#!/usr/bin/env python3
"""crawl_qdnhp_hjbh.py - 黄平县-环境保护 (贵州黔东南州黄平县人民政府)

站点: https://www.qdnhp.gov.cn/zfxxgk/fdzdgknr/zdmsxx/hjbh_5621994/
CMS: TRS WCM (UTF-8, 服务端渲染) — 同站 crawl_qdnhp.py 通知公告栏目不同
列表: <li><a target="_blank" href=".../YYYYMM/tYYYYMMDD_ID.html" title="标题">标题</a><b>2026-07-06</b></li>
      链接: /zfxxgk/fdzdgknr/zdmsxx/hjbh_5621994/YYYYMM/tYYYYMMDD_ID.html (title 属性完整无截断)
分页: createPageHTML(67, 0, "index", "html", 1000) → index_N.html 66 页 (P67 404), 每页 15 条
详情: meta ArticleTitle + meta PubDate + div#zoom 正文(内含 TRS_Editor, 平衡 div)
"""
import sys, os, requests, re, argparse, json
from datetime import datetime, timedelta

sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
from crawler_lib import push_to_searchdb

CUTOFF = (datetime.now() - timedelta(days=365 * 3)).strftime('%Y-%m-%d')
SITE_NAME = '黄平县-环境保护'
GROUP_NAME = '贵州'
HEADERS = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
    'Referer': 'https://www.qdnhp.gov.cn/',
}
BASE = 'https://www.qdnhp.gov.cn'
COL_DIR = '/zfxxgk/fdzdgknr/zdmsxx/hjbh_5621994'
LIST_URL = BASE + COL_DIR + '/'


def fetch(url):
    try:
        r = requests.get(url, timeout=(5, 20), headers=HEADERS, verify=False)
        r.encoding = 'utf-8'
        return r.text
    except Exception as e:
        print(f"[WARN] fetch failed: {url} - {e}", file=sys.stderr)
        return ''


def extract_items(html):
    """列表条目: tYYYYMMDD_ID.html + title 属性 + b 日期"""
    items = []
    for m in re.finditer(r'<a[^>]+href="([^"]*t\d{8}_\d+\.html)"[^>]*title="([^"]*)"[^>]*>(.*?)</a>\s*<b>([^<]+)</b>', html, re.DOTALL):
        href, title_attr, title_html, date_text = m.group(1), m.group(2), m.group(3), m.group(4).strip()
        # 只抓精准子栏目: hjbh_5621994 路径
        if 'hjbh_5621994' not in href:
            continue
        title = (title_attr or re.sub(r'<[^>]+>', '', title_html)).strip()
        title = title.replace('\u200b', '').strip()
        title = re.sub(r'^\s*(?:&middot;|&nbsp;|·|\s)+', '', title)
        if len(title) < 4:
            continue
        url = href if href.startswith('http') else BASE + href
        md = re.search(r'(\d{4}-\d{2}-\d{2})', date_text)
        pd = md.group(1) if md else ''
        items.append({'url': url, 'pub_date': pd, 'title': title})
    return items


def extract_balanced_div(html, open_tag_re):
    """平衡 div 匹配 (大小写不敏感)"""
    m = re.search(open_tag_re, html, re.I)
    if not m:
        return ''
    gt = html.find('>', m.start())
    if gt == -1:
        return ''
    i = gt + 1
    depth = 1
    for mm in re.finditer(r'<div[\s>]|</div>', html[i:], re.I):
        if mm.group(0).startswith('<div') or mm.group(0).startswith('<DIV'):
            depth += 1
        else:
            depth -= 1
            if depth == 0:
                return html[i:i + mm.start()]
    return html[i:]


def fetch_detail(url):
    html = fetch(url)
    if not html:
        return '', '', ''
    # 标题
    title = ''
    m = re.search(r'<meta\s+name="ArticleTitle"\s+content="([^"]+)"', html)
    if m:
        title = m.group(1).strip()
    if not title:
        m = re.search(r'<title>\s*([^<]+?)\s*</title>', html, re.DOTALL)
        if m:
            title = re.sub(r'\s+', '', m.group(1)).strip()
    title = title.replace('\u200b', '').strip()
    title = re.sub(r'^\s*(?:&middot;|&nbsp;|·|\s)+', '', title)
    # 日期
    date_text = ''
    m = re.search(r'<meta\s+name="PubDate"\s+content="([^"]+)"', html)
    if m:
        md = re.search(r'(\d{4}-\d{2}-\d{2})', m.group(1))
        if md:
            date_text = md.group(1)
    # 正文: div#zoom 优先, div.TRS_Editor 回退
    content = ''
    c = extract_balanced_div(html, r'<div[^>]*id="zoom"[^>]*>')
    if c.strip():
        content = c.strip()
    if not content.strip():
        c = extract_balanced_div(html, r'<div[^>]*class="[^"]*TRS_Editor[^"]*"[^>]*>')
        if c.strip():
            content = c.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)
    # 附件/图片相对路径 → 绝对化
    from urllib.parse import urljoin
    content = re.sub(r'href="(?!https?://|//|#|javascript:)([^"]+)"',
                     lambda mm: f'href="{urljoin(url, mm.group(1))}"', content)
    content = re.sub(r'src="(?!https?://|//)([^"]+)"',
                     lambda mm: f'src="{urljoin(url, mm.group(1))}"', content)
    # 清首尾空 p
    content = re.sub(r'^(<p[^>]*>(\s*(<br\s*/?>)?\s*)?</p>)+', '', content)
    content = re.sub(r'(<p[^>]*>(\s*(<br\s*/?>)?\s*)?</p>)+$', '', content)
    return title, date_text, content.strip()


def run(max_pages=5):
    records = []
    seen = set()
    for pg in range(max_pages):
        url = LIST_URL if pg == 0 else BASE + COL_DIR + f'/index_{pg + 1}.html'
        html = fetch(url)
        if not html:
            print(f"  Page {pg + 1}: fetch fail (stop)")
            break
        items = extract_items(html)
        if not items:
            if len(html) < 2000:
                print(f"  Page {pg + 1}: 404/empty (stop)")
            else:
                print(f"  Page {pg + 1}: 0 items (stop)")
            break
        print(f"  Page {pg + 1}: {len(items)} items")
        for it in items:
            if it['pub_date'] and it['pub_date'] < CUTOFF:
                continue
            if it['url'] in seen:
                continue
            seen.add(it['url'])
            title, date_text, content = fetch_detail(it['url'])
            if not title:
                title = it['title']
            if not content:
                print(f"    SKIP empty content: {it['url']}")
                continue
            pd = date_text or it['pub_date']
            # 附件提取
            atts = []
            for am in re.finditer(r'<a[^>]*href="([^"]+)"[^>]*>(.*?)</a>', content, re.DOTALL):
                h, t = am.group(1), re.sub(r'<[^>]+>', '', am.group(2)).strip()
                if re.search(r'\.(pdf|doc|docx|xls|xlsx|zip|rar|caj|wps)$', h, re.I) or 'download' in h.lower():
                    atts.append({'title': t or '附件', 'url': h})
            records.append({
                'title': title, 'url': it['url'], 'source_url': it['url'],
                'pub_date': pd, 'site_name': SITE_NAME, 'content': content,
                'summary': '', 'attachments': json.dumps(atts, ensure_ascii=False) if atts else '[]',
                'group_name': GROUP_NAME,
            })
    print(f"  Total: {len(records)} items")
    valid = [r for r in records if r['content'].strip()]
    if valid:
        push_to_searchdb(valid, "qdnhp_hjbh")
    return len(valid)


if __name__ == '__main__':
    ap = argparse.ArgumentParser()
    ap.add_argument('--pages', type=int, default=5)
    args = ap.parse_args()
    n = run(args.pages)
    print(f"[RESULT] {SITE_NAME}: {n} items")