improve comments readability

Author: ntudy

apps/run-agent/calculate_score_from_log.py

@@ -11,7 +11,7 @@
 
 
 def extract_score_from_log(run_dir, task_score_dict):
-    # 遍历所有task_{task_id}_attempt_*.log的文件，提取score
+    # Traverse all task_{task_id}_attempt_*.log files to extract score
     log_files = glob.glob(os.path.join(run_dir, "task_*_attempt_*.json"))
     for log_file in log_files:
         task_id = log_file.split("/")[-1].split("_")[1]
@@ -35,7 +35,7 @@ def main(results_dir: str, pass_at_k: int = 3):
 
     print(f"Analyzing results from: {results_dir}")
 
-    # 遍历所有run_*目录under results_dir
+    # Traverse all run_* directories under results_dir
     run_dirs = glob.glob(os.path.join(results_dir, "run_*"))
     task_score_dict = {}
     for run_dir in run_dirs:
@@ -50,7 +50,7 @@ def main(results_dir: str, pass_at_k: int = 3):
         else:
             failed_id.append(task)
 
-    # 保存简单的统计结果
+    # Save simple statistical results
     output_file = os.path.join(results_dir, f"average_scores_pass_at_{pass_at_k}.txt")
     with open(output_file, "w") as f:
         f.write("EVALUATION RESULTS\n")

apps/run-agent/common_benchmark.py

@@ -546,7 +546,7 @@ def prepare_task_description(
         if path.is_absolute():
             return task.task_question, str(path.resolve())
 
-        # 构建完整文件路径：数据目录 + 相对路径
+        # Build complete file path: data directory + relative path
         full_file_path = Path(self.data_dir) / path
         return task.task_question, str(full_file_path.resolve())
 
@@ -636,9 +636,9 @@ def main(*args):
     with hydra.initialize_config_dir(config_dir=config_path(), version_base=None):
         cfg = hydra.compose(config_name=config_name(), overrides=list(args))
         _ = bootstrap_logger()
-        # 默认关闭 tracing, 同时不 set key
+        # Default to disable tracing, and don't set key
         set_tracing_disabled(True)
         set_tracing_export_api_key("fake-key")
-        # 压制 trace provider 的报警
+        # Suppress trace provider warnings
         bootstrap_silent_trace_provider()
         asyncio.run(entrypoint(cfg))

apps/visualize-trace/app.py

@@ -8,7 +8,7 @@
 
 app = Flask(__name__)
 
-# 全局变量存储分析器实例
+# Global variable to store analyzer instance
 analyzer = None
 
 
@@ -25,13 +25,13 @@ def list_files():
         directory = request.args.get("directory", "")
 
         if not directory:
-            # 默认行为：检查上级目录
+            # Default behavior: check parent directory
             directory = os.path.abspath("..")
 
-        # 扩展路径（处理~等符号）
+        # Expand path (handle ~ and other symbols)
         directory = os.path.expanduser(directory)
 
-        # 转换为绝对路径
+        # Convert to absolute path
         directory = os.path.abspath(directory)
 
         if not os.path.exists(directory):
@@ -46,7 +46,7 @@ def list_files():
                 if file.endswith(".json"):
                     file_path = os.path.join(directory, file)
                     try:
-                        # 获取文件大小和修改时间
+                        # Get file size and modification time
                         stat = os.stat(file_path)
                         json_files.append(
                             {
@@ -61,7 +61,7 @@ def list_files():
                             {"name": file, "path": file_path, "size": 0, "modified": 0}
                         )
 
-            # 按文件名排序
+            # Sort by filename
             json_files.sort(key=lambda x: x["name"])
 
             return jsonify(
@@ -91,7 +91,7 @@ def load_trace():
     if not file_path:
         return jsonify({"error": "请提供文件路径"}), 400
 
-    # 如果是相对路径，转换为绝对路径
+    # If it's a relative path, convert to absolute path
     if not os.path.isabs(file_path):
         file_path = os.path.abspath(file_path)
 
@@ -193,7 +193,7 @@ def get_raw_messages():
         main_history = analyzer.get_main_agent_history()
         browser_sessions = analyzer.get_browser_agent_sessions()
 
-        # 获取消息结构概览
+        # Get message structure overview
         main_messages = analyzer.get_main_agent_messages()
         message_structure = []
 

libs/miroflow-tool/src/miroflow/tool/manager.py

@@ -50,8 +50,8 @@ async def execute_tool_call(
 class ToolManager(ToolManagerProtocol):
     def __init__(self, server_configs, tool_blacklist=None):
         """
-        初始化 ToolManager。
-        :param server_configs: create_server_parameters() 返回的列表
+        Initialize ToolManager.
+        :param server_configs: List returned by create_server_parameters()
         """
         self.server_configs = server_configs
         self.server_dict = {
@@ -61,7 +61,7 @@ def __init__(self, server_configs, tool_blacklist=None):
         self.tool_blacklist = tool_blacklist if tool_blacklist else set()
 
         logger.info(
-            f"ToolManager 初始化，已加载服务器: {list(self.server_dict.keys())}"
+            f"ToolManager initialized, loaded servers: {list(self.server_dict.keys())}"
         )
 
     def _is_huggingface_dataset_or_space_url(self, url):
@@ -88,14 +88,14 @@ def _should_block_hf_scraping(self, tool_name, arguments):
         )
 
     def get_server_params(self, server_name):
-        """获取指定服务器的参数"""
+        """Get parameters for specified server"""
         return self.server_dict.get(server_name)
 
     async def _find_servers_with_tool(self, tool_name):
         """
-        在所有服务器中查找包含指定工具名称的服务器
-        :param tool_name: 要查找的工具名称
-        :return: 包含该工具的服务器名称列表
+        Find servers containing the specified tool name among all servers
+        :param tool_name: Tool name to search for
+        :return: List of server names containing the tool
         """
         servers_with_tool = []
 
@@ -111,7 +111,7 @@ async def _find_servers_with_tool(self, tool_name):
                         ) as session:
                             await session.initialize()
                             tools_response = await session.list_tools()
-                            # 遵循与 get_all_tool_definitions 相同的 blacklist 逻辑
+                            # Follow the same blacklist logic as get_all_tool_definitions
                             for tool in tools_response.tools:
                                 if (server_name, tool.name) in self.tool_blacklist:
                                     continue
@@ -129,8 +129,8 @@ async def _find_servers_with_tool(self, tool_name):
                             await session.initialize()
                             tools_response = await session.list_tools()
                             for tool in tools_response.tools:
-                                # 与 get_all_tool_definitions 保持一致：SSE 部分没有 blacklist 处理
-                                # 可以在这里添加特定工具的过滤逻辑（如果需要）
+                                # Consistent with get_all_tool_definitions: SSE part has no blacklist processing
+                                # Can add specific tool filtering logic here (if needed)
                                 # if server_name == "tool-excel" and tool.name not in ["get_workbook_metadata", "read_data_from_excel"]:
                                 #     continue
                                 if tool.name == tool_name:
@@ -140,28 +140,28 @@ async def _find_servers_with_tool(self, tool_name):
                     logger.error(
                         f"错误: 服务器 '{server_name}' 的参数类型未知: {type(server_params)}"
                     )
-                    # 对于未知类型，我们跳过而不是抛出异常，因为这是查找功能
+                    # For unknown types, we skip rather than throw an exception, because this is a search function
                     continue
             except Exception as e:
                 logger.error(
-                    f"错误: 无法连接或获取服务器 '{server_name}' 的工具以查找 '{tool_name}': {e}"
+                    f"Error: Cannot connect or get tools from server '{server_name}' to find '{tool_name}': {e}"
                 )
                 continue
 
         return servers_with_tool
 
     async def get_all_tool_definitions(self):
         """
-        连接到所有已配置的服务器，获取它们的工具定义。
-        返回一个适合传递给 Prompt 生成器的列表。
+        Connect to all configured servers and get their tool definitions.
+        Returns a list suitable for passing to Prompt generators.
         """
         all_servers_for_prompt = []
-        # 处理远程服务器工具
+        # Handle remote server tools
         for config in self.server_configs:
             server_name = config["name"]
             server_params = config["params"]
             one_server_for_prompt = {"name": server_name, "tools": []}
-            logger.info(f"正在获取服务器 '{server_name}' 的工具定义...")
+            logger.info(f"Getting tool definitions for server '{server_name}'...")
 
             try:
                 if isinstance(server_params, StdioServerParameters):
@@ -175,7 +175,7 @@ async def get_all_tool_definitions(self):
                             for tool in tools_response.tools:
                                 if (server_name, tool.name) in self.tool_blacklist:
                                     logger.info(
-                                        f"server '{server_name}' 中的工具 '{tool.name}' 被列入黑名单，跳过。"
+                                        f"Tool '{tool.name}' in server '{server_name}' is blacklisted, skipping."
                                     )
                                     continue
                                 one_server_for_prompt["tools"].append(
@@ -196,7 +196,7 @@ async def get_all_tool_definitions(self):
                             await session.initialize()
                             tools_response = await session.list_tools()
                             for tool in tools_response.tools:
-                                # 可以在这里添加特定工具的过滤逻辑（如果需要）
+                                # Can add specific tool filtering logic here (if needed)
                                 # if server_name == "tool-excel" and tool.name not in ["get_workbook_metadata", "read_data_from_excel"]:
                                 #     continue
                                 one_server_for_prompt["tools"].append(
@@ -215,13 +215,13 @@ async def get_all_tool_definitions(self):
                     )
 
                 logger.info(
-                    f"成功获取服务器 '{server_name}' 的 {len(one_server_for_prompt['tools'])} 个工具定义。"
+                    f"Successfully obtained {len(one_server_for_prompt['tools'])} tool definitions for server '{server_name}'."
                 )
                 all_servers_for_prompt.append(one_server_for_prompt)
 
             except Exception as e:
-                logger.error(f"错误: 无法连接或获取服务器 '{server_name}' 的工具: {e}")
-                # 仍然添加服务器条目，但标记工具列表为空或包含错误信息
+                logger.error(f"Error: Cannot connect or get tools from server '{server_name}': {e}")
+                # Still add server entry, but mark tool list as empty or containing error information
                 one_server_for_prompt["tools"] = [
                     {"error": f"Failed to fetch tools: {e}"}
                 ]
@@ -232,11 +232,11 @@ async def get_all_tool_definitions(self):
     @with_timeout(600)
     async def execute_tool_call(self, server_name, tool_name, arguments) -> Any:
         """
-        执行单个工具调用。
-        :param server_name: 服务器名称
-        :param tool_name: 工具名称
-        :param arguments: 工具参数字典
-        :return: 包含结果或错误的字典
+        Execute a single tool call.
+        :param server_name: Server name
+        :param tool_name: Tool name
+        :param arguments: Tool arguments dictionary
+        :return: Dictionary containing result or error
         """
 
         # 原远程服务器调用逻辑

libs/miroflow-tool/src/miroflow/tool/mcp_servers/searching_mcp_server.py

@@ -4,7 +4,7 @@
 import calendar
 from fastmcp import FastMCP
 from mcp.client.stdio import stdio_client
-from mcp import ClientSession, StdioServerParameters  # (已在 config.py 导入)
+from mcp import ClientSession, StdioServerParameters  # (already imported in config.py)
 import wikipedia
 import asyncio
 from .utils.smart_request import smart_request, request_to_json

libs/miroflow-tool/src/miroflow/tool/mcp_servers/utils/smart_request.py

@@ -6,7 +6,7 @@
     ClientSession,
     StdioServerParameters,
     stdio_client,
-)  # (已在 config.py 导入)
+)  # (already imported in config.py)
 import urllib.parse
 from markitdown import MarkItDown
 import io

libs/miroflow-tool/tests/test_reading_mcp_server.py

@@ -71,7 +71,7 @@ async def test_tool_definitions_available(self):
                     "uri": f"file://{files_for_tests_dir}/test-pdf.pdf",
                 },
                 "should_succeed": True,
-                "expected_content_keywords": ["做笔记"],
+                "expected_content_keywords": ["note-taking"],
             },
             {
                 "name": "convert_excel_file",
@@ -81,7 +81,7 @@ async def test_tool_definitions_available(self):
                     "uri": f"file://{files_for_tests_dir}/test-excel.xlsx",
                 },
                 "should_succeed": True,
-                "expected_content_keywords": ["财务"],
+                "expected_content_keywords": ["finance"],
             },
             {
                 "name": "convert_word_file",
@@ -91,7 +91,7 @@ async def test_tool_definitions_available(self):
                     "uri": f"file://{files_for_tests_dir}/test-word.docx",
                 },
                 "should_succeed": True,
-                "expected_content_keywords": ["做笔记"],
+                "expected_content_keywords": ["note-taking"],
             },
             {
                 "name": "convert_powerpoint_file",
@@ -101,7 +101,7 @@ async def test_tool_definitions_available(self):
                     "uri": f"file://{files_for_tests_dir}/test-ppt.pptx",
                 },
                 "should_succeed": True,
-                "expected_content_keywords": ["计算机系统"],
+                "expected_content_keywords": ["computer system"],
             },
             {
                 "name": "convert_image_file",
@@ -111,7 +111,7 @@ async def test_tool_definitions_available(self):
                     "uri": f"file://{files_for_tests_dir}/test-zip.zip",
                 },
                 "should_succeed": True,
-                "expected_content_keywords": ["计算机系统", "做笔记", "财务"],
+                "expected_content_keywords": ["computer system", "note-taking", "finance"],
             },
         ],
     )

libs/miroflow/src/miroflow/llm/provider_client_base.py

@@ -74,7 +74,7 @@ def __post_init__(self):
         self.openrouter_provider: Optional[str] = self.cfg.llm.get(
             "openrouter_provider"
         )
-        # 安全地处理字符串到bool的转换
+        # Safely handle string to bool conversion
         disable_cache_control_val = self.cfg.llm.get("disable_cache_control", False)
         if isinstance(disable_cache_control_val, str):
             self.disable_cache_control: bool = (
@@ -84,10 +84,10 @@ def __post_init__(self):
             self.disable_cache_control: bool = bool(disable_cache_control_val)
 
         logger.info(
-            f"openrouter_provider配置值: {self.openrouter_provider} (类型: {type(self.openrouter_provider)})"
+            f"openrouter_provider config value: {self.openrouter_provider} (type: {type(self.openrouter_provider)})"
         )
         logger.info(
-            f"disable_cache_control配置值: {disable_cache_control_val} (类型: {type(disable_cache_control_val)}) -> 解析为: {self.disable_cache_control}"
+            f"disable_cache_control config value: {disable_cache_control_val} (type: {type(disable_cache_control_val)}) -> parsed as: {self.disable_cache_control}"
         )
 
         self.token_usage = self._reset_token_usage()