#!/usr/bin/env python3
"""crawl_ahyeji_gsgg.py - 六安叶集区-政府信息公开栏目 (LS CMS Ls.ajax接口)"""
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 = '叶集区-通知公告'
HEADERS = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) Chrome/120.0',
           'X-Requested-With': 'XMLHttpRequest',
           'Referer': 'https://www.ahyeji.gov.cn/public/column/6599521?type=4&catId=7260078&action=list&nav=3'}
API = 'https://www.ahyeji.gov.cn/site/label/8888'
ORGAN_ID = '6599521'
CAT_ID = '7260078'
SITE_ID = '6789301'
FILE = '/c2/yj/publicInfoList_newest2020'

def fetch_list(page_idx):
    data = {
        'labelName': 'publicInfoList', 'siteId': SITE_ID, 'organId': ORGAN_ID,
        'pageSize': 15, 'catId': CAT_ID, 'type': 4,
        'dateFormat': 'yyyy年MM月dd日', 'pdfAndWord': 'false', 'file': FILE,
    }
    if page_idx > 1:
        data['pageIndex'] = page_idx
    try:
        r = requests.post(API, timeout=(5, 15), headers=HEADERS, data=data, verify=False)
        r.encoding = 'utf-8'
    except:
        return []
    # 表格: <tr><td>序号</td><td><a href="https://www.ahyeji.gov.cn/public/6599521/ID.html">标题</a></td><td></td><td>2026年05月13日</td></tr>
    items = re.findall(
        r'<a href="([^"]+)"[^>]*>([^<]{4,120})</a>.*?(\d{4})年(\d{2})月(\d{2})日',
        r.text, re.DOTALL
    )
    results = []
    for href, title, y, m, d in items:
        title = re.sub(r'\s+', '', title)
        if len(title) < 4:
            continue
        results.append({'url': href, 'pub_date': f'{y}-{m}-{d}', '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'<h1[^>]*>(.*?)</h1>', r.text, re.DOTALL)
    if m:
        cand = re.sub(r'<[^>]+>', '', m.group(1)).strip()
        if cand and '404' not in cand:
            title = cand
    if not title:
        m = re.search(r'<title>([^<]*)</title>', r.text, re.DOTALL)
        if m:
            title = re.sub(r'<[^>]+>', '', m.group(1)).strip()
    title = re.sub(r'\s+', '', title).strip()
    content = ''
    # 正文容器 gkwz_contnet j-fontContent（含扫描图片），以附件区 xxgk_zclist 为终止锚点
    m = re.search(
        r'<div class="gkwz_contnet[^"]*"[^>]*>(.*?)</div>\s*<div class="xxgk_zclist"',
        r.text, re.DOTALL)
    if not m:
        m = re.search(r'<div class="gkwz_contnet[^"]*"[^>]*>(.*?)</div>', r.text, re.DOTALL)
    if m:
        content = m.group(1).strip()
    if not content:
        m = re.search(r'class="clearfix xxgkcontent minh500"[^>]*>(.*?)</div>', r.text, re.DOTALL)
        if m:
            content = m.group(1).strip()
            # 截掉尾部附件区（xxgk_zclist 起）
            idx = content.find('<div class="xxgk_zclist"')
            if idx > 0:
                content = content[:idx]
    # 意见征集模板 (URL /content/article/): 正文在 zjnr_content（征集公告 tab）
    if not content:
        m = re.search(r'class="zjnr_content"[^>]*>(.*?)</div>', r.text, re.DOTALL)
        if m:
            content = m.group(1).strip()
    # img 相对路径 → 完整域名（排除协议相对 // 与 http 开头）
    content = re.sub(r'src="(/[^"]+)"', r'src="https://www.ahyeji.gov.cn\1"', content)
    return title, content

def run(max_pages=5):
    records = []
    seen = set()
    for pg in range(max_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'], '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, "ahyeji_gsgg")
    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")