME prefix format. Must start with 'data:'"),
+            ("data:unsupported_mine_type;base64,", "Unsupported MIME type. Allowed: ['image/jpeg', 'image/png']"),
         ],
         ids=["empty_prefix", "missing_comma", "unsupported_mine_type", "invalid_mine_type"],
     )
-    def test_avatar_invalid_prefix(self, get_http_api_auth, add_dataset_func, tmp_path, name, avatar_prefix, expected_message):
+    def test_avatar_invalid_prefix(self, get_http_api_auth, add_dataset_func, tmp_path, avatar_prefix, expected_message):
         dataset_id = add_dataset_func
         fn = create_image_file(tmp_path / "ragflow_test.png")
-        payload = {
-            "name": name,
-            "avatar": f"{avatar_prefix}{encode_avatar(fn)}",
-        }
+        payload = {"avatar": f"{avatar_prefix}{encode_avatar(fn)}"}
         res = update_dataset(get_http_api_auth, dataset_id, payload)
         assert res["code"] == 101, res
         assert expected_message in res["message"], res
@@ -312,18 +309,18 @@ class TestDatasetUpdate:
 
     @pytest.mark.p1
     @pytest.mark.parametrize(
-        "name, permission",
+        "permission",
         [
-            ("me", "me"),
-            ("team", "team"),
-            ("me_upercase", "ME"),
-            ("team_upercase", "TEAM"),
+            "me",
+            "team",
+            "ME",
+            "TEAM",
         ],
         ids=["me", "team", "me_upercase", "team_upercase"],
     )
-    def test_permission(self, get_http_api_auth, add_dataset_func, name, permission):
+    def test_permission(self, get_http_api_auth, add_dataset_func, permission):
         dataset_id = add_dataset_func
-        payload = {"name": name, "permission": permission}
+        payload = {"permission": permission}
         res = update_dataset(get_http_api_auth, dataset_id, payload)
         assert res["code"] == 0, res
 
@@ -351,7 +348,7 @@ class TestDatasetUpdate:
     @pytest.mark.p3
     def test_permission_none(self, get_http_api_auth, add_dataset_func):
         dataset_id = add_dataset_func
-        payload = {"name": "test_permission_none", "permission": None}
+        payload = {"permission": None}
         res = update_dataset(get_http_api_auth, dataset_id, payload)
         assert res["code"] == 101, res
         assert "Input should be 'me' or 'team'" in res["message"], res
@@ -697,32 +694,65 @@ class TestDatasetUpdate:
 
         res = list_datasets(get_http_api_auth)
         assert res["code"] == 0, res
-        assert res["data"][0]["parser_config"] == {}
-
-    # @pytest.mark.p2
-    # def test_parser_config_unset(self, get_http_api_auth, add_dataset_func):
-    #     dataset_id = add_dataset_func
-    #     payload = {"name": "default_unset"}
-    #     res = update_dataset(get_http_api_auth, dataset_id, payload)
-    #     assert res["code"] == 0, res
-
-    #     res = list_datasets(get_http_api_auth)
-    #     assert res["code"] == 0, res
-    #     assert res["data"][0]["parser_config"] == {
-    #         "chunk_token_num": 128,
-    #         "delimiter": r"\n",
-    #         "html4excel": False,
-    #         "layout_recognize": "DeepDOC",
-    #         "raptor": {"use_raptor": False},
-    #     }, res
+        assert res["data"][0]["parser_config"] == {
+            "chunk_token_num": 128,
+            "delimiter": r"\n",
+            "html4excel": False,
+            "layout_recognize": "DeepDOC",
+            "raptor": {"use_raptor": False},
+        }, res
 
     @pytest.mark.p3
     def test_parser_config_none(self, get_http_api_auth, add_dataset_func):
         dataset_id = add_dataset_func
         payload = {"parser_config": None}
         res = update_dataset(get_http_api_auth, dataset_id, payload)
-        assert res["code"] == 101, res
-        assert "Input should be a valid dictionary or instance of ParserConfig" in res["message"], res
+        assert res["code"] == 0, res
+
+        res = list_datasets(get_http_api_auth, {"id": dataset_id})
+        assert res["code"] == 0, res
+        assert res["data"][0]["parser_config"] == {
+            "chunk_token_num": 128,
+            "delimiter": r"\n",
+            "html4excel": False,
+            "layout_recognize": "DeepDOC",
+            "raptor": {"use_raptor": False},
+        }, res
+
+    @pytest.mark.p3
+    def test_parser_config_empty_with_chunk_method_change(self, get_http_api_auth, add_dataset_func):
+        dataset_id = add_dataset_func
+        payload = {"chunk_method": "qa", "parser_config": {}}
+        res = update_dataset(get_http_api_auth, dataset_id, payload)
+        assert res["code"] == 0, res
+
+        res = list_datasets(get_http_api_auth)
+        print(res)
+        assert res["code"] == 0, res
+        assert res["data"][0]["parser_config"] == {"raptor": {"use_raptor": False}}, res
+
+    @pytest.mark.p3
+    def test_parser_config_unset_with_chunk_method_change(self, get_http_api_auth, add_dataset_func):
+        dataset_id = add_dataset_func
+        payload = {"chunk_method": "qa"}
+        res = update_dataset(get_http_api_auth, dataset_id, payload)
+        assert res["code"] == 0, res
+
+        res = list_datasets(get_http_api_auth)
+        assert res["code"] == 0, res
+        assert res["data"][0]["parser_config"] == {"raptor": {"use_raptor": False}}, res
+
+    @pytest.mark.p3
+    def test_parser_config_none_with_chunk_method_change(self, get_http_api_auth, add_dataset_func):
+        dataset_id = add_dataset_func
+        payload = {"chunk_method": "qa", "parser_config": None}
+        res = update_dataset(get_http_api_auth, dataset_id, payload)
+        assert res["code"] == 0, res
+
+        res = list_datasets(get_http_api_auth, {"id": dataset_id})
+        print(res)
+        assert res["code"] == 0, res
+        assert res["data"][0]["parser_config"] == {"raptor": {"use_raptor": False}}, res
 
     @pytest.mark.p2
     @pytest.mark.parametrize(
@@ -742,8 +772,35 @@ class TestDatasetUpdate:
             {"unknown_field": "unknown_field"},
         ],
     )
-    def test_unsupported_field(self, get_http_api_auth, add_dataset_func, payload):
+    def test_field_unsupported(self, get_http_api_auth, add_dataset_func, payload):
         dataset_id = add_dataset_func
         res = update_dataset(get_http_api_auth, dataset_id, payload)
         assert res["code"] == 101, res
         assert "Extra inputs are not permitted" in res["message"], res
+
+    @pytest.mark.p2
+    def test_field_unset(self, get_http_api_auth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = list_datasets(get_http_api_auth)
+        assert res["code"] == 0, res
+        original_data = res["data"][0]
+
+        payload = {"name": "default_unset"}
+        res = update_dataset(get_http_api_auth, dataset_id, payload)
+        assert res["code"] == 0, res
+
+        res = list_datasets(get_http_api_auth)
+        assert res["code"] == 0, res
+        assert res["data"][0]["avatar"] == original_data["avatar"], res
+        assert res["data"][0]["description"] == original_data["description"], res
+        assert res["data"][0]["embedding_model"] == original_data["embedding_model"], res
+        assert res["data"][0]["permission"] == original_data["permission"], res
+        assert res["data"][0]["chunk_method"] == original_data["chunk_method"], res
+        assert res["data"][0]["pagerank"] == original_data["pagerank"], res
+        assert res["data"][0]["parser_config"] == {
+            "chunk_token_num": 128,
+            "delimiter": r"\n",
+            "html4excel": False,
+            "layout_recognize": "DeepDOC",
+            "raptor": {"use_raptor": False},
+        }, res