from docxtpl import DocxTemplate
import re
import mammoth

def extract_placeholders(template_path):
    """
    從Word模板中提取所有占位符
    """
    # 轉(zhuǎn)換 Word 文檔為 HTML
    with open(template_path, "rb") as docx_file:
        # 轉(zhuǎn)換為Markdown（相比HTML更易于處理文本）
        result_markdown = mammoth.convert_to_markdown(docx_file)
        markdown_content = result_markdown.value
        
        # 使用正則表達式提取所有占位符
        placeholders = re.compile(r"\{.*?\}").findall(markdown_content)
        
        # 清理占位符，去除花括號
        cleaned_placeholders = [i.replace("{", "").replace("}", "") for i in placeholders]
        
        # 去重并返回
        return list(set(cleaned_placeholders))

# 測試提取功能
template_path = "1-行政許可受理通知書模版.docx"
placeholders = extract_placeholders(template_path)
print("提取到的占位符:", placeholders)

def generate_placeholder_descriptions(placeholders):
    """
    為占位符生成中文描述（模擬LLM功能）
    實際應(yīng)用中可接入OpenAI GPT等大模型
    """
    # 此處簡化處理，實際應(yīng)用中可調(diào)用LLM API
    description_mapping = {}
    for placeholder in placeholders:
        # 根據(jù)占位符名稱推測其含義（實際應(yīng)用中使用LLM增強準確性）
        if 'info0' in placeholder:
            description_mapping[placeholder] = "序列編號"
        elif 'info1' in placeholder:
            description_mapping[placeholder] = "日期信息"
        elif 'info2' in placeholder:
            description_mapping[placeholder] = "申請單位/個人名稱"
        else:
            description_mapping[placeholder] = f"未知參數(shù){placeholder}"
    
    return description_mapping

def get_user_input(placeholder_descriptions):
    """
    根據(jù)占位符描述獲取用戶輸入
    """
    context = {}
    for placeholder, description in placeholder_descriptions.items():
        user_input = input(f"請輸入{description} ({placeholder}): ")
        context[placeholder] = user_input
    
    return context

# 生成占位符描述
placeholder_descriptions = generate_placeholder_descriptions(placeholders)
print("占位符描述映射:", placeholder_descriptions)

# 根據(jù)描述獲取用戶輸入（實際應(yīng)用中可從數(shù)據(jù)庫/API獲取）
context = get_user_input(placeholder_descriptions)
print("填充數(shù)據(jù):", context)

def render_template(template_path, context, output_path):
    """
    使用提供的上下文數(shù)據(jù)渲染W(wǎng)ord模板
    """
    # 加載模板文件
    doc = DocxTemplate(template_path)
    
    # 渲染模板
    doc.render(context)
    
    # 保存生成的文檔
    doc.save(output_path)
    print(f"文檔已生成: {output_path}")

# 渲染模板（使用前一步獲取的context數(shù)據(jù)）
render_template("1-行政許可受理通知書模版.docx", context, "output.docx")

from docxtpl import DocxTemplate, InlineImage
from docx.shared import Mm

def advanced_template_rendering():
    """
    高級模板渲染示例：包含循環(huán)、條件和圖片
    """
    # 加載模板
    doc = DocxTemplate("advanced_template.docx")
    
    # 準備包含復雜結(jié)構(gòu)的上下文數(shù)據(jù)
    context = {
        'company_name': 'ABC科技有限公司',
        'current_date': '2025-10-20',
        'department': '技術(shù)部',
        'employees': [
            {'name': '張三', 'position': '工程師', 'work_years': 3},
            {'name': '李四', 'position': '設(shè)計師', 'work_years': 5},
            {'name': '王五', 'position': '項目經(jīng)理', 'work_years': 7}
        ],
        'has_bonus': True,
        'bonus_amount': 5000,
        'company_logo': InlineImage(doc, 'logo.png', width=Mm(30))
    }
    
    # 渲染并保存
    doc.render(context)
    doc.save("advanced_output.docx")

# 執(zhí)行高級渲染示例
# advanced_template_rendering()