Python Excel转HDF5打包exe后报'sequence item 2'类型错误求助

我编写了一个将多工作表Excel文件转换为.h5格式的Python程序，在Python环境中运行正常，但使用PyInstaller打包为exe文件后，convert_to_hdf5函数抛出错误：'sequence item 2: expected str instance, NoneType found'。已多次尝试添加各类hidden-import但无法解决，怀疑缺少未识别的依赖库，求解决思路。

程序代码如下：

import tkinter as tk
from tkinter import filedialog, messagebox
import pandas as pd
import numpy as np
from pathlib import Path
import threading
import os
from os.path import expanduser
from datetime import datetime

class ExcelToHDF5Converter:
    def __init__(self, root):
        self.root = root
        self.root.title("Excel转HDF5")
        self.folder_path = ""
        self.setup_ui()

    def setup_ui(self):
        self.center_window(350, 180)
        # set window
        root.configure(background='#f0f0f0')
        
        self.convert_button = tk.Button(self.root, text="Select a Excel File", width=30, command=self.start_convert, padx=10, pady=5)
        self.convert_button.pack(pady=20)

        self.open_path_button = tk.Button(self.root, text="open folder", width=30, command=self.open_path, padx=10, pady=5)
        self.open_path_button.pack(pady=20)
        self.open_path_button.config(state=tk.DISABLED)

    def center_window(self, width, height):
        screen_width = self.root.winfo_screenwidth()
        screen_height = self.root.winfo_screenheight()
        x = (screen_width / 2) - (width / 2)
        y = (screen_height / 3) - (height / 2)
        self.root.geometry('%dx%d+%d+%d' % (width, height, x, y))

    def start_convert(self):
        file_path = filedialog.askopenfilename(filetypes=[("Excel文件", "*.xlsx *.xls")])
        if not file_path:
            # messagebox.showinfo("message", "select excel file")  
            return
        
        self.convert_button.config(text="Converting, please wait...", state=tk.DISABLED)
        threading.Thread(target=self.convert_to_hdf5, args=(file_path,)).start()

    def convert_to_hdf5(self, file_path):
        error_message = ''  # error_message

        downloads_path = os.path.join(expanduser("~"), 'Downloads')
        current_time = datetime.now().strftime("%Y%m%d%H%M%S")

        original_path = Path(file_path)
        new_filename = f"{original_path.stem}-{current_time}"
        hdf5_filename = Path(new_filename).with_suffix('.h5')
        hdf5_path = Path(downloads_path) / hdf5_filename

        try:
            xls = pd.ExcelFile(file_path)
            with pd.HDFStore(hdf5_path, 'w') as hdf:
                for sheet_name in xls.sheet_names:
                    df = pd.read_excel(xls, sheet_name=sheet_name, dtype=str)
                    df.fillna("", inplace=True)  # replace NaN to ""
                    df = df.astype(str)  
                    safe_sheet_name = sheet_name.replace(' ', '_').replace('-', '_')
                    if df.shape[0] < 3:  # if data line <3，pass to next sheet
                        print(f"Skipping empty sheet: {sheet_name}")
                        continue
                    print(sheet_name)  
                    print(df)

                    hdf.put(safe_sheet_name, df, data_columns=False)

                self.folder_path = downloads_path
                self.root.after(0, lambda: self.open_path_button.config(text="Conversion successful, open folder", state=tk.NORMAL))

        except Exception as e:
            error_message = f"Conversion failed: {e}"

        finally:
            if error_message:
                self.root.after(0, lambda: messagebox.showerror("Error", error_message))
                print(error_message)
            self.root.after(0, lambda: self.convert_button.config(text="Select a Excel File", state=tk.NORMAL))
            
    def open_path(self):
        folder_path = self.folder_path
        if folder_path:
            try:
                os.startfile(folder_path)
            except Exception as e:
                messagebox.showerror("Error", f"Failed to open folder: {e}")

if __name__ == "__main__":
    root = tk.Tk()
    app = ExcelToHDF5Converter(root)
    root.mainloop()

解决思路与方案 #

这个错误通常和HDF5相关依赖的动态库未被PyInstaller正确捕获有关，以下是针对性的解决步骤：

1. 显式添加HDF5相关的hidden-import #

PyInstaller经常会遗漏pandas读写HDF5的底层依赖，打包时添加以下参数：

pyinstaller --hidden-import=h5py --hidden-import=h5py._hl.files --hidden-import=pandas.io.hdf --hidden-import=tables your_script.py

其中tables是pandas处理HDF5的核心依赖库，大概率是被遗漏的关键项。

2. 手动复制缺失的动态库 #

找到Python环境中HDF5相关的动态库文件（Windows下为.dll，路径通常在Lib/site-packages/h5py或Lib/site-packages/tables目录），打包完成后将这些文件复制到exe所在的目录中。

3. 修改HDFStore初始化方式 #

尝试显式指定HDF5驱动，避免打包后驱动加载异常：

with pd.HDFStore(hdf5_path, 'w', driver='H5FD_SEC2') as hdf:
    # 原有代码逻辑

4. 排查路径问题（若使用--onefile模式） #

如果用--onefile打包，PyInstaller会在临时目录解压文件，可能导致依赖路径异常。可以：

• 改用--onedir模式打包（更易排查缺失文件）
• 或在代码中添加路径适配逻辑：

  import sys
  def resource_path(relative_path):
      if hasattr(sys, '_MEIPASS'):
          return os.path.join(sys._MEIPASS, relative_path)
      return os.path.join(os.path.abspath("."), relative_path)
  

5. 升级依赖版本 #

可能是版本兼容性问题，升级相关库到最新稳定版：

pip install --upgrade pandas h5py tables pyinstaller

6. 打印完整错误栈定位问题 #

修改代码中的异常捕获逻辑，打印完整traceback，精准定位错误发生位置：

import traceback
# ...
except Exception as e:
    error_message = f"Conversion failed: {e}\n{traceback.format_exc()}"

内容的提问来源于stack exchange，提问作者JUNYAN WANG