fix

Author: jackalcooper

tests/mock_server.py

@@ -452,13 +452,7 @@ async def generate(
             )
 
         if model_spec.is_reasoning and params.enable_thinking:
-            return JSONResponse(
-                status_code=400,
-                content={
-                    "code": "InternalError",
-                    "message": "Error code: 400 - {'code': 20015, 'message': 'Value error, current model does not support parameter `enable_thinking`.', 'data': None}",
-                },
-            )
+            params.enable_thinking = False
 
         proxy_stop_list: List[str] = []
         if params.stop:

tests/test_doc.py

@@ -166,12 +166,7 @@ def test_r1_enable_thinking_unsupported(self):
         }
         response = make_request(payload)
 
-        # 表格显示此处返回的是 InternalError，且 message 是上游透传回来的原始错误
-        assert_exact_error(
-            response,
-            expected_code_str="InternalError",
-            expected_message=ERR_MSG_R1_THINKING,
-        )
+        assert response.status_code == 200
 
 
 class TestFunctionalFixes:
@@ -188,11 +183,13 @@ def test_r1_usage_structure_no_text_tokens(self):
         payload = {
             "model": "pre-siliconflow/deepseek-r1",
             "input": {"messages": [{"role": "user", "content": "你好"}]},
-            "parameters": {"max_tokens": 10} # 限制输出以加快测试
+            "parameters": {"max_tokens": 10},  # 限制输出以加快测试
         }
         response = make_request(payload)
 
-        assert response.status_code == 200, f"Request failed with {response.status_code}"
+        assert (
+            response.status_code == 200
+        ), f"Request failed with {response.status_code}"
 
         # 解析 SSE 流或直接读取 JSON (假设是非流式或读取最后一块)
         # 这里简化处理：如果是流式，我们需要解析最后一个包含 usage 的块
@@ -201,8 +198,12 @@ def test_r1_usage_structure_no_text_tokens(self):
 
         # 简单起见，读取整个流并查找 usage
         content = response.text
-        assert '"text_tokens"' not in content, "Response output_tokens_details should NOT contain 'text_tokens'"
-        assert '"reasoning_tokens"' in content, "Response output_tokens_details MUST contain 'reasoning_tokens'"
+        assert (
+            '"text_tokens"' not in content
+        ), "Response output_tokens_details should NOT contain 'text_tokens'"
+        assert (
+            '"reasoning_tokens"' in content
+        ), "Response output_tokens_details MUST contain 'reasoning_tokens'"
 
     def test_tool_choice_invalid_error_mapping(self):
         """
@@ -213,7 +214,9 @@ def test_tool_choice_invalid_error_mapping(self):
         payload = {
             "model": "pre-siliconflow/deepseek-r1",
             "input": {
-                "messages": [{"role": "user", "content": "What is the weather like in Boston?"}]
+                "messages": [
+                    {"role": "user", "content": "What is the weather like in Boston?"}
+                ]
             },
             "parameters": {
                 "result_format": "message",
@@ -228,12 +231,12 @@ def test_tool_choice_invalid_error_mapping(self):
                             "parameters": {
                                 "type": "object",
                                 "properties": {"location": {"type": "string"}},
-                                "required": ["location"]
-                            }
-                        }
+                                "required": ["location"],
+                            },
+                        },
                     }
-                ]
-            }
+                ],
+            },
         }
         response = make_request(payload)
         data = response.json()
@@ -242,8 +245,9 @@ def test_tool_choice_invalid_error_mapping(self):
 
         assert data.get("code") == "InternalError"
         # 验证错误信息是否透传了上游的详细校验失败信息
-        assert expected_msg_snippet in data.get("message"), \
-            f"Expected error message to contain '{expected_msg_snippet}', got: {data.get('message')}"
+        assert expected_msg_snippet in data.get(
+            "message"
+        ), f"Expected error message to contain '{expected_msg_snippet}', got: {data.get('message')}"
 
     def test_history_tool_call_fix(self):
         """
@@ -256,40 +260,36 @@ def test_history_tool_call_fix(self):
             "model": "pre-siliconflow/deepseek-v3.2",
             "input": {
                 "messages": [
-                    {
-                        "role": "system",
-                        "content": "你是一个智能助手。"
-                    },
-                    {
-                        "role": "user",
-                        "content": "外部轴设置"
-                    },
+                    {"role": "system", "content": "你是一个智能助手。"},
+                    {"role": "user", "content": "外部轴设置"},
                     {
                         "role": "assistant",
                         "tool_calls": [
                             {
                                 "function": {
-                                    "arguments": "{\"input_text\": \"外部轴设置\"}",
-                                    "name": "KB20250625001"
+                                    "arguments": '{"input_text": "外部轴设置"}',
+                                    "name": "KB20250625001",
                                 },
                                 "id": "call_6478091069c2448b83f38e",
-                                "type": "function"
+                                "type": "function",
                             }
-                        ]
+                        ],
                     },
                     {
                         "role": "tool",
                         "content": "界面用于用户进行快速配置。",
-                        "tool_call_id": "call_6478091069c2448b83f38e"
-                    }
+                        "tool_call_id": "call_6478091069c2448b83f38e",
+                    },
                 ]
             },
             # 确保不开启思考，避免干扰测试 tool history 功能
-            "parameters": {"enable_thinking": False}
+            "parameters": {"enable_thinking": False},
         }
 
         response = make_request(payload)
-        assert response.status_code == 200, f"Previously 500 error scenario failed. Got: {response.text}"
+        assert (
+            response.status_code == 200
+        ), f"Previously 500 error scenario failed. Got: {response.text}"
 
     def test_prefix_completion_success(self):
         """
@@ -303,19 +303,17 @@ def test_prefix_completion_success(self):
             "input": {
                 "messages": [
                     {"role": "user", "content": "你好"},
-                    {
-                        "role": "assistant",
-                        "partial": True,
-                        "content": "你好，我是"
-                    }
+                    {"role": "assistant", "partial": True, "content": "你好，我是"},
                 ]
             },
             # 明确关闭 thinking 以测试单纯的前缀续写功能
-            "parameters": {"enable_thinking": False}
+            "parameters": {"enable_thinking": False},
         }
 
         response = make_request(payload)
-        assert response.status_code == 200, f"Prefix completion failed with {response.status_code}. Body: {response.text}"
+        assert (
+            response.status_code == 200
+        ), f"Prefix completion failed with {response.status_code}. Body: {response.text}"
 
         # 可选：验证返回内容确实是以前缀开始的续写
         # context = response.text...