parse_file

Instructions

Implementation Reference

• mcp_tool/tools/file_tool.py:58-109 (handler)

  The execute method of FileTool class — the main handler for the 'parse_file' tool. It receives arguments, validates file_path, determines file extension, and delegates to the appropriate specialized tool (pdf, word, excel, csv, markdown).

  async def execute(self, arguments: Dict[str, Any]) -> List[types.TextContent | types.ImageContent | types.EmbeddedResource]:
      """
      解析文件内容
      
      Args:
          arguments: 参数字典，必须包含'file_path'键
      
      Returns:
          解析结果列表
      """
      if "file_path" not in arguments:
          return [types.TextContent(
              type="text",
              text="错误: 缺少必要参数 'file_path'"
          )]
      
      file_path = arguments["file_path"]
      # 处理文件路径，支持挂载目录的转换
      file_path = self.process_file_path(file_path)
      
      if not os.path.exists(file_path):
          return [types.TextContent(
              type="text",
              text=f"错误: 文件不存在: {file_path}"
          )]
      
      # 获取文件扩展名（转换为小写）
      file_ext = os.path.splitext(file_path)[1].lower()
      
      try:
          # 根据文件扩展名选择处理工具
          if file_ext == '.pdf':
              return await self.pdf_tool.execute(arguments)
          elif file_ext in ['.doc', '.docx']:
              return await self.word_tool.execute(arguments)
          elif file_ext in ['.xls', '.xlsx', '.xlsm']:
              return await self.excel_tool.execute(arguments)
          elif file_ext == '.csv':
              return await self.csv_tool.execute(arguments)
          elif file_ext == '.md':
              return await self.markdown_tool.execute(arguments)
          else:
              return [types.TextContent(
                  type="text",
                  text=f"错误: 不支持的文件类型: {file_ext}"
              )]
      except Exception as e:
          error_details = traceback.format_exc()
          return [types.TextContent(
              type="text",
              text=f"错误: 处理文件时发生错误: {str(e)}\n{error_details}"
          )] 

• mcp_tool/tools/file_tool.py:38-47 (schema)

  The input_schema defines the required 'file_path' parameter (a string describing the local file path) for the parse_file tool.

  input_schema = {
      "type": "object",
      "required": ["file_path"],
      "properties": {
          "file_path": {
              "type": "string",
              "description": "文件的本地路径，例如'/path/to/document.pdf'",
          }
      },
  }

• mcp_tool/tools/file_tool.py:24-50 (registration)

  FileTool class is decorated with @ToolRegistry.register, which registers it under the name 'parse_file' (line 36). This makes the tool available to the MCP server.

  @ToolRegistry.register
  class FileTool(BaseTool):
      """
      综合文件处理工具，根据文件扩展名自动选择合适的处理方式
      支持的文件类型：
      - PDF文件 (.pdf)
      - Word文档 (.doc, .docx)
      - Excel文件 (.xls, .xlsx, .xlsm)
      - CSV文件 (.csv)
      - Markdown文件 (.md)
      """
      
      name = "parse_file"
      description = "解析文件内容，支持PDF、Word、Excel、CSV和Markdown格式"
      input_schema = {
          "type": "object",
          "required": ["file_path"],
          "properties": {
              "file_path": {
                  "type": "string",
                  "description": "文件的本地路径，例如'/path/to/document.pdf'",
              }
          },
      }
      
      def __init__(self):
          """初始化各种文件处理工具"""

• mcp_tool/tools/__init__.py:38-42 (registration)

  ToolRegistry class that provides the @register decorator. When @ToolRegistry.register is applied to a class, it stores the class in _tools dict keyed by its name (e.g., 'parse_file').

          return file_path
  
  
  # 工具注册器
  class ToolRegistry:

• mcp_tool/tools/loader.py:12-23 (helper)

  The load_tools() function auto-discovers and imports all tool modules (including file_tool.py), which triggers the @ToolRegistry.register decorators, registering 'parse_file' among other tools.

  def load_tools() -> List[Type[BaseTool]]:
      """
      自动加载tools目录下的所有工具模块
      
      Returns:
          List[Type[BaseTool]]: 已加载的工具类列表
      """
      # 获取当前模块的路径
      package_path = os.path.dirname(__file__)
      
      # 获取所有子模块
      for _, name, is_pkg in pkgutil.iter_modules([package_path]):