#!/usr/bin/env python3
"""crawl_qianxian_hjbh.py - 乾县人民政府-环境保护

站点: https://snqianxian.gov.cn/zfxxgk/fdzdgknr/ggjg/hjbh/
CMS: TRS 系 (咸阳市生态环境局乾县分局公告)
列表: ul > li[rowno] > a[href=./YYYYMM/tYYYYMMDD_ID.html] 标题 + span 日期
分页: 首页(无后缀)=最新页; index_N.html 向后 (每页20条, 30+页)
详情: div.tit 标题 + meta PubDate 日期 + div.trs_editor_view.TRS_UEDITOR 正文(平衡div)
附件: ./P0...doc 相对详情目录 → urljoin
标题: strip ·  实体前缀
"""
import sys, os, re, json, argparse
import requests
from urllib.parse import urljoin

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

CUTOFF = (datetime.now() - timedelta(days=365 * 3)).strftime('%Y-%m-%d')
SITE_NAME = '乾县-环境保护'
GROUP_NAME = '陕西'
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',
}
BASE = 'https://snqianxian.gov.cn'
LIST_DIR = '/zfxxgk/fdzdgknr/ggjg/hjbh/'


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_balanced_div(html, open_tag_re):
    """平衡 div 匹配"""
    m = re.search(open_tag_re, html)
    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:]):
        if mm.group(0).startswith('<div'):
            depth += 1
        else:
            depth -= 1
            if depth == 0:
                return html[i:i + mm.start()]
    return html[i:]


def fetch_list(page_idx):
    """列表页: 0=首页(最新), N=index_N.html"""
    if page_idx == 0:
        url = BASE + LIST_DIR
    else:
        url = BASE + LIST_DIR + f'index_{page_idx}.html'
    html = fetch(url)
    if not html:
        return []
    items = []
    for m in re.finditer(r'<li[^>]*>\s*<a href="([^"]+)"[^>]*>(.*?)</a>\s*<span>\s*([^<]*)', html, re.DOTALL):
        href, title_html, date_text = m.group(1).strip(), m.group(2), m.group(3).strip()
        title = re.sub(r'<[^>]+>', '', title_html).strip()
        title = re.sub(r'\s+', ' ', title).strip()
        # strip &middot;&nbsp; 实体前缀
        title = re.sub(r'^\s*(?:&middot;|&nbsp;|·|\s)+', '', title)
        if len(title) < 4:
            continue
        url = urljoin(BASE + LIST_DIR, 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 fetch_detail(url):
    """详情页: tit 标题 + PubDate + trs_editor_view 正文"""
    html = fetch(url)
    if not html:
        return '', '', ''
    # 标题
    title = ''
    m = re.search(r'<div[^>]*class="[^"]*tit[^"]*"[^>]*>(.*?)</div>', html, re.DOTALL)
    if m:
        title = re.sub(r'<[^>]+>', '', m.group(1)).strip()
        title = re.sub(r'\s+', ' ', title).strip()
    if not title:
        m = re.search(r'<meta\s+name="ArticleTitle"\s+content="([^"]+)"', html)
        if m:
            title = m.group(1).strip()
    title = re.sub(r'^\s*(?:&middot;|&nbsp;|·|\s)+', '', title)
    title = re.sub(r'_(乾县人民政府|乾县|乾县人民政府门户网站)$', '', title)
    # 日期
    date_text = ''
    m = re.search(r'<meta\s+name="PubDate"\s+content="([^"]+)"', html)
    if m:
        md = re.search(r'(\d{4}-\d{2}-\d{2})', m.group(1))
        if md:
            date_text = md.group(1)
    # 正文: trs_editor_view
    content = ''
    c = extract_balanced_div(html, r'<div[^>]*class="[^"]*trs_editor_view[^"]*"[^>]*>')
    if c.strip():
        content = c.strip()
    if not content.strip():
        c = extract_balanced_div(html, r'<div[^>]*class="[^"]*(?:TRS_Editor|TRS_UEDITOR|view)[^"]*"[^>]*>')
        if c.strip():
            content = c.strip()
    content = re.sub(r'<script.*?</script>', '', content, flags=re.DOTALL | re.I)
    content = re.sub(r'<style.*?</style>', '', content, flags=re.DOTALL | re.I)
    # 附件/图片相对路径 → 详情页目录 urljoin
    content = re.sub(r'href="(?!https?://|//|#|javascript:)([^"]+)"',
                     lambda mm: f'href="{urljoin(url, mm.group(1))}"', content)
    content = re.sub(r'src="(?!https?://|//)([^"]+)"',
                     lambda mm: f'src="{urljoin(url, mm.group(1))}"', content)
    # 清首尾空 p
    content = re.sub(r'^(<p[^>]*>(\s*(<br\s*/?>)?\s*)?</p>)+', '', content)
    content = re.sub(r'(<p[^>]*>(\s*(<br\s*/?>)?\s*)?</p>)+$', '', content)
    return title, date_text, content.strip()


def run(max_pages=5):
    records = []
    seen = set()
    for pg in range(max_pages):
        items = fetch_list(pg)
        if not items:
            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 content:
                print(f"    SKIP empty content: {it['url']}")
                continue
            if not title:
                title = it['title']
            pd = date_text or it['pub_date']
            # 附件
            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 'attach' in h.lower() or 'wcm' in h.lower():
                    if h.startswith('http'):
                        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, "qianxian_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")