pip install pandas openpyxl pdfplumber python-dotenv schedule

import os
import shutil
from pathlib import Path
import pandas as pd
import pdfplumber
from datetime import datetime
import schedule
import smtplib
from email.mime.multipart import MIMEMultipart
from email.mime.text import MIMEText
from email.mime.base import MIMEBase
from email import encoders

# 配置
INVOICE_DIR = Path("./invoices")
REPORT_PATH = Path("./daily_report.xlsx")
ARCHIVE_DIR = Path("./archived")

def extract_invoice_data(pdf_path):
    """从PDF发票中提取关键字段"""
    data = {
        "filename": pdf_path.name,
        "invoice_date": "",
        "amount": 0,
        "seller": ""
    }
    
    with pdfplumber.open(pdf_path) as pdf:
        for page in pdf.pages:
            text = page.extract_text()
            if not text:
                continue
            
            # 提取日期（匹配常见格式）
            for line in text.split('\n'):
                if '日期' in line or 'date' in line.lower():
                    parts = line.split(':')
                    if len(parts) > 1:
                        data["invoice_date"] = parts[-1].strip()
                
                # 提取金额
                if '金额' in line or 'total' in line.lower():
                    parts = line.split(':')
                    if len(parts) > 1:
                        try:
                            data["amount"] = float(parts[-1].strip().replace(',', ''))
                        except ValueError:
                            pass
                
                # 提取卖方
                if '销售方' in line or 'seller' in line.lower():
                    parts = line.split(':')
                    if len(parts) > 1:
                        data["seller"] = parts[-1].strip()
    
    return data

def process_invoices():
    """处理所有发票并生成报表"""
    results = []
    
    for pdf_file in INVOICE_DIR.glob("*.pdf"):
        try:
            data = extract_invoice_data(pdf_file)
            results.append(data)
            
            # 移动到归档目录
            ARCHIVE_DIR.mkdir(exist_ok=True)
            shutil.move(str(pdf_file), str(ARCHIVE_DIR / pdf_file.name))
        except Exception as e:
            results.append({
                "filename": pdf_file.name,
                "invoice_date": "",
                "amount": 0,
                "seller": f"ERROR: {str(e)}"
            })
    
    # 生成Excel报表
    df = pd.DataFrame(results)
    df.to_excel(REPORT_PATH, index=False)
    
    # 计算汇总
    total_amount = df['amount'].sum()
    print(f"今日处理 {len(df)} 张发票，总金额 ¥{total_amount:,.2f}")
    
    return df

def send_report_email(df):
    """发送日报邮件"""
    msg = MIMEMultipart()
    msg['From'] = "your_email@company.com"
    msg['To'] = "team@company.com"
    msg['Subject'] = f"发票日报 - {datetime.now().strftime('%Y-%m-%d')}"
    
    body = f"""
    各位同事，今日发票处理完成。
    
    - 处理数量：{len(df)} 张
    - 总金额：¥{df['amount'].sum():,.2f}
    - 详细报表请见附件
    
    此邮件由RPA自动发送。
    """
    
    msg.attach(MIMEText(body, 'html', 'utf-8'))
    
    # 附加Excel文件
    with open(REPORT_PATH, 'rb') as f:
        part = MIMEBase('application', 'octet-stream')
        part.set_payload(f.read())
        encoders.encode_base64(part)
        part.add_header('Content-Disposition', 
                       f'attachment; filename="{REPORT_PATH.name}"')
        msg.attach(part)
    
    # 发送邮件
    server = smtplib.SMTP('smtp.company.com', 587)
    server.starttls()
    server.login("your_email@company.com", "password")
    server.send_message(msg)
    server.quit()
    print("邮件已发送")

if __name__ == "__main__":
    df = process_invoices()
    send_report_email(df)

schedule.every().day.at("09:00").do(process_invoices)

while True:
    schedule.run_pending()
    import time
    time.sleep(60)

# 编辑crontab
crontab -e

# 添加这一行（每天早上9点执行）
0 9 * * * cd /path/to/project && python automate.py >> logs/automation.log 2>&1