#!/usr/bin/env python3
"""
法律文书模板抓取脚本 - 最高人民法院
抓取所有诉讼文书样式模板
"""

import os
import re
import time
import json
import subprocess
from urllib.parse import urljoin

BASE_URL = "https://www.court.gov.cn"
LISTING_PAGES = [
    # 人民法院制作民事诉讼文书样式（当事人参考类）
    "https://www.court.gov.cn/susongyangshi/66.html",  # 管辖
    "https://www.court.gov.cn/susongyangshi/67.html",  # 回避
    "https://www.court.gov.cn/susongyangshi/68.html",  # 诉讼参加人
    "https://www.court.gov.cn/susongyangshi/69.html",  # 证据
    "https://www.court.gov.cn/susongyangshi/70.html",  # 期间、送达
    "https://www.court.gov.cn/susongyangshi/71.html",  # 调解
    "https://www.court.gov.cn/susongyangshi/72.html",  # 保全和先予执行
    "https://www.court.gov.cn/susongyangshi/73.html",  # 对妨害民事诉讼的强制措施
    "https://www.court.gov.cn/susongyangshi/74.html",  # 诉讼费用
    "https://www.court.gov.cn/susongyangshi/75.html",  # 第一审普通程序
    "https://www.court.gov.cn/susongyangshi/76.html",  # 简易程序
    "https://www.court.gov.cn/susongyangshi/77.html",  # 简易程序中的小额诉讼
    "https://www.court.gov.cn/susongyangshi/78.html",  # 公益诉讼
    "https://www.court.gov.cn/susongyangshi/79.html",  # 第三人撤销之诉
    "https://www.court.gov.cn/susongyangshi/80.html",  # 执行异议之诉
    "https://www.court.gov.cn/susongyangshi/81.html",  # 第二审程序
    "https://www.court.gov.cn/susongyangshi/83.html",  # 选民资格案件
    "https://www.court.gov.cn/susongyangshi/84.html",  # 宣告失踪、宣告死亡案件
    "https://www.court.gov.cn/susongyangshi/85.html",  # 认定公民民事行为能力案件
    "https://www.court.gov.cn/susongyangshi/86.html",  # 认定财产无主案件
    "https://www.court.gov.cn/susongyangshi/87.html",  # 确认调解协议案件
    "https://www.court.gov.cn/susongyangshi/88.html",  # 实现担保物权案件
    "https://www.court.gov.cn/susongyangshi/89.html",  # 监护权特别程序案件
    "https://www.court.gov.cn/susongyangshi/90.html",  # 确认仲裁协议效力案件
    "https://www.court.gov.cn/susongyangshi/91.html",  # 撤销仲裁裁决案件
    "https://www.court.gov.cn/susongyangshi/92.html",  # 人身安全保护令案件
    "https://www.court.gov.cn/susongyangshi/93.html",  # 其他非诉程序
    "https://www.court.gov.cn/susongyangshi/94.html",  # 审判监督程序
    "https://www.court.gov.cn/susongyangshi/95.html",  # 督促程序
    "https://www.court.gov.cn/susongyangshi/96.html",  # 公示催告程序
    "https://www.court.gov.cn/susongyangshi/97.html",  # 执行程序
    "https://www.court.gov.cn/susongyangshi/98.html",  # 涉外民事诉讼程序的特别规定
    
    # 行政文书样式
    "https://www.court.gov.cn/susongyangshi/102.html",
    
    # 国家赔偿文书样式
    "https://www.court.gov.cn/susongyangshi/100.html",  # 本院赔偿文书样式
    "https://www.court.gov.cn/susongyangshi/101.html",  # 赔偿委员会文书样式
]

def get_html(url):
    """使用curl获取页面HTML"""
    try:
        result = subprocess.run(
            ['curl', '-s', '-L', '-A', 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36', url],
            capture_output=True, text=True, timeout=30
        )
        return result.stdout
    except Exception as e:
        print(f"Error fetching {url}: {e}")
        return ""

def extract_template_urls(html):
    """从列表页面提取模板详情页URL"""
    pattern = r'href="(/susongyangshi/xiangqing/\d+\.html)"[^>]*>([^<]+)</a>'
    matches = re.findall(pattern, html)
    urls = []
    for url, title in matches:
        urls.append((url, title.strip()))
    return urls

def get_all_pages_for_listing(base_list_url):
    """获取列表的所有分页"""
    pages = [base_list_url]
    html = get_html(base_list_url)
    
    # 查找分页链接模式 like 8_2.html, 8_3.html
    page_pattern = re.compile(r'href="(/susongyangshi/\d+_\d+\.html)"')
    page_urls = set(page_pattern.findall(html))
    
    # 提取基础页码（如 8_2 中的 8）
    base_num_pattern = re.compile(r'/(\d+)_\d+\.html')
    for url in page_urls:
        match = base_num_pattern.search(url)
        if match:
            page_num = int(match.group(1))
            # 添加所有分页
            for i in range(2, 100):  # 假设最多99页
                next_page = f"/susongyangshi/{page_num}_{i}.html"
                next_html = get_html(BASE_URL + next_page)
                if "找不到" in next_html or not next_html:
                    break
                pages.append(BASE_URL + next_page)
    
    return pages

def main():
    print("=" * 60)
    print("法律文书模板抓取开始")
    print("=" * 60)
    
    all_templates = {}
    
    for list_url in LISTING_PAGES:
        print(f"\n正在处理: {list_url}")
        html = get_html(list_url)
        
        # 提取当前页的模板
        templates = extract_template_urls(html)
        print(f"  发现 {len(templates)} 个模板")
        
        for url, title in templates:
            if url not in all_templates:
                all_templates[url] = title
        
        # 处理分页
        page_pattern = re.compile(r'/(\d+_\d+)\.html"')
        page_nums = set(page_pattern.findall(html))
        
        for page_str in page_nums:
            base_num = page_str.split('_')[0]
            for i in range(2, 100):
                next_url = f"{list_url.rsplit('/', 1)[0]}/{base_num}_{i}.html"
                next_html = get_html(next_url)
                if "找不到" in next_html or len(next_html) < 1000:
                    break
                templates = extract_template_urls(next_html)
                for url, title in templates:
                    if url not in all_templates:
                        all_templates[url] = title
                time.sleep(0.5)
    
    print(f"\n总共找到 {len(all_templates)} 个模板")
    
    # 保存URL列表
    with open('/root/.openclaw/workspace/知识库/法律文书模板库/全部模板URL列表.json', 'w', encoding='utf-8') as f:
        json.dump(all_templates, f, ensure_ascii=False, indent=2)
    
    print("URL列表已保存")
    return all_templates

if __name__ == "__main__":
    main()