pst_parser.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
import pypff
import csv
from datetime import datetime, timezone
from pathlib import Path
import re
from bs4 import BeautifulSoup

def clean_html_content(html_content):
    """清理HTML内容，只保留文本"""
    if not html_content:
        return ""
    try:
        # 使用BeautifulSoup解析HTML
        soup = BeautifulSoup(html_content, 'html.parser')
        # 获取纯文本内容
        text = soup.get_text(separator=' ', strip=True)
        # 清理多余的空白字符
        text = re.sub(r'\s+', ' ', text)
        return text
    except Exception as e:
        print(f"清理HTML内容时出错: {str(e)}")
        return html_content

# 添加归一化主题函数，用于邮件线程的归类
def normalize_subject(subject):
    """归一化邮件主题，去除常见前缀，如 're:'、'fw:' 以便识别邮件线程"""
    if not subject:
        return "无主题"
    subject = subject.lower().strip()
    # 去除 "re:", "fw:" 等前缀
    subject = re.sub(r'^(re:|fw:|fwd:)\s*', '', subject)
    return subject

def parse_pst(pst_path, output_csv):
    """解析PST文件并提取关键元数据"""
    pst = pypff.file()
    pst.open(pst_path)
    
    # 使用列表来累积所有邮件记录，用于后续按时间序列排序和归纳
    records = []
    message_counts = {}
    
    def process_folder(folder, folder_path=""):
        # 获取文件夹名称，如果为None则使用空字符串
        folder_name = folder.name if hasattr(folder, 'name') and folder.name else ""
        current_path = f"{folder_path}/{folder_name}" if folder_path else folder_name
        
        # 设定需要排除的文件夹关键字，不处理这些文件夹（例如日历、联系人、通讯录、任务等）
        current_path_lower = current_path.lower()
        exclude_keywords = ['日历', 'calendar', '联系人', '通讯录', '任务']
        if any(keyword in current_path_lower for keyword in exclude_keywords):
            # 跳过该文件夹及其子文件夹
            return
        
        # 判断当前文件夹类型
        folder_type = "收件箱" if "收件箱" in current_path_lower else "发件箱" if "发件箱" in current_path_lower else "其他"
        
        for message in folder.sub_messages:
            try:
                message_id = str(message.identifier)
                
                if message_id in message_counts:
                    message_counts[message_id] += 1
                    print(f"发现重复邮件: ID={message_id}, 主题={message.subject}, 路径={current_path}")
                    continue
                
                message_counts[message_id] = 1
                
                # 尝试获取收件人列表
                recipients = []
                try:
                    # 如果message有收件人数属性，则循环提取
                    if hasattr(message, 'number_of_recipients') and message.number_of_recipients:
                        for i in range(message.number_of_recipients):
                            try:
                                recipient = message.get_recipient(i)
                                # 尝试获取收件人名称；如果没有则使用"未知收件人"
                                recipient_name = getattr(recipient, 'name', None)
                                if not recipient_name:
                                    recipient_name = "未知收件人"
                                recipients.append(recipient_name)
                            except Exception as e:
                                print(f"获取单个收件人时出错: {e}")
                except Exception as e:
                    print(f"获取收件人信息失败: {e}")
                recipients_str = '; '.join(recipients) if recipients else "无收件人"
                
                # 处理邮件
                sender_name = message.sender_name or "未知"
                headers = message.transport_headers or ""
                from_header = [line.partition(": ")[2].strip() for line in headers.split("\r\n") if line.lower().startswith("from:")]
                sender_address = from_header[0].split("<")[-1].split(">")[0] if from_header else "未知发件人"
                
                timestamp = message.client_submit_time
                if timestamp:
                    dt = timestamp.astimezone(timezone.utc).replace(tzinfo=None)
                    formatted_time = dt.strftime('%Y-%m-%d %H:%M:%S')
                else:
                    dt = None
                    formatted_time = "无时间戳"
                
                attachments = []
                for attachment in message.attachments:
                    att_info = {
                        'name': getattr(attachment, 'filename', '未命名'),
                        'size': getattr(attachment, 'size', 0),
                        'type': getattr(attachment, 'content_type', '未知类型')
                    }
                    attachments.append(f"{att_info['name']} ({att_info['type']}, {att_info['size']}字节)")
                
                # 提取并清理正文
                body = message.plain_text_body
                if not body and message.html_body:
                    # 如果纯文本不存在，清理HTML正文
                    body = clean_html_content(message.html_body)
                if not body:
                    body = "无正文"
                else:
                    # 处理编码问题
                    if isinstance(body, bytes):
                        try:
                            body = body.decode('utf-8')
                        except UnicodeDecodeError:
                            try:
                                body = body.decode('gbk')
                            except UnicodeDecodeError:
                                body = body.decode('utf-8', errors='ignore')
                    # 清理正文中的特殊字符和多余空白
                    body = re.sub(r'\r\n|\r|\n', ' ', body)
                    body = re.sub(r'\s+', ' ', body)
                    body = body.strip()
                
                # 归一化主题以便识别邮件线程
                conversation = normalize_subject(message.subject or "无主题")
                
                # 累积邮件记录，增加"收件人"字段
                record = [
                    f"{sender_name} <{sender_address}>",
                    formatted_time,
                    message.subject or "无主题",
                    conversation,
                    len(attachments),
                    '; '.join(attachments) if attachments else '无附件',
                    current_path,
                    folder_type,
                    recipients_str,
                    body
                ]
                records.append((dt, record))
            except Exception as e:
                error_msg = f"邮件ID:{message.identifier} 错误类型:{type(e).__name__} 详细信息:{str(e)}"
                print(error_msg)
                records.append((None, ["错误记录", "", error_msg, "", "", "", current_path, "", ""]))
                continue
        
        # 处理子文件夹
        for subfolder in folder.sub_folders:
            process_folder(subfolder, current_path)
    
    # 从根文件夹开始处理
    process_folder(pst.root_folder)
    
    # 按时间排序记录，缺失时间的记录放在最后
    records.sort(key=lambda x: (x[0] is None, x[0]))
    
    # 写入CSV文件
    with open(output_csv, 'w', newline='', encoding='utf-8') as csvfile:
        writer = csv.writer(csvfile)
        writer.writerow(['发件人', '时间戳', '主题', '邮件线程', '附件数量', '附件信息', '文件夹路径', '文件夹类型', '收件人', '邮件正文'])
        for _, row in records:
            writer.writerow(row)
    
    pst.close()
    print(f"\n解析完成，结果已保存至：{output_csv}")

if __name__ == "__main__":
    input_pst = r"D:\Projects\Email_Audit\V4-\input.pst"
    output_dir = Path(input_pst).parent / "output"
    output_dir.mkdir(exist_ok=True)
    
    output_csv = output_dir / "metadata.csv"
    parse_pst(input_pst, output_csv)