From 0569b50fed95ee702db62e128a1e2820c01ca210 Mon Sep 17 00:00:00 2001
From: Billy Bao <newyorkupperbay@gmail.com>
Date: Mon, 17 Nov 2025 15:27:19 +0800
Subject: [PATCH] Fix: create dataset return type inconsistent (#11272)

### What problem does this PR solve?

Fix: create dataset return type inconsistent #11167

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
---
 api/apps/kb_app.py                             |  4 ++++
 api/db/services/knowledgebase_service.py       | 18 +++++++++---------
 .../test/test_frontend_api/test_dataset.py     |  4 ++--
 3 files changed, 15 insertions(+), 11 deletions(-)

diff --git a/api/apps/kb_app.py b/api/apps/kb_app.py
index b7cf58a20..f173b56a0 100644
--- a/api/apps/kb_app.py
+++ b/api/apps/kb_app.py
@@ -55,6 +55,10 @@ def create():
         **req
     )
 
+    code = req.get("code")
+    if code:
+        return get_data_error_result(code=code, message=req.get("message"))
+
     try:
         if not KnowledgebaseService.save(**req):
             return get_data_error_result()
diff --git a/api/db/services/knowledgebase_service.py b/api/db/services/knowledgebase_service.py
index ca30ca074..28936576d 100644
--- a/api/db/services/knowledgebase_service.py
+++ b/api/db/services/knowledgebase_service.py
@@ -24,9 +24,9 @@ from common.time_utils import current_timestamp, datetime_format
 from api.db.services import duplicate_name
 from api.db.services.user_service import TenantService
 from common.misc_utils import get_uuid
-from common.constants import StatusEnum
+from common.constants import StatusEnum, RetCode
 from api.constants import DATASET_NAME_LIMIT
-from api.utils.api_utils import get_parser_config, get_data_error_result
+from api.utils.api_utils import get_parser_config
 
 class KnowledgebaseService(CommonService):
     """Service class for managing knowledge base operations.
@@ -391,12 +391,12 @@ class KnowledgebaseService(CommonService):
         """
         # Validate name
         if not isinstance(name, str):
-            return get_data_error_result(message="Dataset name must be string.")
+            return {"code": RetCode.DATA_ERROR, "message": "Dataset name must be string."}
         dataset_name = name.strip()
-        if dataset_name == "":
-            return get_data_error_result(message="Dataset name can't be empty.")
+        if len(dataset_name) == 0:
+            return {"code": RetCode.DATA_ERROR, "message": "Dataset name can't be empty."}
         if len(dataset_name.encode("utf-8")) > DATASET_NAME_LIMIT:
-            return get_data_error_result(message=f"Dataset name length is {len(dataset_name)} which is larger than {DATASET_NAME_LIMIT}")
+            return {"code": RetCode.DATA_ERROR, "message": f"Dataset name length is {len(dataset_name)} which is larger than {DATASET_NAME_LIMIT}"}
 
         # Deduplicate name within tenant
         dataset_name = duplicate_name(
@@ -409,7 +409,7 @@ class KnowledgebaseService(CommonService):
         # Verify tenant exists
         ok, _t = TenantService.get_by_id(tenant_id)
         if not ok:
-            return False, "Tenant not found."
+            return {"code": RetCode.DATA_ERROR, "message": "Tenant does not exist."}
 
         # Build payload
         kb_id = get_uuid()
@@ -419,10 +419,10 @@ class KnowledgebaseService(CommonService):
             "tenant_id": tenant_id,
             "created_by": tenant_id,
             "parser_id": (parser_id or "naive"),
-            **kwargs
+            **kwargs # Includes optional fields such as description, language, permission, avatar, parser_config, etc.
         }
 
-        # Default parser_config (align with kb_app.create) — do not accept external overrides
+        # Update parser_config (always override with validated default/merged config)
         payload["parser_config"] = get_parser_config(parser_id, kwargs.get("parser_config"))
 
         return payload
diff --git a/sdk/python/test/test_frontend_api/test_dataset.py b/sdk/python/test/test_frontend_api/test_dataset.py
index b57a3543b..3fb3e93ff 100644
--- a/sdk/python/test/test_frontend_api/test_dataset.py
+++ b/sdk/python/test/test_frontend_api/test_dataset.py
@@ -104,7 +104,7 @@ def test_invalid_name_dataset(get_auth):
     assert res['code'] == 100
 
     res = create_dataset(get_auth, "")
-    assert res['code'] == 100
+    assert res['code'] == 102
 
     long_string = ""
 
@@ -112,7 +112,7 @@ def test_invalid_name_dataset(get_auth):
         long_string += random.choice(string.ascii_letters + string.digits)
 
     res = create_dataset(get_auth, long_string)
-    assert res['code'] == 100
+    assert res['code'] == 102
     print(res)