#!/usr/bin/env python3
"""
宿豫区 - 公示公告 (suyu.gov.cn)
http://www.suyu.gov.cn/suyu/gsgg/xxgk_list.shtml
UCAP CMS, 详情: /suyu/gsgg/202605/uuid.shtml
正文: UCAPCONTENT 或 xxgkcont > article-content
"""
import sys, os, re, time
from datetime import datetime, timezone, timedelta
import requests, urllib3
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
from crawler_lib import push_to_searchdb
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)

SITE_NAME = "宿豫区公示公告"
BASE_URL = "http://www.suyu.gov.cn"
THREE_YEARS_AGO = (datetime.now(timezone.utc) - timedelta(days=365*3)).strftime("%Y-%m-%d")
MAX_PAGES = 5
HEADERS = {"User-Agent": "Mozilla/5.0"}

def fetch(url):
    try:
        r = requests.get(url, headers=HEADERS, timeout=20, verify=False)
        r.encoding = 'utf-8'
        return r.text
    except: return None

def parse_list(html):
    items = []
    pattern = r'<a[^>]*href="(/suyu/gsgg/\d+[^"]*\.shtml)"[^>]*>(.*?)</a>'
    for m in re.finditer(pattern, html, re.DOTALL):
        href = m.group(1)
        title = re.sub(r'<[^>]+>', '', m.group(2)).strip()
        if not title or len(title) < 5: continue
        if 'xxgk_list' in href or 'gsgg' == href.split('/')[-1].replace('.shtml',''):
            continue
        items.append((title, BASE_URL + href))
    return items

def get_page_url(page):
    if page == 1: return "http://www.suyu.gov.cn/suyu/gsgg/xxgk_list.shtml"
    return f"http://www.suyu.gov.cn/suyu/gsgg/xxgk_list_{page}.shtml"

def extract_nested(html, tag_start, tag='div'):
    """找到匹配的闭合标签，支持嵌套div"""
    close_tag = f'</{tag}>'
    open_tag = f'<{tag}'
    depth = 0
    i = tag_start
    n = len(html)
    while i < n:
        if html[i:i+4] == '<!--':
            end = html.find('-->', i+4)
            if end > i: i = end + 3
            else: i += 1
            continue
        if html[i:i+len(close_tag)] == close_tag:
            if depth == 0:
                return html[tag_start:i]
            depth -= 1
            i += len(close_tag)
            continue
        if html[i:i+len(open_tag)] == open_tag and html[i+len(open_tag)] in (' ', '>', '\n', '\t', '\r'):
            depth += 1
            i += len(open_tag)
            continue
        i += 1
    return html[tag_start:]

def fetch_detail(url):
    html = fetch(url)
    if not html: return None, None, None
    result = {}
    # 标题: UCAPTITLE 或 h1.article-title
    m = re.search(r'<UCAPTITLE>(.*?)</UCAPTITLE>', html, re.DOTALL)
    if m: result["title"] = m.group(1).strip()
    if not result.get("title"):
        m = re.search(r'<h1[^>]*class="article-title"[^>]*>(.*?)</h1>', html, re.DOTALL)
        if m: result["title"] = re.sub(r"<[^>]+>", "", m.group(1)).strip()
    if not result.get("title"):
        m = re.search(r'<h1[^>]*>(.*?)</h1>', html, re.DOTALL)
        if m: result["title"] = re.sub(r"<[^>]+>", "", m.group(1)).strip()
    # 日期: publishtime 或 meta PubDate
    m = re.search(r'<publishtime>(\d{4}-\d{1,2}-\d{1,2})</publishtime>', html)
    if m: result["publish_date"] = m.group(1)
    if not result.get("publish_date"):
        m = re.search(r'<meta[^>]*name="PubDate"[^>]*content="(\d{4}-\d{1,2}-\d{1,2})', html)
        if m: result["publish_date"] = m.group(1)
    # 正文: xxgkArticle（含元数据表+正文）, 其次 UCAPCONTENT
    m = re.search(r'class="xxgkArticle"[^>]*>', html)
    if m:
        content = extract_nested(html, m.end()).strip()
        content = re.sub(r"<script[^>]*>.*?</script>", "", content, flags=re.DOTALL|re.I)
        content = re.sub(r"<style[^>]*>.*?</style>", "", content, flags=re.DOTALL|re.I)
        if len(content) > 100:
            result["content"] = content
    if not result.get("content"):
        m = re.search(r'<UCAPCONTENT>(.*?)</UCAPCONTENT>', html, re.DOTALL)
        if m and len(m.group(1).strip()) > 50:
            content = m.group(1).strip()
            result["content"] = content
    if not result.get("content"):
        # 兜底: 常见容器
        for cls in ['xxgkcont', 'article-content', 'content', 'article', 'text']:
            m = re.search(r'<(div|section)[^>]*class="' + cls + r'[^\"]*"[^>]*>(.*?)</\\1>', html, re.DOTALL)
            if m and len(m.group(2)) > 100:
                content = m.group(2).strip()
                content = re.sub(r"<script[^>]*>.*?</script>", "", content, flags=re.DOTALL|re.I)
                result["content"] = content
                break
    return result.get("title"), result.get("content"), result.get("publish_date")

def main():
    print(f"\n{'='*50}\n🏠 {SITE_NAME}\n{'='*50}")
    all_list = []
    for page in range(1, MAX_PAGES+1):
        url = get_page_url(page)
        print(f"\n📄 第 {page} 页...", end=" ", flush=True)
        html = fetch(url)
        if not html: print("❌"); break
        items = parse_list(html)
        if not items: print("0 条"); break
        print(f"✅ {len(items)} 条")
        all_list.extend(items)
    print(f"\n📊 列表总计: {len(all_list)} 条")
    all_items, seen = [], set()
    for i, (title, url) in enumerate(all_list):
        if "项目" not in title: continue
        if url in seen: continue
        seen.add(url)
        print(f"  [{i+1}/{len(all_list)}] {title[:50]}...", end=" ", flush=True)
        dt, content, date = fetch_detail(url)
        if dt: title = dt
        summary = re.sub(r"<[^>]+>", " ", content or "").strip()[:300]
        summary = re.sub(r"\s+", " ", summary)
        all_items.append({
            "site_name": SITE_NAME, "title": title, "url": url,
            "content": content or "", "pub_date": date or "",
            "summary": summary, "tags": SITE_NAME,
        })
        print("✅")
        time.sleep(0.3)
    if all_items:
        push_to_searchdb(all_items, "suyu_gsgg")
    print(f"\n✅ 完成! 共 {len(all_items)} 条")

if __name__ == "__main__":
    t0 = time.time()
    main()
    print(f"⏱ 耗时: {time.time()-t0:.1f}s")