使用Python从Outlook提取员工信息突然变慢的技术求助

我之前用Python的pywin32库从指定Outlook通讯组列表提取成员的姓名、邮箱、职位等信息，100-200个成员仅需1-2分钟。但更换公司笔记本后，相同任务耗时长达1小时，网上未找到相关解决方案，求指导。

原代码示例 #

import win32com.client
import pandas as pd
import progressbar as pb

outApp = win32com.client.gencache.EnsureDispatch('Outlook.Application').GetNamespace("MAPI")
entries = outApp.AddressLists
dist_lists = entries['All Distribution Lists']

outlookdf=pd.DataFrame(columns=['User Name','Email','Job Level','Location','Department','DL Name'])

dl_names_list = ["<insert name of distribution list here>"]

print("Distribution Lists being extracted are: ")
for i in dl_names_list:
    print(str(i))
print("----------------------------")

#########GET DL Names from list and put it in a dataframe

for i in dl_names_list:
    print("Distribution List now being extracted: ")
    print(str(i))

    #initialize widgets
    widgets = ['Remaining Time: ', pb.Percentage(), ' ', 
                pb.Bar(marker=pb.RotatingMarker()), ' ', pb.ETA()]
    #initialize timer
    timer = pb.ProgressBar(widgets=widgets,        maxval=len(dist_lists.AddressEntries.Item(str(i)).GetExchangeDistributionList().Members)).start()
    ind = 0
    m_list_1 = []
    m_list_2 = []
    m_list_3 = []
    m_list_4 = []
    m_list_5 = []
    for m in dist_lists.AddressEntries.Item(str(i)).GetExchangeDistributionList().Members:
            user=m.GetExchangeUser()
            try:
                if len(user.Name) > 0 and (user.Name.find(', ') != -1):
                        value1 = user.Name
                        m_list_1.append(value1)

                        value2 = user.PrimarySmtpAddress  
                        m_list_2.append(value2)

                        value3 = user.JobTitle 
                        m_list_3.append(value3)
                        
                        value4 = user.OfficeLocation
                        m_list_4.append(value4)
                        
                        value5 = user.Department
                        m_list_5.append(value5)
                timer.update(ind)
                ind = ind + 1
             except:
                continue
    timer.finish()
    print("---------------------------------------------------")

    df_m = pd.DataFrame(pd.DataFrame(
    {'User Name': m_list_1,
     'Email': m_list_2,
     'Job Level': m_list_3,
     'Location': m_list_4,
     'Department': m_list_5
    }))
    df_m["DL Name"] = str(dist_lists.AddressEntries.Item(str(i)).GetExchangeDistributionList().Name)
    outlookdf = pd.concat([outlookdf,df_m])

outlookdf = outlookdf.drop_duplicates()
outlookdf = outlookdf.reset_index(drop = True)

针对性优化建议 #

• 缓存重复调用的COM对象：原代码多次重复调用dist_lists.AddressEntries.Item(str(i)).GetExchangeDistributionList()，每次都会触发Outlook接口交互，是核心耗时点。提前缓存对象可大幅减少交互次数：

  # 循环内提前缓存通讯组和成员对象
  dl = dist_lists.AddressEntries.Item(str(i)).GetExchangeDistributionList()
  members = dl.Members
  timer = pb.ProgressBar(widgets=widgets, maxval=len(members)).start()
  

• 简化DataFrame构建逻辑：用字典列表直接构建DataFrame，减少中间列表的创建与维护开销，同时避免重复拼接DataFrame：

  member_data = []
  for m in members:
      user = m.GetExchangeUser()
      try:
          if user.Name and ', ' in user.Name:
              member_data.append({
                  'User Name': user.Name,
                  'Email': user.PrimarySmtpAddress,
                  'Job Level': user.JobTitle,
                  'Location': user.OfficeLocation,
                  'Department': user.Department,
                  'DL Name': dl.Name
              })
          ind += 1
          timer.update(ind)
      except:
          continue
  df_m = pd.DataFrame(member_data)
  

• 调整Outlook运行模式：新笔记本可能开启了Outlook缓存模式，同步大量本地数据导致接口响应延迟。可切换至在线模式测试：打开Outlook→文件→账户设置→账户设置→双击Exchange账户→取消勾选“使用缓存Exchange模式”。

• 检查版本兼容性：确认新笔记本的pywin32版本、Outlook版本与旧电脑是否一致，尝试更新pywin32至最新稳定版，或回退到旧电脑的匹配版本。

• 优化进度条更新逻辑：将timer.update(ind)移至try块外，避免条件判断带来的额外开销，确保每次循环仅执行一次进度更新。

内容的提问来源于stack exchange，提问作者Samrat