Python脚本自动化实现Word转PDF全攻略(建议收藏)

import comtypes.client
wdFormatPDF = 17

def doc_to_pdf(input_file_path, output_file_path):
    word = comtypes.client.CreateObject('Word.Application')
    doc = word.Documents.Open(input_file_path)
    doc.SaveAs(output_file_path, FileFormat=wdFormatPDF)
    doc.Close()
    word.Quit()

# 导入COM对象客户端库，用于控制Windows应用程序
import comtypes.client

# 定义Word中PDF格式的常量值
wdFormatPDF = 17

def doc_to_pdf(input_file_path, output_file_path):
    """
    将Word文档转换为PDF格式
    
    参数:
        input_file_path (str): 输入的Word文档路径
        output_file_path (str): 输出的PDF文件路径
    """
    # 创建Word应用程序的COM对象
    word = comtypes.client.CreateObject('Word.Application')
    
    try:
        # 打开指定的Word文档
        doc = word.Documents.Open(input_file_path)
        
        # 将文档另存为PDF格式
        # wdFormatPDF=17 是Word中PDF格式的标识符
        doc.SaveAs(output_file_path, FileFormat=wdFormatPDF)
        
        # 关闭文档
        doc.Close()
    finally:
        # 无论转换是否成功，都要退出Word应用程序
        word.Quit()

import os

def batch_convert(input_folder, output_folder):
    """
    批量转换Word文档为PDF
    
    参数:
        input_folder (str): 包含Word文档的文件夹路径
        output_folder (str): 输出PDF的文件夹路径
    """
    # 确保输出文件夹存在
    if not os.path.exists(output_folder):
        os.makedirs(output_folder)
    
    # 遍历输入文件夹中的所有文件
    for filename in os.listdir(input_folder):
        if filename.endswith('.docx') or filename.endswith('.doc'):
            input_path = os.path.join(input_folder, filename)
            output_path = os.path.join(output_folder, 
                                      os.path.splitext(filename)[0] + '.pdf')
            try:
                doc_to_pdf(input_path, output_path)
                print(f"成功转换: {filename}")
            except Exception as e:
                print(f"转换失败 {filename}: {str(e)}")

# 使用示例
batch_convert('input_docs', 'output_pdfs')

import smtplib
from email.mime.multipart import MIMEMultipart
from email.mime.base import MIMEBase
from email import encoders

def send_pdf_via_email(pdf_path, recipient):
    """
    发送PDF文件作为邮件附件
    
    参数:
        pdf_path (str): PDF文件路径
        recipient (str): 收件人邮箱
    """
    # 创建邮件对象
    msg = MIMEMultipart()
    msg['From'] = 'your_email@example.com'
    msg['To'] = recipient
    msg['Subject'] = '培训材料 - 请查收'
    
    # 添加PDF附件
    with open(pdf_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={os.path.basename(pdf_path)}')
        msg.attach(part)
    
    # 连接SMTP服务器并发送
    with smtplib.SMTP('smtp.example.com') as server:
        server.login('username', 'password')
        server.send_message(msg)

# 结合转换和发送
def convert_and_send(input_doc, recipient):
    output_pdf = os.path.splitext(input_doc)[0] + '.pdf'
    doc_to_pdf(input_doc, output_pdf)
    send_pdf_via_email(output_pdf, recipient)

import os
import smtplib
from email.mime.multipart import MIMEMultipart
from email.mime.base import MIMEBase
from email import encoders

def batch_convert_and_email(input_folder, department_emails):
    """
    批量转换Word文档为PDF并通过邮件发送给指定部门
    
    参数:
        input_folder (str): 包含Word文档的文件夹路径
        department_emails (dict): 部门名称到邮箱列表的映射
    """
    # 确保输出文件夹存在
    output_folder = 'output_pdfs'
    if not os.path.exists(output_folder):
        os.makedirs(output_folder)
    
    # 遍历输入文件夹中的所有文件
    for filename in os.listdir(input_folder):
        if filename.endswith('.docx') or filename.endswith('.doc'):
            input_path = os.path.join(input_folder, filename)
            
            # 从文件名中提取部门信息（假设格式为"部门_培训材料.docx"）
            try:
                department = filename.split('_')[0]
                if department in department_emails:
                    output_path = os.path.join(output_folder, 
                                              os.path.splitext(filename)[0] + '.pdf')
                    
                    # 转换文档
                    doc_to_pdf(input_path, output_path)
                    
                    # 发送给相关部门
                    for email in department_emails[department]:
                        send_pdf_via_email(output_path, email)
            except Exception as e:
                print(f"处理 {filename} 时出错: {str(e)}")

# 部门邮箱映射
departments = {
    'HR': ['hr@company.com', 'manager_hr@company.com'],
    'IT': ['it_team@company.com'],
    'Finance': ['finance@company.com']
}

# 执行批量转换和发送
batch_convert_and_email('training_materials', departments)

import logging

logging.basicConfig(filename='conversion.log', level=logging.INFO)

def doc_to_pdf_robust(input_file_path, output_file_path):
    try:
        word = comtypes.client.CreateObject('Word.Application')
        word.Visible = False  # 后台运行
        doc = word.Documents.Open(input_file_path)
        doc.SaveAs(output_file_path, FileFormat=wdFormatPDF)
        doc.Close()
        word.Quit()
        logging.info(f"成功转换: {input_file_path} -> {output_file_path}")
    except Exception as e:
        logging.error(f"转换失败 {input_file_path}: {str(e)}")
        raise