#!/usr/bin/env python3
"""crawl_aysthj_c143398.py - 安远县-生态环境 (ay.gov.cn 重点领域→公共监管→生态环境)

站点: http://www.ay.gov.cn/ayxxxgk/c143398/xxgk_list.shtml (安远县信息公开-生态环境栏目)
CMS: 安远县信息公开集群 (UTF-8, meta ArticleTitle/PubDate)
列表: ul.pageList.newsList > li > a(title, href=/ayxxxgk/sthj20220326/{YYYYMM}/{hash}.shtml) + span.time(YYYY-MM-DD)
分页: createPageHTML('page_div',8,1,'xxgk_list','shtml',152) → xxgk_list_{N}.shtml (共8页, 每页~20条, 152条)
详情: meta ArticleTitle + meta PubDate + div.article-content (正文)
注意: ay.gov.cn=安远县(江西), 非河南安阳; 现有crawl_anyang_sthj.py抓的是zfxxkc指南栏目(仅2条), 本脚本抓c143398生态环境栏目
"""
import sys, os, requests, re, argparse
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 = '安远县-生态环境'
HEADERS = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) Chrome/120.0'}
BASE = 'http://www.ay.gov.cn'
LIST_URL = BASE + '/ayxxxgk/c143398/xxgk_list.shtml'
TOTAL_PAGES = 8

def fetch_list(page_idx):
    url = LIST_URL if page_idx == 1 else LIST_URL.replace('xxgk_list.shtml', f'xxgk_list_{page_idx}.shtml')
    try:
        r = requests.get(url, timeout=(5, 15), headers=HEADERS, verify=False)
        r.encoding = 'utf-8'
    except:
        return []
    items = re.findall(
        r'<li><a href="([^"]+)"[^>]*title="([^"]*)"[^>]*>.*?</a><span class="time">(\d{4}-\d{2}-\d{2})</span></li>',
        r.text, re.DOTALL
    )
    results = []
    for href, title, pd in items:
        title = re.sub(r'\s+', '', title).strip()
        if len(title) < 4:
            continue
        url = href if href.startswith('http') else BASE + href
        results.append({'url': url, 'pub_date': pd, 'title': title})
    return results

def fetch_detail(url):
    for _attempt in range(2):
        try:
            r = requests.get(url, timeout=(5, 15), headers=HEADERS, verify=False)
            r.encoding = 'utf-8'
            break
        except:
            r = None
    if r is None:
        return '', ''
    title = ''
    m = re.search(r'<meta name="ArticleTitle" content="([^"]*)"', r.text)
    if m:
        title = m.group(1).strip()
    if not title:
        m = re.search(r'<title>\s*([^<]+?)\s*</title>', r.text, re.DOTALL)
        if m:
            title = re.sub(r'\s+', '', m.group(1)).strip()
            title = re.split(r'[-–—|]', title)[0].strip()
    content = ''
    m = re.search(r'class="article-content[^"]*"[^>]*>(.*?)</div>', r.text, re.DOTALL)
    if m:
        content = m.group(1).strip()
    if not content.strip():
        m = re.search(r'class="content[^"]*"[^>]*>(.*?)</div>', r.text, re.DOTALL)
        if m:
            content = m.group(1).strip()
    # img 相对路径 → 完整域名
    content = re.sub(r'src="(?!https?://|//)(/[^"]+)"', lambda mm: f'src="{BASE}{mm.group(1)}"', content)
    return title, content

def run(max_pages=5):
    records = []
    seen = set()
    pages = min(max_pages, TOTAL_PAGES)
    for pg in range(pages):
        items = fetch_list(pg + 1)
        if not items:
            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, content = fetch_detail(it['url'])
            if not title:
                title = it['title']
            if not content.strip():
                continue
            records.append({
                'title': title, 'url': it['url'], 'source_url': it['url'],
                'pub_date': it['pub_date'],
                'site_name': SITE_NAME, 'content': content, 'summary': '',
            })
    print(f"  Total: {len(records)} items")
    valid = [r for r in records if r['content'].strip()]
    if valid:
        push_to_searchdb(valid, "aysthj_c143398")
    return len(valid)

if __name__ == '__main__':
    ap = argparse.ArgumentParser()
    ap.add_argument('--pages', type=int, default=5)
    args = ap.parse_args()
    cnt = run(max_pages=args.pages)
    print(f"Done: {cnt} records")