#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
crawl_xiangfuqu_hjbh.py - 开封祥符区-环境保护
CMS: UCAP (JS渲染), 需Playwright渲染列表
列表: /kfsxfqwz/c00511/pc/list.html (JS渲染, 15条/页, 4页共56条)
      Playwright点击翻页: a.num[2], a.num[3]
详情: /kfsxfqwz/c00511/pc/content/content_{id}.html (可直接HTTP访问)
标题: <meta ArticleTitle>
日期: <meta PubDate>
正文: div.article-content > p
"""

import requests
import re
import sys
import os
import time
from bs4 import BeautifulSoup
from playwright.sync_api import sync_playwright

DB_PATH = "/root/search.db"
BASE_URL = "https://www.xiangfuqu.gov.cn"
SITE_NAME = "祥符区-环境保护"
GROUP = "河南"
INDUSTRY = "环境公示"
MAX_PAGES = 4  # 56条 / 15条每页

HEADERS = {
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
}

SESSION = None


def get(url):
    global SESSION
    if SESSION is None:
        SESSION = requests.Session()
        SESSION.headers.update(HEADERS)
    try:
        r = SESSION.get(url, timeout=60)
        r.encoding = "utf-8"
        return r.text
    except Exception as e:
        print(f"ERR:{e}")
        return None


def fetch_all_list_items(max_pages=MAX_PAGES):
    """Use Playwright to render JS list and get all items across pages"""
    all_items = []
    with sync_playwright() as pw:
        browser = pw.chromium.launch(headless=True)
        page = browser.new_page()
        page.goto(f"{BASE_URL}/kfsxfqwz/c00511/pc/list.html",
                  wait_until="networkidle", timeout=30000)

        for pg in range(1, max_pages + 1):
            if pg > 1:
                # Click page N
                try:
                    page.click(f'a.num:text("{pg}")', timeout=5000)
                    page.wait_for_timeout(2000)
                except:
                    print(f"  无法翻到第{pg}页，停止")
                    break

            # Extract items
            items = page.eval_on_selector_all(
                'a[href*="content_"]',
                'els => els.map(el => ({title: el.textContent.trim(), href: el.href}))'
            )
            print(f"  第{pg}页: {len(items)}条", flush=True)
            for item in items:
                if item['href'].startswith(f"{BASE_URL}/kfsxfqwz/c00511/pc/content/"):
                    all_items.append((item['href'], item['title']))

        browser.close()
    return all_items


def extract_detail(html, url):
    soup = BeautifulSoup(html, "html.parser")

    # Title from <meta ArticleTitle>
    title = ""
    meta_title = soup.find("meta", {"name": "ArticleTitle"})
    if meta_title:
        title = meta_title.get("content", "").strip()

    # Date from <meta PubDate>
    date_str = ""
    pubdate_meta = soup.find("meta", {"name": "PubDate"})
    if pubdate_meta:
        content = pubdate_meta.get("content", "")
        m = re.search(r"(\d{4}-\d{2}-\d{2})", content)
        if m:
            date_str = m.group(1)

    # Content from div.article-content
    content = ""
    ac = soup.find("div", class_="article-content")
    if ac:
        parts = []
        for p in ac.find_all("p"):
            txt = p.get_text(strip=True)
            if txt:
                parts.append(txt)
        for tbl in ac.find_all("table"):
            parts.append(str(tbl))
        if parts:
            content = "\n\n".join(parts)
        else:
            content = ac.get_text(separator="\n").strip()

    if content:
        content = re.sub(r"&nbsp;", "", content)

    summary = content[:200] if content else ""
    return title, date_str, content, summary


def save_to_db(items_data):
    import sqlite3
    conn = sqlite3.connect(DB_PATH, timeout=10)
    c = conn.cursor()
    inserted = 0
    fts_batch = []
    for title, date_str, content, summary, page_url in items_data:
        try:
            c.execute("SELECT id FROM gov_raw WHERE page_url = ?", (page_url,))
            if c.fetchone():
                continue
            c.execute(
                """INSERT INTO gov_raw 
                   (page_url, title, site_name, publish_date, content, summary,
                    source_url, category, industry, group_name, script_name)
                   VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)""",
                (page_url, title, SITE_NAME, date_str, content, summary,
                 page_url, "环境公示", INDUSTRY, GROUP, "crawl_xiangfuqu_hjbh.py"),
            )
            new_id = c.lastrowid
            if new_id:
                fts_batch.append((new_id, title, SITE_NAME, summary or ""))
            inserted += 1
        except Exception as e:
            print(f"  [DB ERROR] {title[:30]}: {e}")
    conn.commit()
    if fts_batch:
        for rid, title, site, summary in fts_batch:
            try:
                c.execute(
                    "INSERT OR IGNORE INTO gov_search(rowid, title, site_name, summary) VALUES (?, ?, ?, ?)",
                    (rid, title, site, summary),
                )
            except Exception as e:
                print(f"  [FTS ERROR] id={rid}: {e}")
        conn.commit()
        print(f"  FTS同步: {len(fts_batch)}条")
    conn.close()
    return inserted


def main():
    import argparse
    parser = argparse.ArgumentParser(description=f"{SITE_NAME}爬虫")
    parser.add_argument("--pages", type=int, default=1, help="爬取页数(共4页)")
    parser.add_argument("--full", action="store_true", help="全量爬取(4页)")
    args = parser.parse_args()

    max_pages = MAX_PAGES if args.full else args.pages
    print(f"[{time.strftime('%H:%M:%S')}] Playwright渲染列表...", flush=True)
    all_items = fetch_all_list_items(max_pages)
    if not all_items:
        print("无数据")
        return

    print(f"\n共获取 {len(all_items)} 条链接")
    total_new = 0
    batch = []

    for href, title in all_items:
        detail_html = get(href)
        if not detail_html:
            continue
        dt, dd, content, summary = extract_detail(detail_html, href)
        batch.append((dt or title, dd or "", content, summary, href))
        time.sleep(0.3)

    saved = save_to_db(batch)
    total_new += saved
    print(f"入库{saved}/{len(batch)}条")

    print(f"\n===== 完成 =====")
    print(f"新增入库: {total_new} 条")
    print(f"站点: {SITE_NAME}")


if __name__ == "__main__":
    main()