#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
crawl_zhuhai_spqgs.py - 珠海市生态环境局-审批前公示
CMS: NFCMS (广东政府网站群)
列表: /zxfw/xmgsgg/spqgs/index.html (page1) → index_{N}.html (pages 2+, 96页)
      div.wendangListC > li > a[href] + date prefix
详情: /zxfw/xmgsgg/spqgs/content/post_XXXXXXX.html
标题: <meta ArticleTitle>
日期: <meta PubDate>
正文: div.xxxq_text_cont > p
"""

import requests
import re
import sys
import os
import time
from bs4 import BeautifulSoup

DB_PATH = "/root/search.db"
BASE_URL = "https://ssthjj.zhuhai.gov.cn"
SITE_NAME = "珠海市-审批前公示"
GROUP = "广东"
INDUSTRY = "环境公示"

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",
}

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_list_page(page_no):
    if page_no == 1:
        url = f"{BASE_URL}/zxfw/xmgsgg/spqgs/index.html"
    else:
        url = f"{BASE_URL}/zxfw/xmgsgg/spqgs/index_{page_no}.html"
    return get(url)


def parse_list(html):
    """Extract (url, title, date) from list page"""
    items = []
    soup = BeautifulSoup(html, "html.parser")

    wl = soup.find("div", class_="wendangListC")
    if not wl:
        return items

    for li in wl.find_all("li"):
        a = li.find("a", href=True)
        if not a:
            continue
        href = a.get("href", "").strip()
        if not href or "post_" not in href:
            continue

        txt = li.get_text(strip=True)
        # Date: YYYY-MM-DD prefix
        date_match = re.search(r"(\d{4}-\d{2}-\d{2})", txt)
        date_str = date_match.group(1) if date_match else ""

        # Title: text without the date prefix
        title = re.sub(r"^\d{4}-\d{2}-\d{2}", "", txt).strip()
        if not title:
            title = a.get_text(strip=True)

        if href.startswith("http"):
            pass  # Full URL already
        elif href.startswith("/"):
            href = BASE_URL + href
        else:
            href = f"{BASE_URL}/{href}"

        items.append((href, title, date_str))

    return 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.xxxq_text_cont
    content = ""
    tc = soup.find("div", class_="xxxq_text_cont")
    if tc:
        parts = []
        for el in tc.children:
            if el.name == "p":
                txt = el.get_text(strip=True)
                if txt:
                    parts.append(txt)
            elif el.name == "table":
                # Preserve table HTML (not convert to text)
                tbl_html = str(el)
                parts.append(tbl_html)
            elif el.name == "div":
                # Check for p/table inside div
                for child in el.children:
                    if child.name == "p":
                        txt = child.get_text(strip=True)
                        if txt:
                            parts.append(txt)
                    elif child.name == "table":
                        parts.append(str(child))
        if parts:
            content = "\n\n".join(parts)
        else:
            content = tc.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_zhuhai_spqgs.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96页)")
    parser.add_argument("--full", action="store_true", help="全量爬取(96页)")
    args = parser.parse_args()

    max_pages = 96 if args.full else args.pages
    total_new = 0

    for page_no in range(1, max_pages + 1):
        print(f"[{time.strftime('%H:%M:%S')}] 第{page_no}页...", end=" ", flush=True)
        html = fetch_list_page(page_no)
        if not html:
            print("请求失败，停止")
            break

        items = parse_list(html)
        if not items:
            print("无数据，停止")
            break
        print(f"{len(items)}条", flush=True)

        batch = []
        for href, title, date_str in 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 date_str, content, summary, href))
            time.sleep(0.5)

        saved = save_to_db(batch)
        total_new += saved
        print(f"  入库{saved}/{len(batch)}条 (累计{total_new})", flush=True)
        if not args.full and saved == 0 and len(batch) > 0:
            print("  全部已存在，增量停止")
            break

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


if __name__ == "__main__":
    main()