from langchain_core.tools import tool @tool(parse_docstring=True) def get_text_file(file_name: str) -> str: """ Loads a text file by its name and returns its content as a string. Args: file_name (str): The name of the text file to load. The file name should include a task ID followed by a file extension (e.g., "12345.txt"). Returns: str: The content of the text file as a string. """ print(f"Loading text file: {file_name}") try: import questions task_id, ext = file_name.rsplit(".", 1) if not task_id or not ext: return "Invalid file name format. Please provide a valid file name with task ID and extension." file_content = questions.get_file(task_id) if not file_content: return f"File with task ID {task_id} not found." return file_content.decode("utf-8") except Exception as e: return f"Error loading text file: {e}" @tool(parse_docstring=True) def load_excel_file(file_name: str) -> str: """ Loads an Excel file by its name and returns its content as a string. Args: file_name (str): The name of the Excel file to load. The file name should include a task ID followed by a file extension (e.g., "12345.xlsx"). Returns: str: The content of the Excel file as a string. """ print(f"Loading Excel file: {file_name}") try: from io import BytesIO import pandas as pd import questions task_id, ext = file_name.rsplit(".", 1) if not task_id or not ext: return "Invalid file name format. Please provide a valid file name with task ID and extension." if ext.lower() not in ["xlsx", "xls"]: return "Invalid file format. Please provide an Excel file." file_bytes = questions.get_file(task_id) if not file_bytes: return f"File with task ID {task_id} not found." data = pd.read_excel(BytesIO(file_bytes)) print(f"Data shape: {data.shape}") print(f"Data columns: {data.columns.tolist()}") print(f"Data preview:\n{data.head()}") return data.to_markdown(index=False) except Exception as e: return f"Error loading Excel file: {e}" @tool(parse_docstring=True) def sum_rows_cols(file_name: str, rows: list[str], cols: list[str]) -> float | str: """ Sums all data from the specified rows and columns in an Excel file. Args: file_name (str): The name of the Excel file to load. The file name should include a task ID followed by a file extension (e.g., "12345.xlsx"). rows (list[str]): List of row labels to sum. These should match the index labels in the Excel file. cols (list[str]): List of column labels to sum. These should match the column headers in the Excel file. Returns: float | str: The total sum of the specified rows and columns, or an error message if the operation fails. """ try: from io import BytesIO import pandas as pd import questions task_id, ext = file_name.rsplit(".", 1) if not task_id or not ext or ext.lower() not in ["xlsx", "xls"]: return "Invalid file name format. Please provide a valid Excel file." file_bytes = questions.get_file(task_id) if not file_bytes: return f"File with task ID {task_id} not found." data = pd.read_excel(BytesIO(file_bytes), index_col=0) print(f"Data shape: {data.shape}") print(f"Data columns: {data.columns.tolist()}") # Ensure rows and cols are in the DataFrame if not all(row in data.index for row in rows): return f"Some specified rows {rows} are not found in the data." if not all(col in data.columns for col in cols): return f"Some specified columns {cols} are not found in the data." # Sum all data from the specified rows and columns total_sum = data.loc[rows, cols].sum().sum() return total_sum except Exception as e: return f"Error processing Excel file: {e}"