get_data_info

Instructions

Implementation Reference

• src/stata_mcp/api/get_data_info.py:18-42 (handler)

  Primary handler function for the get_data_info tool. Accepts data_path, optional vars_list, encoding, and config_file. Resolves the path, determines file extension, fetches the appropriate data handler class, instantiates it, and returns JSON-serialized dataset info.

  def get_data_info(
      data_path: str,
      vars_list: List[str] | None = None,
      encoding: str = "utf-8",
      config_file: str | Path | None = None,
  ) -> str:
      """Return descriptive statistics for a supported dataset."""
      runtime = create_runtime_context(config_file=config_file)
      resolved_data_path = Path(data_path).expanduser().resolve()
      data_extension = resolved_data_path.suffix.lower().strip(".")
  
      data_info_cls = get_data_handler(data_extension)
      if not data_info_cls:
          return f"Unsupported file extension now: {data_extension}"
  
      data_info = data_info_cls(
          resolved_data_path,
          vars_list,
          encoding=encoding,
          cache_dir=runtime.tmp_base_path,
      )
      try:
          return json.dumps(data_info.info, ensure_ascii=False)
      except Exception as error:
          return f"Failed to generate data summary for {resolved_data_path}: {error}"

• src/stata_mcp/mcp_servers.py:379-425 (handler)

  Alternative (legacy/mcp-server) handler for get_data_info. Same core logic but uses config.STATA_MCP_FOLDER.TMP for caching and supports a head parameter (row preview). Also includes logging and cache awareness.

  def get_data_info(
          data_path: str,
          vars_list: List[str] | None = None,
          encoding: str = "utf-8",
          head: int = 0,
  ) -> str:
      """
      Return descriptive statistics for a supported data file.
  
      Args:
          data_path (str): Absolute path to .dta, .csv, .xlsx, .xls, .sav file.
          vars_list (List[str] | None): Optional variable subset (default: all variables).
          encoding (str): File encoding (ignored for .dta).
          head (int): Number of preview rows (0 = disabled).
  
      Returns:
          str: JSON string with overview, variable details, and config.
  
      Examples:
          >>> get_data_info("/Applications/Stata/auto.dta")
          >>> get_data_info("/Applications/Stata/auto.dta", vars_list=["price", "mpg"], head=5)
      """
      data_path = Path(data_path).expanduser().resolve()
      data_extension = data_path.suffix.lower().strip(".")
  
      # Lazy import: pandas/numpy/requests are heavy, only load when needed
      from .data_info import get_data_handler
  
      # Get the appropriate data handler class from the registry
      data_info_cls = get_data_handler(data_extension)
  
      if not data_info_cls:
          logging.error(f"Unsupported file extension: {data_extension} for data file: {data_path}")
          return f"Unsupported file extension now: {data_extension}"
  
      data_info = data_info_cls(data_path, vars_list, encoding=encoding, cache_dir=config.STATA_MCP_FOLDER.TMP, head=head)
      try:
          info = data_info.info
          if data_info.is_cache:
              saved_path = info.get("saved_path", None)
              logging.info(f"Successfully generated data summary for {data_path}, saved to {saved_path}")
          else:
              logging.info(f"Successfully generated data summary for {data_path}")
          return json.dumps(info, ensure_ascii=False)
      except Exception as e:
          logging.error(f"Failed to generate data summary for {data_path}: {str(e)}")
          return f"Failed to generate data summary for {data_path}: {str(e)}"

• src/stata_mcp/mcp_servers.py:597-606 (registration)

  Registration entry for get_data_info in the _TOOL_REGISTRY dict. Maps the tool name to its description, the handler function, and the profiles ('core', 'all') under which it is registered.

  "get_data_info": {
      "description": (
          "Get descriptive statistics and a data preview for a data file "
          "(dta, csv, xlsx). Returns overview, variable details, "
          "and optional head rows filtered by requested variables. "
          "Use when you need to understand a dataset or have no prior knowledge of the data."
      ),
      "func": get_data_info,
      "profiles": {"core", "all"},
  },

• src/stata_mcp/api/__init__.py:12-27 (schema)

  Re-export of get_data_info from the api package, making it available via from ..api import get_data_info.

  from .get_data_info import get_data_info
  from .read_log import read_log
  from .stata_do import stata_do
  from .stata_help import stata_help
  from .write_dofile import write_dofile
  
  __all__ = [
      "RuntimeContext",
      "create_runtime_context",
      "ado_package_install",
      "get_data_info",
      "read_log",
      "stata_do",
      "stata_help",
      "write_dofile",
  ]

• src/stata_mcp/cli/_parsers.py:106-106 (helper)

  CLI parser definition flagging get_data_info as a 'core' tool in the --core argument's help text.

  help="Register only core tools (stata_do, get_data_info, help)",