如何在Python界面中正确显示Outlook邮件的内嵌图片与签名图片？

问题分析 #

你遇到的核心问题是：普通Outlook邮件正文、签名中的内嵌图片无法加载（显示损坏链接），仅全HTML营销邮件的图片正常。现有代码仅处理了Type=6的内嵌附件，但普通邮件的内嵌图片可能不符合这个判断条件，同时存在以下几个问题：

1. 未初始化temp_dir变量，导致图片保存失败
2. 硬编码图片MIME类型为image/jpeg，无法适配png、gif等格式
3. 未处理cid带尖括号（如<image001.png@01D9XXXX>）的情况
4. 普通Outlook邮件的内嵌图片可能不标记为Type=6，需通过PR_ATTACH_CONTENT_ID属性判断

修复步骤 #

1. 初始化临时目录 #

在代码开头添加临时目录创建逻辑，避免路径错误：

temp_dir = tempfile.mkdtemp()

2. 改进内嵌图片识别逻辑 #

不再仅依赖Attachment.Type == 6，而是检查是否存在PR_ATTACH_CONTENT_ID属性（即cid），这样能覆盖更多内嵌图片场景：

def save_inline_images(message):
    inline_images = {}
    attachments = message.Attachments

    for attachment in attachments:
        try:
            # 尝试获取cid，存在则视为内嵌图片
            cid = attachment.PropertyAccessor.GetProperty("http://schemas.microsoft.com/mapi/proptag/0x3712001F")
            if cid:
                # 处理带尖括号的cid
                cid = cid.strip('<>')
                file_path = os.path.join(temp_dir, attachment.FileName)
                attachment.SaveAsFile(file_path)
                
                # 获取图片MIME类型，避免硬编码
                mime_type = attachment.PropertyAccessor.GetProperty("http://schemas.microsoft.com/mapi/proptag/0x370E001F") or "image/jpeg"
                
                with open(file_path, "rb") as f:
                    image_data = f.read()
                    encoded_image = base64.b64encode(image_data).decode("utf-8")
                inline_images[cid] = (mime_type, encoded_image)
        except Exception as e:
            # 忽略无cid的普通附件
            continue

    return inline_images

3. 动态替换图片src #

根据实际MIME类型生成data URI，同时处理HTML中cid带尖括号的情况：

def replace_inline_images(html_body, inline_images):
    soup = BeautifulSoup(html_body, 'html.parser')

    for img in soup.find_all('img'):
        src = img.get('src', '')
        if src.startswith('cid:'):
            cid = src[4:]
            # 处理HTML中cid带尖括号的情况
            cid = cid.strip('<>')
            if cid in inline_images:
                mime_type, encoded_image = inline_images[cid]
                img['src'] = f"data:{mime_type};base64,{encoded_image}"

    return str(soup)

4. 清理临时目录（可选） #

在程序退出时删除临时目录，避免残留文件：

import atexit
atexit.register(lambda: os.rmdir(temp_dir))

完整修复代码 #

import sys
import os
import tempfile
import atexit
from PyQt6.QtWidgets import QApplication, QMainWindow, QVBoxLayout, QWidget, QListWidget, QListWidgetItem, QDialog
from PyQt6.QtWebEngineWidgets import QWebEngineView
import win32com.client
from bs4 import BeautifulSoup
import base64

# 创建临时目录并注册清理
temp_dir = tempfile.mkdtemp()
atexit.register(lambda: os.rmdir(temp_dir))

# Function to fetch emails from Outlook
def fetch_emails():
    try:
        outlook = win32com.client.Dispatch("Outlook.Application").GetNamespace("MAPI")
        inbox = outlook.GetDefaultFolder(6)  # 6 refers to the inbox folder
        messages = inbox.Items
        messages.Sort("[ReceivedTime]", True)

        emails = []
        for i in range(min(10, len(messages))):  # Fetching last 10 emails
            message = messages[i]
            try:
                # Check if HTMLBody is available and not empty
                if hasattr(message, 'HTMLBody') and message.HTMLBody:
                    html_body = str(message.HTMLBody)  # Convert HTMLBody to string
                    inline_images = save_inline_images(message)
                    html_body = replace_inline_images(html_body, inline_images)

                    emails.append({
                        'subject': message.Subject,
                        'html_body': html_body,
                        'datetime_received': message.ReceivedTime,
                    })
                else:
                    emails.append({
                        'subject': message.Subject,
                        'html_body': '<p>This email does not contain HTML content.</p>',
                        'datetime_received': message.ReceivedTime,
                    })
                    print(f"Email {i} ('{message.Subject}') does not have an HTMLBody.")
            except Exception as e:
                print(f"An error occurred while accessing email {i} ('{message.Subject}'): {e}")
                continue
        return emails
    except Exception as e:
        print(f"An error occurred: {e}")
        return []

# Function to save inline images from an email
def save_inline_images(message):
    inline_images = {}
    attachments = message.Attachments

    for attachment in attachments:
        try:
            # 获取cid，存在则视为内嵌图片
            cid = attachment.PropertyAccessor.GetProperty("http://schemas.microsoft.com/mapi/proptag/0x3712001F")
            if cid:
                # 移除cid可能带的尖括号
                cid = cid.strip('<>')
                file_path = os.path.join(temp_dir, attachment.FileName)
                attachment.SaveAsFile(file_path)
                
                # 获取图片的MIME类型，适配多种格式
                mime_type = attachment.PropertyAccessor.GetProperty("http://schemas.microsoft.com/mapi/proptag/0x370E001F") or "image/jpeg"
                
                with open(file_path, "rb") as f:
                    image_data = f.read()
                    encoded_image = base64.b64encode(image_data).decode("utf-8")
                inline_images[cid] = (mime_type, encoded_image)
        except Exception as e:
            # 跳过无cid的普通附件
            continue

    return inline_images

# Function to replace inline images in the HTML content
def replace_inline_images(html_body, inline_images):
    soup = BeautifulSoup(html_body, 'html.parser')

    for img in soup.find_all('img'):
        src = img.get('src', '')
        if src.startswith('cid:'):
            cid = src[4:]
            # 移除HTML中cid的尖括号
            cid = cid.strip('<>')
            if cid in inline_images:
                mime_type, encoded_image = inline_images[cid]
                img['src'] = f"data:{mime_type};base64,{encoded_image}"

    return str(soup)

class EmailViewer(QMainWindow):
    def __init__(self):
        super().__init__()
        self.setWindowTitle("Outlook Email Viewer")
        self.setGeometry(100, 100, 800, 600)

        # Main layout
        main_layout = QVBoxLayout()
        main_widget = QWidget()
        main_widget.setLayout(main_layout)
        self.setCentralWidget(main_widget)

        # Email list
        self.email_list = QListWidget()
        main_layout.addWidget(self.email_list)

        # Fetch and display emails
        self.emails = fetch_emails()
        if not self.emails:
            self.email_list.addItem(QListWidgetItem("No emails to display"))

        for email in self.emails:
            item = QListWidgetItem(f"{email['subject']} ({email['datetime_received']})")
            self.email_list.addItem(item)

        # Connect list item click to email display
        self.email_list.itemClicked.connect(self.open_email_viewer)

    def open_email_viewer(self, item):
        index = self.email_list.row(item)
        email = self.emails[index]

        # Create a new window to display the email
        self.email_window = QDialog(self)
        self.email_window.setWindowTitle(email['subject'])
        self.email_window.setGeometry(150, 150, 800, 600)

        # Layout for the new window
        email_layout = QVBoxLayout()
        self.email_window.setLayout(email_layout)

        # QWebEngineView to display the email content
        email_viewer = QWebEngineView()
        email_layout.addWidget(email_viewer)
        email_viewer.setHtml(email['html_body'])

        self.email_window.exec()

# Run the application
if __name__ == "__main__":
    app = QApplication(sys.argv)
    viewer = EmailViewer()
    viewer.show()
    sys.exit(app.exec())

关键修改说明 #

• 临时目录：创建并自动清理，避免路径错误和文件残留
• 图片识别：通过PR_ATTACH_CONTENT_ID判断内嵌图片，覆盖普通邮件和签名图片场景
• MIME类型动态获取：适配jpeg、png、gif等多种图片格式
• cid处理：移除可能存在的尖括号，确保HTML中的cid和附件cid匹配

内容的提问来源于stack exchange，提问作者Joshua S