#!/usr/bin/env python3
"""crawl_jiangshan_hpsp.py - 江山市-环评信息公示

站点: https://www.jiangshan.gov.cn/col/col1229857236/index.html?uid=9057648
CMS: 浙江 JPAAS (unitbuild.js, webId=1839, 同扬州体系)
列表: /api-gateway/jpaas-publish-server/front/page/build/unit?parseType=bulidstatic&webId=1839
      &tplSetId=DdaSIvZkD2fqQMzw19yIN&pageType=column&tagId=文章列表&pageId=1229857236
      &paramJson={"pageNo":N,"pageSize":15} → JSON data.html
      ⚠️ 两种链接格式: 新 art_<32hex>.html (/col/col.../art/YYYY/) + 旧 /art/YYYY/M/D/art_<colid>_<numid>.html
      ⚠️ P2 稀疏(4条) 后 P3 恢复 15 条(旧格式) — 混合分页, 用 pageNo 循环+去重
详情: meta ArticleTitle + meta PubDate + div.article_text 正文(平衡div)
⚠️ WAF: 须 Chrome UA + Referer, 否则 403
"""
import sys, os, re, json, argparse, urllib.parse
import requests

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 = '江山市-环评信息公示'
GROUP_NAME = '浙江'
BASE = 'https://www.jiangshan.gov.cn'
API_URL = BASE + '/api-gateway/jpaas-publish-server/front/page/build/unit'
PAGE_ID = '1229857236'
WEB_ID = '1839'
TPL_ID = 'DdaSIvZkD2fqQMzw19yIN'
PAGE_SIZE = 15
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': BASE + '/col/col1229857236/index.html?uid=9057648',
}


def fetch(url):
    try:
        r = requests.get(url, timeout=(5, 25), 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 fetch_api(page_no):
    """JPAAS API 列表"""
    params = {
        'parseType': 'bulidstatic', 'webId': WEB_ID, 'tplSetId': TPL_ID,
        'pageType': 'column', 'tagId': '文章列表', 'editType': 'null',
        'pageId': PAGE_ID,
        'paramJson': json.dumps({'pageNo': page_no, 'pageSize': PAGE_SIZE}, ensure_ascii=False),
    }
    url = API_URL + '?' + urllib.parse.urlencode(params)
    try:
        r = requests.get(url, timeout=(5, 25), headers=HEADERS, verify=False)
        data = r.json()
        return data.get('data', {}).get('html', '')
    except Exception as e:
        print(f"[WARN] API fetch failed page {page_no}: {e}", file=sys.stderr)
        return ''


def extract_items(html):
    """从 API HTML 提取条目, 只抓本栏目 (两种链接格式)"""
    items = []
    for m in re.finditer(r'<a\s+href="([^"]+)"[^>]*title="([^"]*)"[^>]*>(.*?)</a>\s*<span>([^<]*)</span>', 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: hex 新格式 或 数字旧格式(art_1229857236_)
        if not (re.search(r'art_[0-9a-f]{32}\.html', href) or re.search(r'art_1229857236_\d+\.html', href)):
            continue
        title = (title_attr or 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 = BASE + href if href.startswith('/') else 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):
    """详情页: meta ArticleTitle + meta PubDate + article_text 正文"""
    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'<h1[^>]*>(.*?)</h1>', html, re.DOTALL)
        if m:
            title = re.sub(r'<[^>]+>', '', m.group(1)).strip()
            title = re.sub(r'\s+', ' ', title).strip()
    title = re.sub(r'^\s*(?:&middot;|&nbsp;|·|\s)+', '', title)
    title = re.sub(r'_(江山市人民政府|江山市|江山市人民政府网站|建设项目环境影响评价信息公示)$', '', title)
    # 日期
    date_text = ''
    m = re.search(r'<meta\s+name="PubDate"\s+content="([^"]+)"', html)
    if m:
        md = re.search(r'(\d{4}-\d{2}-\d{2})', m.group(1))
        if md:
            date_text = md.group(1)
    # 正文: div.article_text
    content = ''
    c = extract_balanced_div(html, r'<div[^>]*class="[^"]*article_text[^"]*"[^>]*>')
    if c.strip():
        content = c.strip()
    if not content.strip():
        c = extract_balanced_div(html, r'<div[^>]*class="[^"]*(?:article-content|content|TRS_Editor|TRS_UEDITOR)[^"]*"[^>]*>')
        if c.strip():
            content = c.strip()
    content = re.sub(r'<script.*?</script>', '', content, flags=re.DOTALL | re.I)
    content = re.sub(r'<style.*?</style>', '', content, flags=re.DOTALL | re.I)
    # 附件/图片相对路径 → 详情页目录 urljoin
    from urllib.parse import urljoin
    content = re.sub(r'href="(?!https?://|//|#|javascript:)([^"]+)"',
                     lambda mm: f'href="{urljoin(url, mm.group(1))}"', content)
    content = re.sub(r'src="(?!https?://|//)([^"]+)"',
                     lambda mm: f'src="{urljoin(url, mm.group(1))}"', content)
    # 清首尾空 p
    content = re.sub(r'^(<p[^>]*>(\s*(<br\s*/?>)?\s*)?</p>)+', '', content)
    content = re.sub(r'(<p[^>]*>(\s*(<br\s*/?>)?\s*)?</p>)+$', '', content)
    return title, date_text, content.strip()


def run(max_pages=5):
    records = []
    seen = set()
    for pg in range(max_pages):
        html = fetch_api(pg + 1)
        if not html:
            print(f"  Page {pg + 1}: 0 items (stop)")
            break
        items = extract_items(html)
        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']
            # 附件提取 (从正文)
            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() or 'fileUrl' 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, "jiangshan_hpsp")
    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")