feat(api): add unified index API and dataset management endpoints (#14222)

### What problem does this PR solve? ## Summary Refactor the dataset API layer into a clean service/REST separation pattern, add a unified `/index` API for graph/raptor/mindmap operations, and introduce several new dataset management endpoints with full test coverage. ## Changes ### Service Layer (`dataset_api_service.py`) - Added `trace_index(dataset_id, tenant_id, index_type)` — unified trace function for all index types - Added `run_index`, `delete_index` service functions - Added `get_dataset`, `get_ingestion_summary`, `list_ingestion_logs`, `get_ingestion_log` - Added `run_embedding`, `list_tags`, `aggregate_tags`, `delete_tags`, `rename_tag` - Added `get_flattened_metadata`, `get_auto_metadata`, `update_auto_metadata` ### REST API Layer (`dataset_api.py`) **New unified routes:** | Method | Route | Description | |--------|-------|-------------| | POST | `/datasets/<id>/index?type=graph\|raptor\|mindmap` | Run index task | | GET | `/datasets/<id>/index?type=graph\|raptor\|mindmap` | Trace index task | | DELETE | `/datasets/<id>/<index_type>` | Delete index | | GET | `/datasets/<id>` | Get dataset details | | GET | `/datasets/<id>/ingestions/summary` | Ingestion summary | | GET | `/datasets/<id>/ingestions` | List ingestion logs | | GET | `/datasets/<id>/ingestions/<log_id>` | Get single ingestion log | | POST | `/datasets/<id>/embedding` | Run embedding | | GET | `/datasets/<id>/tags` | List tags | | GET | `/datasets/tags/aggregation` | Aggregate tags across datasets | | DELETE | `/datasets/<id>/tags` | Delete tags | | PUT | `/datasets/<id>/tags` | Rename tag | | GET | `/datasets/metadata/flattened` | Get flattened metadata | | GET/PUT | `/datasets/<id>/metadata/config` | New metadata config path | **Removed routes (replaced by unified `/index`):** - `POST /datasets/<id>/mindmap` - `GET /datasets/<id>/mindmap` **Preserved legacy routes (backward compatibility):** - `/run_graphrag`, `/trace_graphrag`, `/run_raptor`, `/trace_raptor` - `/auto_metadata` GET/PUT ### Test Suite - Updated `common.py` helpers: added `trace_index`, removed `run_mindmap`/`trace_mindmap` - Added 7 new test files with 39 test cases total: | Test File | Cases | |-----------|-------| | `test_get_dataset.py` | 4 | | `test_ingestion_summary.py` | 2 | | `test_ingestion_logs.py` | 5 | | `test_index_api.py` | 14 | | `test_embedding.py` | 2 | | `test_tags.py` | 8 | | `test_flattened_metadata.py` | 4 | - Deleted `test_mindmap_tasks.py` (covered by unified index tests) ## Design Decisions 1. **Unified `/index?type=...`** — single endpoint replaces 3 separate route pairs for graph/raptor/mindmap 2. **Backward compatibility** — old routes (`/run_graphrag`, `/run_raptor`, `/auto_metadata`) preserved alongside new paths 3. **`_VALID_INDEX_TYPES = {"graph", "raptor", "mindmap"}`** — input validation via constant set 4. **`_INDEX_TYPE_TO_TASK_ID_FIELD`** — maps index type to KB model task ID field for clean dispatch ## Files Changed - `api/apps/restful_apis/dataset_api.py` - `api/apps/services/dataset_api_service.py` - `sdk/python/ragflow_sdk/modules/dataset.py` - `test/testcases/test_http_api/common.py` - `test/testcases/test_http_api/test_dataset_management/` (7 new files) ### Type of change - [x] New Feature (non-breaking change which adds functionality) - [x] Refactoring --------- Signed-off-by: noob <yixiao121314@outlook.com>
2026-07-29 04:08:12 +08:00 · 2026-04-27 01:38:01 +00:00
parent fb95136f39
commit 4dcc42e0e1
51 changed files with 1765 additions and 4381 deletions
--- a/test/testcases/test_http_api/common.py
+++ b/test/testcases/test_http_api/common.py
@@ -23,7 +23,8 @@ from utils.file_utils import create_txt_file
 HEADERS = {"Content-Type": "application/json"}
 DATASETS_API_URL = f"/api/{VERSION}/datasets"
 FILE_API_URL = f"/api/{VERSION}/datasets/{{dataset_id}}/documents"
-FILE_CHUNK_API_URL = f"/api/{VERSION}/datasets/{{dataset_id}}/chunks"
+FILE_PARSE_API_URL = f"/api/{VERSION}/datasets/{{dataset_id}}/documents/parse"
+FILE_STOP_PARSE_API_URL = f"/api/{VERSION}/datasets/{{dataset_id}}/documents/stop"
 CHUNK_API_URL = f"/api/{VERSION}/datasets/{{dataset_id}}/documents/{{document_id}}/chunks"
 CHAT_ASSISTANT_API_URL = f"/api/{VERSION}/chats"
 SESSION_WITH_CHAT_ASSISTANT_API_URL = f"/api/{VERSION}/chats/{{chat_id}}/sessions"
@@ -136,15 +137,15 @@ def delete_all_documents(auth, dataset_id, *, page_size=1000):
    return delete_documents(auth, dataset_id, {"ids": None, "delete_all": True})


-def parse_documents(auth, dataset_id, payload=None):
-    url = f"{HOST_ADDRESS}{FILE_CHUNK_API_URL}".format(dataset_id=dataset_id)
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+def parse_documents(auth, dataset_id, payload=None, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{FILE_PARSE_API_URL}".format(dataset_id=dataset_id)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


 def stop_parse_documents(auth, dataset_id, payload=None):
-    url = f"{HOST_ADDRESS}{FILE_CHUNK_API_URL}".format(dataset_id=dataset_id)
-    res = requests.delete(url=url, headers=HEADERS, auth=auth, json=payload)
+    url = f"{HOST_ADDRESS}{FILE_STOP_PARSE_API_URL}".format(dataset_id=dataset_id)
+    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
    return res.json()


@@ -161,9 +162,9 @@ def bulk_upload_documents(auth, dataset_id, num, tmp_path):


 # CHUNK MANAGEMENT WITHIN DATASET
-def add_chunk(auth, dataset_id, document_id, payload=None):
+def add_chunk(auth, dataset_id, document_id, payload=None, *, headers=HEADERS):
    url = f"{HOST_ADDRESS}{CHUNK_API_URL}".format(dataset_id=dataset_id, document_id=document_id)
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


@@ -195,9 +196,9 @@ def delete_all_chunks(auth, dataset_id, document_id, *, page_size=1000):
    return delete_chunks(auth, dataset_id, document_id, {"chunk_ids": None, "delete_all": True})


-def retrieval_chunks(auth, payload=None):
+def retrieval_chunks(auth, payload=None, *, headers=HEADERS):
    url = f"{HOST_ADDRESS}{RETRIEVAL_API_URL}"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


@@ -210,9 +211,9 @@ def batch_add_chunks(auth, dataset_id, document_id, num):


 # CHAT ASSISTANT MANAGEMENT
-def create_chat_assistant(auth, payload=None):
+def create_chat_assistant(auth, payload=None, *, headers=HEADERS):
    url = f"{HOST_ADDRESS}{CHAT_ASSISTANT_API_URL}"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


@@ -259,9 +260,9 @@ def batch_create_chat_assistants(auth, num):


 # SESSION MANAGEMENT
-def create_session_with_chat_assistant(auth, chat_assistant_id, payload=None):
+def create_session_with_chat_assistant(auth, chat_assistant_id, payload=None, *, headers=HEADERS):
    url = f"{HOST_ADDRESS}{SESSION_WITH_CHAT_ASSISTANT_API_URL}".format(chat_id=chat_assistant_id)
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


@@ -297,13 +298,13 @@ def batch_add_sessions_with_chat_assistant(auth, chat_assistant_id, num):

 # DATASET GRAPH AND TASKS
 def knowledge_graph(auth, dataset_id, params=None):
-    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/knowledge_graph"
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/graph/search"
    res = requests.get(url=url, headers=HEADERS, auth=auth, params=params)
    return res.json()


 def delete_knowledge_graph(auth, dataset_id, payload=None):
-    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/knowledge_graph"
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/graph"
    if payload is None:
        res = requests.delete(url=url, headers=HEADERS, auth=auth)
    else:
@@ -311,39 +312,15 @@ def delete_knowledge_graph(auth, dataset_id, payload=None):
    return res.json()


-def run_graphrag(auth, dataset_id, payload=None):
-    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/run_graphrag"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
-    return res.json()
-
-
-def trace_graphrag(auth, dataset_id, params=None):
-    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/trace_graphrag"
-    res = requests.get(url=url, headers=HEADERS, auth=auth, params=params)
-    return res.json()
-
-
-def run_raptor(auth, dataset_id, payload=None):
-    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/run_raptor"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
-    return res.json()
-
-
-def trace_raptor(auth, dataset_id, params=None):
-    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/trace_raptor"
-    res = requests.get(url=url, headers=HEADERS, auth=auth, params=params)
-    return res.json()
-
-
 def metadata_summary(auth, dataset_id, params=None):
    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/metadata/summary"
    res = requests.get(url=url, headers=HEADERS, auth=auth, params=params)
    return res.json()


-def metadata_batch_update(auth, dataset_id, payload=None):
+def metadata_batch_update(auth, dataset_id, payload=None, *, headers=HEADERS):
    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/metadata/update"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


@@ -358,16 +335,16 @@ def update_documents_metadata(auth, dataset_id, payload=None):


 # CHAT COMPLETIONS AND RELATED QUESTIONS
-def related_questions(auth, payload=None):
+def related_questions(auth, payload=None, *, headers=HEADERS):
    url = f"{HOST_ADDRESS}/api/{VERSION}/sessions/related_questions"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


 # AGENT MANAGEMENT AND SESSIONS
-def create_agent(auth, payload=None):
+def create_agent(auth, payload=None, *, headers=HEADERS):
    url = f"{HOST_ADDRESS}{AGENT_API_URL}"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
    return res.json()


@@ -439,7 +416,7 @@ def chat_completions(auth, chat_id=None, payload=None):
    return res.json()


-def chat_completions_openai(auth, chat_id, payload=None):
+def chat_completions_openai(auth, chat_id, payload=None, *, headers=HEADERS):
    """
    Send a request to the OpenAI-compatible chat completions endpoint.

@@ -454,5 +431,88 @@ def chat_completions_openai(auth, chat_id, payload=None):
        Response JSON in OpenAI chat completions format with usage information
    """
    url = f"{HOST_ADDRESS}/api/{VERSION}/chats_openai/{chat_id}/chat/completions"
-    res = requests.post(url=url, headers=HEADERS, auth=auth, json=payload)
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
+    return res.json()
+
+
+# NEW DATASET ENDPOINTS
+def get_dataset(auth, dataset_id, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}"
+    res = requests.get(url=url, headers=headers, auth=auth)
+    return res.json()
+
+
+def get_ingestion_summary(auth, dataset_id, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/ingestions/summary"
+    res = requests.get(url=url, headers=headers, auth=auth)
+    return res.json()
+
+
+def list_ingestion_logs(auth, dataset_id, params=None, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/ingestions"
+    res = requests.get(url=url, headers=headers, auth=auth, params=params)
+    return res.json()
+
+
+def get_ingestion_log(auth, dataset_id, log_id, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/ingestions/{log_id}"
+    res = requests.get(url=url, headers=headers, auth=auth)
+    return res.json()
+
+
+def run_index(auth, dataset_id, index_type, payload=None, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/index"
+    params = {"type": index_type}
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload, params=params)
+    return res.json()
+
+
+def trace_index(auth, dataset_id, index_type, params=None, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/index"
+    all_params = {"type": index_type}
+    if params:
+        all_params.update(params)
+    res = requests.get(url=url, headers=headers, auth=auth, params=all_params)
+    return res.json()
+
+
+def delete_index(auth, dataset_id, index_type, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/{index_type}"
+    res = requests.delete(url=url, headers=headers, auth=auth)
+    return res.json()
+
+
+def run_embedding(auth, dataset_id, payload=None, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/embedding"
+    res = requests.post(url=url, headers=headers, auth=auth, json=payload)
+    return res.json()
+
+
+def list_tags(auth, dataset_id, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/tags"
+    res = requests.get(url=url, headers=headers, auth=auth)
+    return res.json()
+
+
+def aggregate_tags(auth, dataset_ids, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/tags/aggregation"
+    res = requests.get(url=url, headers=headers, auth=auth, params={"dataset_ids": ",".join(dataset_ids)})
+    return res.json()
+
+
+def delete_tags(auth, dataset_id, tags, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/tags"
+    res = requests.delete(url=url, headers=headers, auth=auth, json={"tags": tags})
+    return res.json()
+
+
+def rename_tag(auth, dataset_id, from_tag, to_tag, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/{dataset_id}/tags"
+    res = requests.put(url=url, headers=headers, auth=auth, json={"from_tag": from_tag, "to_tag": to_tag})
+    return res.json()
+
+
+def get_flattened_metadata(auth, dataset_ids, *, headers=HEADERS):
+    url = f"{HOST_ADDRESS}{DATASETS_API_URL}/metadata/flattened"
+    res = requests.get(url=url, headers=headers, auth=auth, params={"dataset_ids": ",".join(dataset_ids)})
    return res.json()
--- a/test/testcases/test_http_api/conftest.py
+++ b/test/testcases/test_http_api/conftest.py
@@ -43,7 +43,7 @@ from utils.file_utils import (
 )


-@wait_for(30, 1, "Document parsing timeout")
+@wait_for(200, 1, "Document parsing timeout")
 def condition(_auth, _dataset_id):
    res = list_documents(_auth, _dataset_id)
    for doc in res["data"]["docs"]:
--- a/test/testcases/test_http_api/test_dataset_management/test_embedding.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_embedding.py
@@ -0,0 +1,32 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+#
+import pytest
+from common import run_embedding
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestRunEmbedding:
+    @pytest.mark.p2
+    def test_run_embedding_no_documents(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = run_embedding(HttpApiAuth, dataset_id)
+        assert res["code"] == 102, res
+        assert "No documents in Dataset" in res.get("message", ""), res
+
+    @pytest.mark.p2
+    def test_run_embedding_invalid_id(self, HttpApiAuth):
+        res = run_embedding(HttpApiAuth, "invalid_id")
+        assert res["code"] != 0, res
--- a/test/testcases/test_http_api/test_dataset_management/test_flattened_metadata.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_flattened_metadata.py
@@ -0,0 +1,42 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+#
+import pytest
+from common import get_flattened_metadata
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestFlattenedMetadata:
+    @pytest.mark.p2
+    def test_get_flattened_metadata_success(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = get_flattened_metadata(HttpApiAuth, [dataset_id])
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_get_flattened_metadata_multiple_datasets(self, HttpApiAuth, add_datasets_func):
+        dataset_ids = add_datasets_func
+        res = get_flattened_metadata(HttpApiAuth, dataset_ids)
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_get_flattened_metadata_empty_ids(self, HttpApiAuth):
+        res = get_flattened_metadata(HttpApiAuth, [])
+        assert res["code"] != 0, res
+
+    @pytest.mark.p2
+    def test_get_flattened_metadata_invalid_id(self, HttpApiAuth):
+        res = get_flattened_metadata(HttpApiAuth, ["invalid_id"])
+        assert res["code"] != 0, res
--- a/test/testcases/test_http_api/test_dataset_management/test_get_dataset.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_get_dataset.py
@@ -0,0 +1,45 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+#
+import pytest
+from common import get_dataset
+from libs.auth import RAGFlowHttpApiAuth
+from configs import INVALID_API_TOKEN
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestGetDataset:
+    @pytest.mark.p2
+    def test_get_dataset_success(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = get_dataset(HttpApiAuth, dataset_id)
+        assert res["code"] == 0, res
+        assert res["data"]["id"] == dataset_id, res
+
+    @pytest.mark.p2
+    def test_get_dataset_invalid_id(self, HttpApiAuth):
+        res = get_dataset(HttpApiAuth, "invalid_dataset_id")
+        assert res["code"] != 0, res
+
+    @pytest.mark.p2
+    def test_get_dataset_unauthorized(self, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = get_dataset(RAGFlowHttpApiAuth(INVALID_API_TOKEN), dataset_id)
+        assert res["code"] != 0, res
+
+    @pytest.mark.p2
+    def test_get_dataset_nonexistent(self, HttpApiAuth):
+        res = get_dataset(HttpApiAuth, "0" * 32)
+        assert res["code"] != 0, res
--- a/test/testcases/test_http_api/test_dataset_management/test_graphrag_tasks.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_graphrag_tasks.py
@@ -1,89 +0,0 @@
-#
-#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
-#
-#  Licensed under the Apache License, Version 2.0 (the "License");
-#  you may not use this file except in compliance with the License.
-#  You may obtain a copy of the License at
-#
-#      http://www.apache.org/licenses/LICENSE-2.0
-#
-#  Unless required by applicable law or agreed to in writing, software
-#  distributed under the License is distributed on an "AS IS" BASIS,
-#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#  See the License for the specific language governing permissions and
-#  limitations under the License.
-#
-import pytest
-from common import bulk_upload_documents, list_documents, parse_documents, run_graphrag, trace_graphrag
-from utils import wait_for
-
-
-@wait_for(200, 1, "Document parsing timeout")
-def _parse_done(auth, dataset_id, document_ids=None):
-    res = list_documents(auth, dataset_id)
-    target_docs = res["data"]["docs"]
-    if document_ids is None:
-        return all(doc.get("run") == "DONE" for doc in target_docs)
-    target_ids = set(document_ids)
-    for doc in target_docs:
-        if doc.get("id") in target_ids and doc.get("run") != "DONE":
-            return False
-    return True
-
-
-class TestGraphRAGTasks:
-    @pytest.mark.p2
-    def test_trace_graphrag_before_run(self, HttpApiAuth, add_dataset_func):
-        dataset_id = add_dataset_func
-        res = trace_graphrag(HttpApiAuth, dataset_id)
-        assert res["code"] == 0, res
-        assert res["data"] == {}, res
-
-    @pytest.mark.p2
-    def test_run_graphrag_no_documents(self, HttpApiAuth, add_dataset_func):
-        dataset_id = add_dataset_func
-        res = run_graphrag(HttpApiAuth, dataset_id)
-        assert res["code"] == 102, res
-        assert "No documents in Dataset" in res.get("message", ""), res
-
-    @pytest.mark.p3
-    def test_run_graphrag_returns_task_id(self, HttpApiAuth, add_dataset_func, tmp_path):
-        dataset_id = add_dataset_func
-        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
-        res = run_graphrag(HttpApiAuth, dataset_id)
-        assert res["code"] == 0, res
-        assert res["data"].get("graphrag_task_id"), res
-
-    @pytest.mark.p3
-    def test_trace_graphrag_until_complete(self, HttpApiAuth, add_dataset_func, tmp_path):
-        dataset_id = add_dataset_func
-        document_ids = bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
-        res = parse_documents(HttpApiAuth, dataset_id, {"document_ids": document_ids})
-        assert res["code"] == 0, res
-        _parse_done(HttpApiAuth, dataset_id, document_ids)
-
-        res = run_graphrag(HttpApiAuth, dataset_id)
-        assert res["code"] == 0, res
-
-        last_res = {}
-
-        @wait_for(200, 1, "GraphRAG task timeout")
-        def condition():
-            res = trace_graphrag(HttpApiAuth, dataset_id)
-            if res["code"] != 0:
-                return False
-            data = res.get("data") or {}
-            if not data:
-                return False
-            if data.get("task_type") != "graphrag":
-                return False
-            progress = data.get("progress")
-            if progress in (-1, 1, -1.0, 1.0):
-                last_res["res"] = res
-                return True
-            return False
-
-        condition()
-        res = last_res["res"]
-        assert res["data"]["task_type"] == "graphrag", res
-        assert res["data"].get("progress") in (-1, 1, -1.0, 1.0), res
--- a/test/testcases/test_http_api/test_dataset_management/test_index_api.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_index_api.py
@@ -0,0 +1,166 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+#
+import pytest
+from common import (
+    bulk_upload_documents,
+    list_documents,
+    run_index,
+    trace_index,
+    delete_index,
+)
+from utils import wait_for
+
+
+@wait_for(200, 1, "Document parsing timeout")
+def _parse_done(auth, dataset_id, document_ids=None):
+    res = list_documents(auth, dataset_id)
+    if res.get("code") != 0:
+        return False
+    target_docs = res.get("data", {}).get("docs", [])
+    if not target_docs:
+        return False
+    if document_ids is None:
+        return all(doc.get("run") == "DONE" for doc in target_docs)
+    target_ids = set(document_ids)
+    seen_ids = set()
+    for doc in target_docs:
+        doc_id = doc.get("id")
+        if doc_id in target_ids:
+            seen_ids.add(doc_id)
+            if doc.get("run") != "DONE":
+                return False
+    return seen_ids == target_ids
+
+
+@wait_for(60, 1, "Index task creation timeout")
+def _index_task_created(auth, dataset_id, index_type):
+    res = trace_index(auth, dataset_id, index_type)
+    if res.get("code") != 0:
+        return False
+    return bool(res.get("data", {}).get("id"))
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestRunIndex:
+    @pytest.mark.p2
+    def test_run_index_graph(self, HttpApiAuth, add_dataset_func, tmp_path):
+        dataset_id = add_dataset_func
+        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
+        res = run_index(HttpApiAuth, dataset_id, "graph")
+        assert res["code"] == 0, res
+        assert res["data"].get("task_id"), res
+
+    @pytest.mark.p2
+    def test_run_index_raptor(self, HttpApiAuth, add_dataset_func, tmp_path):
+        dataset_id = add_dataset_func
+        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
+        res = run_index(HttpApiAuth, dataset_id, "raptor")
+        assert res["code"] == 0, res
+        assert res["data"].get("task_id"), res
+
+    @pytest.mark.p2
+    def test_run_index_mindmap(self, HttpApiAuth, add_dataset_func, tmp_path):
+        dataset_id = add_dataset_func
+        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
+        res = run_index(HttpApiAuth, dataset_id, "mindmap")
+        assert res["code"] == 0, res
+        assert res["data"].get("task_id"), res
+
+    @pytest.mark.p2
+    def test_run_index_invalid_type(self, HttpApiAuth, add_dataset_func, tmp_path):
+        dataset_id = add_dataset_func
+        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
+        res = run_index(HttpApiAuth, dataset_id, "invalid_type")
+        assert res["code"] != 0, res
+
+    @pytest.mark.p2
+    def test_run_index_no_documents(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = run_index(HttpApiAuth, dataset_id, "raptor")
+        assert res["code"] == 102, res
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestDeleteIndex:
+    @pytest.mark.p2
+    def test_delete_graph(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = delete_index(HttpApiAuth, dataset_id, "graph")
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_delete_raptor(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = delete_index(HttpApiAuth, dataset_id, "raptor")
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_delete_mindmap(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = delete_index(HttpApiAuth, dataset_id, "mindmap")
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_delete_invalid_type(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = delete_index(HttpApiAuth, dataset_id, "invalid_type")
+        assert res["code"] != 0, res
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestTraceIndex:
+    @pytest.mark.p2
+    def test_trace_index_graph(self, HttpApiAuth, add_dataset_func, tmp_path):
+        dataset_id = add_dataset_func
+        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
+        res = run_index(HttpApiAuth, dataset_id, "graph")
+        assert res["code"] == 0, res
+        _index_task_created(HttpApiAuth, dataset_id, "graph")
+        res = trace_index(HttpApiAuth, dataset_id, "graph")
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_trace_index_raptor(self, HttpApiAuth, add_dataset_func, tmp_path):
+        dataset_id = add_dataset_func
+        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
+        res = run_index(HttpApiAuth, dataset_id, "raptor")
+        assert res["code"] == 0, res
+        _index_task_created(HttpApiAuth, dataset_id, "raptor")
+        res = trace_index(HttpApiAuth, dataset_id, "raptor")
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_trace_index_mindmap(self, HttpApiAuth, add_dataset_func, tmp_path):
+        dataset_id = add_dataset_func
+        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
+        res = run_index(HttpApiAuth, dataset_id, "mindmap")
+        assert res["code"] == 0, res
+        _index_task_created(HttpApiAuth, dataset_id, "mindmap")
+        res = trace_index(HttpApiAuth, dataset_id, "mindmap")
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_trace_index_invalid_type(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = trace_index(HttpApiAuth, dataset_id, "invalid_type")
+        assert res["code"] != 0, res
+
+    @pytest.mark.p2
+    def test_trace_index_no_task(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = trace_index(HttpApiAuth, dataset_id, "graph")
+        assert res["code"] == 0, res
+        assert res["data"] == {}
--- a/test/testcases/test_http_api/test_dataset_management/test_ingestion_logs.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_ingestion_logs.py
@@ -0,0 +1,53 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+#
+import pytest
+from common import list_ingestion_logs, get_ingestion_log
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestListIngestionLogs:
+    @pytest.mark.p2
+    def test_list_ingestion_logs_success(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = list_ingestion_logs(HttpApiAuth, dataset_id)
+        assert res["code"] == 0, res
+        assert "total" in res["data"], res
+        assert "logs" in res["data"], res
+
+    @pytest.mark.p2
+    def test_list_ingestion_logs_with_pagination(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = list_ingestion_logs(HttpApiAuth, dataset_id, params={"page": 1, "page_size": 10})
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_list_ingestion_logs_invalid_id(self, HttpApiAuth):
+        res = list_ingestion_logs(HttpApiAuth, "invalid_id")
+        assert res["code"] != 0, res
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestGetIngestionLog:
+    @pytest.mark.p2
+    def test_get_ingestion_log_not_found(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = get_ingestion_log(HttpApiAuth, dataset_id, "nonexistent_log_id")
+        assert res["code"] != 0, res
+
+    @pytest.mark.p2
+    def test_get_ingestion_log_invalid_dataset(self, HttpApiAuth):
+        res = get_ingestion_log(HttpApiAuth, "invalid_id", "some_log_id")
+        assert res["code"] != 0, res
--- a/test/testcases/test_http_api/test_dataset_management/test_ingestion_summary.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_ingestion_summary.py
@@ -0,0 +1,35 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+#
+import pytest
+from common import get_ingestion_summary
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestIngestionSummary:
+    @pytest.mark.p2
+    def test_ingestion_summary_success(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = get_ingestion_summary(HttpApiAuth, dataset_id)
+        assert res["code"] == 0, res
+        assert "doc_num" in res["data"], res
+        assert "chunk_num" in res["data"], res
+        assert "token_num" in res["data"], res
+        assert "status" in res["data"], res
+
+    @pytest.mark.p2
+    def test_ingestion_summary_invalid_id(self, HttpApiAuth):
+        res = get_ingestion_summary(HttpApiAuth, "invalid_id")
+        assert res["code"] != 0, res
--- a/test/testcases/test_http_api/test_dataset_management/test_raptor_tasks.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_raptor_tasks.py
@@ -1,89 +0,0 @@
-#
-#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
-#
-#  Licensed under the Apache License, Version 2.0 (the "License");
-#  you may not use this file except in compliance with the License.
-#  You may obtain a copy of the License at
-#
-#      http://www.apache.org/licenses/LICENSE-2.0
-#
-#  Unless required by applicable law or agreed to in writing, software
-#  distributed under the License is distributed on an "AS IS" BASIS,
-#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#  See the License for the specific language governing permissions and
-#  limitations under the License.
-#
-import pytest
-from common import bulk_upload_documents, list_documents, parse_documents, run_raptor, trace_raptor
-from utils import wait_for
-
-
-@wait_for(200, 1, "Document parsing timeout")
-def _parse_done(auth, dataset_id, document_ids=None):
-    res = list_documents(auth, dataset_id)
-    target_docs = res["data"]["docs"]
-    if document_ids is None:
-        return all(doc.get("run") == "DONE" for doc in target_docs)
-    target_ids = set(document_ids)
-    for doc in target_docs:
-        if doc.get("id") in target_ids and doc.get("run") != "DONE":
-            return False
-    return True
-
-
-class TestRaptorTasks:
-    @pytest.mark.p2
-    def test_trace_raptor_before_run(self, HttpApiAuth, add_dataset_func):
-        dataset_id = add_dataset_func
-        res = trace_raptor(HttpApiAuth, dataset_id)
-        assert res["code"] == 0, res
-        assert res["data"] == {}, res
-
-    @pytest.mark.p2
-    def test_run_raptor_no_documents(self, HttpApiAuth, add_dataset_func):
-        dataset_id = add_dataset_func
-        res = run_raptor(HttpApiAuth, dataset_id)
-        assert res["code"] == 102, res
-        assert "No documents in Dataset" in res.get("message", ""), res
-
-    @pytest.mark.p3
-    def test_run_raptor_returns_task_id(self, HttpApiAuth, add_dataset_func, tmp_path):
-        dataset_id = add_dataset_func
-        bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
-        res = run_raptor(HttpApiAuth, dataset_id)
-        assert res["code"] == 0, res
-        assert res["data"].get("raptor_task_id"), res
-
-    @pytest.mark.p3
-    def test_trace_raptor_until_complete(self, HttpApiAuth, add_dataset_func, tmp_path):
-        dataset_id = add_dataset_func
-        document_ids = bulk_upload_documents(HttpApiAuth, dataset_id, 1, tmp_path)
-        res = parse_documents(HttpApiAuth, dataset_id, {"document_ids": document_ids})
-        assert res["code"] == 0, res
-        _parse_done(HttpApiAuth, dataset_id, document_ids)
-
-        res = run_raptor(HttpApiAuth, dataset_id)
-        assert res["code"] == 0, res
-
-        last_res = {}
-
-        @wait_for(200, 1, "RAPTOR task timeout")
-        def condition():
-            res = trace_raptor(HttpApiAuth, dataset_id)
-            if res["code"] != 0:
-                return False
-            data = res.get("data") or {}
-            if not data:
-                return False
-            if data.get("task_type") != "raptor":
-                return False
-            progress = data.get("progress")
-            if progress in (-1, 1, -1.0, 1.0):
-                last_res["res"] = res
-                return True
-            return False
-
-        condition()
-        res = last_res["res"]
-        assert res["data"]["task_type"] == "raptor", res
-        assert res["data"].get("progress") in (-1, 1, -1.0, 1.0), res
--- a/test/testcases/test_http_api/test_dataset_management/test_tags.py
+++ b/test/testcases/test_http_api/test_dataset_management/test_tags.py
@@ -0,0 +1,84 @@
+#
+#  Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+#
+import pytest
+from common import (
+    list_tags,
+    aggregate_tags,
+    delete_tags,
+    rename_tag,
+)
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestListTags:
+    @pytest.mark.p2
+    def test_list_tags_success(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = list_tags(HttpApiAuth, dataset_id)
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_list_tags_invalid_id(self, HttpApiAuth):
+        res = list_tags(HttpApiAuth, "invalid_id")
+        assert res["code"] != 0, res
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestAggregateTags:
+    @pytest.mark.p2
+    def test_aggregate_tags_success(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = aggregate_tags(HttpApiAuth, [dataset_id])
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_aggregate_tags_multiple_datasets(self, HttpApiAuth, add_datasets_func):
+        dataset_ids = add_datasets_func
+        res = aggregate_tags(HttpApiAuth, dataset_ids)
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_aggregate_tags_empty_ids(self, HttpApiAuth):
+        res = aggregate_tags(HttpApiAuth, [])
+        assert res["code"] != 0, res
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestDeleteTags:
+    @pytest.mark.p2
+    def test_delete_tags_missing_body(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = delete_tags(HttpApiAuth, dataset_id, [])
+        assert res["code"] == 0, res
+
+    @pytest.mark.p2
+    def test_delete_tags_invalid_id(self, HttpApiAuth):
+        res = delete_tags(HttpApiAuth, "invalid_id", ["tag1"])
+        assert res["code"] != 0, res
+
+
+@pytest.mark.usefixtures("clear_datasets")
+class TestRenameTag:
+    @pytest.mark.p2
+    def test_rename_tag_empty_names(self, HttpApiAuth, add_dataset_func):
+        dataset_id = add_dataset_func
+        res = rename_tag(HttpApiAuth, dataset_id, "", "")
+        assert res["code"] != 0, res
+
+    @pytest.mark.p2
+    def test_rename_tag_invalid_id(self, HttpApiAuth):
+        res = rename_tag(HttpApiAuth, "invalid_id", "old_tag", "new_tag")
+        assert res["code"] != 0, res
--- a/test/testcases/test_http_api/test_file_management_within_dataset/test_metadata_retrieval.py
+++ b/test/testcases/test_http_api/test_file_management_within_dataset/test_metadata_retrieval.py
@@ -27,11 +27,14 @@ from common import (
    delete_datasets,
    list_documents,
    update_document,
+    upload_documents,
+    parse_documents,
+    retrieval_chunks,
 )
 from utils import wait_for


-@wait_for(30, 1, "Document parsing timeout")
+@wait_for(120, 1, "Document parsing timeout")
 def _condition_parsing_complete(_auth, dataset_id):
    res = list_documents(_auth, dataset_id)
    if res["code"] != 0:
@@ -39,7 +42,7 @@ def _condition_parsing_complete(_auth, dataset_id):

    for doc in res["data"]["docs"]:
        status = doc.get("run", "UNKNOWN")
-        if status == "FAILED":
+        if status in ("FAIL", "FAILED"):
            pytest.fail(f"Document parsing failed: {doc}")
            return False
        if status != "DONE":
@@ -62,35 +65,17 @@ def add_dataset_with_metadata(HttpApiAuth):
    import requests
    from configs import HOST_ADDRESS, VERSION

-    metadata_config = {
-        "type": "object",
-        "properties": {
-            "character": {
-                "description": "Historical figure name",
-                "type": "string"
-            },
-            "era": {
-                "description": "Historical era",
-                "type": "string"
-            },
-            "achievements": {
-                "description": "Major achievements",
-                "type": "array",
-                "items": {
-                    "type": "string"
-                }
-            }
-        }
-    }
-
-    res = requests.post(
-        url=f"{HOST_ADDRESS}/{VERSION}/kb/update_metadata_setting",
+    res = requests.put(
+        url=f"{HOST_ADDRESS}/api/{VERSION}/datasets/{dataset_id}/metadata/config",
        headers={"Content-Type": "application/json"},
        auth=HttpApiAuth,
        json={
-            "kb_id": dataset_id,
-            "metadata": metadata_config,
-            "enable_metadata": False
+            "enabled": False,
+            "fields": [
+                {"name": "character", "type": "string", "description": "Historical figure name"},
+                {"name": "era", "type": "string", "description": "Historical era"},
+                {"name": "achievements", "type": "list", "description": "Major achievements"},
+            ]
        }
    ).json()

@@ -112,8 +97,6 @@ class TestMetadataWithRetrieval:

        Verifies that chunks are only retrieved from documents matching the metadata condition.
        """
-        from common import upload_documents, parse_documents, retrieval_chunks
-
        dataset_id = add_dataset_with_metadata

        # Create two documents with different metadata
--- a/test/testcases/test_http_api/test_file_management_within_dataset/test_metadata_summary.py
+++ b/test/testcases/test_http_api/test_file_management_within_dataset/test_metadata_summary.py
@@ -28,16 +28,12 @@ def _summary_to_counts(summary):

 class TestMetadataSummary:
    @pytest.mark.p2
-    def test_metadata_summary_missing_kb_id(self, HttpApiAuth, add_document_func):
+    def test_metadata_summary_nonexistent_kb_id(self, HttpApiAuth, add_document_func):
        """
        Call with non-existent dataset
-        :param HttpApiAuth:
-        :param add_document_func:
-        :return:
        """
-        res = metadata_summary(HttpApiAuth, "")
-        assert res["code"] == 404, res
-        assert res["message"] == "Not Found: /api/v1/datasets//metadata/summary", res
+        res = metadata_summary(HttpApiAuth, "0" * 32)
+        assert res["code"] == 102, res

    @pytest.mark.p2
    def test_metadata_summary_invalid_kb_id(self, HttpApiAuth, add_document_func):
--- a/test/testcases/test_http_api/test_file_management_within_dataset/test_parse_documents.py
+++ b/test/testcases/test_http_api/test_file_management_within_dataset/test_parse_documents.py
@@ -58,11 +58,11 @@ class TestAuthorization:
    @pytest.mark.parametrize(
        "invalid_auth, expected_code, expected_message",
        [
-            (None, 0, "`Authorization` can't be empty"),
+            (None, 401, "<Unauthorized '401: Unauthorized'>"),
            (
                RAGFlowHttpApiAuth(INVALID_API_TOKEN),
-                109,
-                "Authentication error: API key is invalid!",
+                401,
+                "<Unauthorized '401: Unauthorized'>",
            ),
        ],
    )
@@ -101,7 +101,7 @@ class TestDocumentsParse:
    @pytest.mark.parametrize(
        "dataset_id, expected_code, expected_message",
        [
-            ("", 100, "<MethodNotAllowed '405: Method Not Allowed'>"),
+            ("", 102, "You don't own the dataset ."),
            (
                "invalid_dataset_id",
                102,
--- a/test/testcases/test_http_api/test_file_management_within_dataset/test_stop_parse_documents.py
+++ b/test/testcases/test_http_api/test_file_management_within_dataset/test_stop_parse_documents.py
@@ -48,11 +48,11 @@ class TestAuthorization:
    @pytest.mark.parametrize(
        "invalid_auth, expected_code, expected_message",
        [
-            (None, 0, "`Authorization` can't be empty"),
+            (None, 401, "<Unauthorized '401: Unauthorized'>"),
            (
                RAGFlowHttpApiAuth(INVALID_API_TOKEN),
-                109,
-                "Authentication error: API key is invalid!",
+                401,
+                "<Unauthorized '401: Unauthorized'>",
            ),
        ],
    )
@@ -105,7 +105,7 @@ class TestDocumentsParseStop:
    @pytest.mark.parametrize(
        "invalid_dataset_id, expected_code, expected_message",
        [
-            ("", 100, "<MethodNotAllowed '405: Method Not Allowed'>"),
+            ("", 102, "You don't own the dataset ."),
            (
                "invalid_dataset_id",
                102,