#!/usr/bin/env python3
"""
满洲里市人民政府 — 生态环境-实施情况 (manzhouli.gov.cn)
===================================================
OpennessTarget CMS
列表: /OpennessContent/showList/645/84182/page_1.html → page_2.html (71页×15条=1064条)
详情: /OpennessContent/show/{id}.html
标题: div.text-center.u-title
日期: meta PubDate (含时间)
正文: div#zoom.g-detailbox
"""
import sys, os, re, time
from datetime import datetime, timezone, timedelta
import requests, urllib3
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
from crawler_lib import push_to_searchdb, clean_html
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)

SITE_NAME = "满洲里市-生态环境实施情况"
BASE_URL = "https://www.manzhouli.gov.cn"
LIST_TPL = "https://www.manzhouli.gov.cn/OpennessContent/showList/645/84182/page_{page}.html"
MAX_PAGES = 5  # pagination: pagecount=71, total=1064
CUTOFF = (datetime.now(timezone.utc) - timedelta(days=365*3)).strftime("%Y-%m-%d")

def fetch(url):
    try:
        r = requests.get(url, timeout=20, verify=False, allow_redirects=True)
        r.encoding = 'utf-8'
        return r.text
    except:
        return None

def parse_list(html):
    """<li><a href="/OpennessContent/show/{id}.html" title="Title">Title</a><span>YYYY-MM-DD</span></li>"""
    items = []
    # 主内容区的li（排除侧边栏的li，取OpennessContent/show/的链接）
    for m in re.finditer(
        r'<li>\s*<a[^>]*href="(/OpennessContent/show/\d+\.html)"[^>]*title="([^"]*)"[^>]*>.*?</a>\s*<span>(\d{4}-\d{2}-\d{2})</span>',
        html, re.DOTALL
    ):
        href = m.group(1).strip()
        title = m.group(2).strip()
        date = m.group(3).strip()
        if not href.startswith('http'):
            href = BASE_URL + href
        if title and date:
            items.append((title, href, date))
    return items

def fetch_detail(url):
    """提取详情页"""
    html = fetch(url)
    if not html:
        return "", "", ""

    # 标题 — div.text-center.u-title
    title = ""
    m = re.search(r'<div[^>]*class="text-center u-title"[^>]*>(.*?)</div>', html, re.DOTALL)
    if m:
        title = re.sub(r'<[^>]+>', '', m.group(1)).strip()

    # 日期 — meta PubDate (含时间，只取日期部分)
    pub_date = ""
    m = re.search(r'<meta[^>]*name="PubDate"[^>]*content="(\d{4}-\d{1,2}-\d{1,2})', html, re.I)
    if m:
        pub_date = m.group(1)
    if not pub_date:
        m = re.search(r'发布时间[：:]\s*(\d{4}-\d{1,2}-\d{1,2})', html)
        if m:
            pub_date = m.group(1)

    # 正文 — div#zoom.g-detailbox
    content = ""
    m = re.search(r'<div[^>]*(?:id="zoom"|class="[^"]*g-detailbox[^"]*")[^>]*>(.*?)</div>\s*</div>', html, re.DOTALL)
    if not m:
        m = re.search(r'id="zoom"[^>]*>(.*?)</div>\s*</div>', html, re.DOTALL)
    if m:
        content = m.group(1).strip()

    if content and len(content) > 50:
        content = clean_html(content)
        content = re.sub(r'>\s+<', '><', content)

    return title, pub_date, content


def run(args=None):
    incremental = False
    if args and '1' in args:
        incremental = True
    print(f"爬取: {SITE_NAME}")
    results = []

    pages_to_crawl = 1 if incremental else MAX_PAGES
    for page in range(1, pages_to_crawl + 1):
        page_url = LIST_TPL.format(page=page)
        page_html = fetch(page_url)
        if not page_html or len(page_html) < 500:
            print(f"  第{page}页: 无内容，停止翻页")
            break

        items = parse_list(page_html)
        if not items:
            print(f"  第{page}页: 0 条")
            if page > 1:
                break
            continue
        print(f"  第{page}页: {len(items)} 条")

        for i, (title, url, date) in enumerate(items):
            print(f"  [{i+1}/{len(items)}] {title[:35]}...", end=" ", flush=True)
            if date < CUTOFF:
                print("⏭ 超时范围")
                continue

            detail_title, pub_date, content = fetch_detail(url)
            if not content or len(content) < 50:
                print("⚠ 无正文")
                continue

            final_title = detail_title or title
            final_date = pub_date or date

            summary = re.sub(r'<[^>]+>', '', content)[:200].strip()
            results.append({
                "site_name": SITE_NAME,
                "title": final_title,
                "url": url,
                "content": content,
                "summary": summary,
                "pub_date": final_date,
                "category": "实施情况",
                "tags": "满洲里,生态环境,实施情况",
            })
            print(f"✅ ({len(content)}字)")
            time.sleep(0.3)

        if incremental:
            break

    if results:
        print(f"\n  入库 {len(results)} 条")
        push_to_searchdb(results, "manzhouli_ssqk")
    else:
        print("  ⏭ 无新数据入库")
    print("完成")


if __name__ == "__main__":
    run(sys.argv[1] if len(sys.argv) > 1 else None)