#!/usr/bin/env python3
"""
crawl_ahjx.py — 泾县人民政府·建设项目环境影响评价
=================================================
CMS: 泾县人民政府信息公开平台

列表：/XxgkContent/showList/336/82020/page_N.html (15条/页)
      需要 token_verified=true cookie 绕过反爬
详情：/OpennessContent/show/{id}.html
      标题 <meta name="ArticleTitle">
      日期 <meta name="PubDate">
      正文 <div id="zoom">

用法:
    python3 crawl_ahjx.py               # 全量
    python3 crawl_ahjx.py 1             # 增量（仅最新页）
"""

import os
import re
import sys
import requests
from bs4 import BeautifulSoup
from urllib.parse import urljoin
from datetime import datetime, timedelta
from concurrent.futures import ThreadPoolExecutor, as_completed

DB_PATH = os.getenv("SEARCH_DB", "/root/search.db")
SITE_NAME = "泾县人民政府"
COLUMN_NAME = "建设项目环境影响评价"
BASE_URL = "https://www.ahjx.gov.cn"
LIST_URL = f"{BASE_URL}/XxgkContent/showList/336/82020/page_{{}}.html"

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",
    "Cookie": "token_verified=true",
}

MAX_WORKERS = 20
PAGE_SIZE = 15


def log(msg):
    print(f"[ahjx] {msg}")


def fetch_list_page(page_no):
    """Fetch one list page. Returns list of (url, title, date_str)."""
    url = LIST_URL.format(page_no)
    try:
        resp = requests.get(url, headers=HEADERS, timeout=30)
        resp.encoding = "utf-8"
        html = resp.text
    except Exception as e:
        log(f"  ERROR list page {page_no}: {e}")
        return []

    items = []
    # <li>...<a href="..." title="TITLE">...</a><span>DATE</span></li>
    for m in re.finditer(
        r'<a\s+href="(/OpennessContent/show/\d+\.html)"\s+title="([^"]*)"[^>]*>.*?</a>'
        r'\s*<span>(\d{4}-\d{2}-\d{2})</span>',
        html,
        re.DOTALL,
    ):
        href = m.group(1)
        title = m.group(2).strip()
        date_str = m.group(3)
        full_url = href if href.startswith("http") else urljoin(BASE_URL, href)
        items.append((full_url, title, date_str))

    return items


def fetch_detail(url):
    """Return (content_html, date_str, title) or (None, None, None)."""
    try:
        resp = requests.get(url, headers=HEADERS, timeout=30)
        resp.encoding = "utf-8"
        html = resp.text
    except Exception as e:
        log(f"  ERROR {url}: {e}")
        return None, None, None

    soup = BeautifulSoup(html, "html.parser")

    # Title
    meta_title = soup.find("meta", attrs={"name": "ArticleTitle"})
    title = meta_title["content"].strip() if meta_title and meta_title.get("content") else ""

    # Date
    date_str = ""
    meta_date = soup.find("meta", attrs={"name": "PubDate"})
    if meta_date and meta_date.get("content"):
        m = re.match(r"(\d{4}-\d{2}-\d{2})", meta_date["content"].strip())
        if m:
            date_str = m.group(1)

    # Content: <div id="zoom">
    content_div = soup.find(id="zoom")
    content_html = str(content_div) if content_div else ""

    if not content_html:
        log(f"  WARNING: no content for {url}")
        return None, date_str, title

    return content_html, date_str, title


def fetch_detail_batch(urls):
    results = {}
    with ThreadPoolExecutor(max_workers=MAX_WORKERS) as ex:
        fut_map = {ex.submit(fetch_detail, url): url for url in urls}
        for fut in as_completed(fut_map):
            url = fut_map[fut]
            try:
                results[url] = fut.result()
            except Exception:
                results[url] = (None, None, None)
    return results


def save_to_db(items):
    if not items:
        return 0
    import sqlite3
    conn = sqlite3.connect(DB_PATH, timeout=60)
    c = conn.cursor()
    inserted = 0
    for url, title, date_str, content_html in items:
        try:
            c.execute(
                """INSERT OR REPLACE INTO gov_raw (site_name, page_url, title, publish_date, content, category, script_name) VALUES (?, ?, ?, ?, ?, ?, 'crawl_ahjx.py')""",
                (SITE_NAME, url, title, date_str, content_html, COLUMN_NAME),
            )
            inserted += 1
        except Exception as e:
            log(f"  DB error for {url}: {e}")
    conn.commit()
    conn.close()
    return inserted


def rebuild_fts():
    try:
        import sqlite3
        conn = sqlite3.connect(DB_PATH, timeout=60)
        conn.execute(
            """INSERT OR REPLACE INTO gov_search(rowid, title, site_name, summary)
               SELECT rowid, title, site_name,
                      CASE WHEN length(content) > 200 THEN substr(content, 1, 200) ELSE content END
               FROM gov_raw WHERE category = ?""",
            (COLUMN_NAME,),
        )
        conn.commit()
        conn.close()
    except Exception as e:
        log(f"FTS note: {e}")


def main():
    incremental = len(sys.argv) > 1 and sys.argv[1] == "1"
    three_years_ago = (datetime.now() - timedelta(days=365 * 3)).strftime("%Y-%m-%d")
    log(f"3-year cutoff: {three_years_ago}")

    # Phase 1: collect articles from all list pages
    all_articles = []
    page = 1

    while True:
        items = fetch_list_page(page)
        if not items:
            if page == 1:
                log("No articles found")
                return
            break

        log(f"Page {page}: {len(items)} articles")

        # Filter by date (3 years)
        filtered = [(u, t, d) for u, t, d in items if d and d >= three_years_ago]
        all_articles.extend(filtered)

        if incremental:
            log("Incremental: page 1 only")
            break

        # Check if we've passed the 3-year cutoff
        if filtered and len(filtered) < len(items):
            log(f"Page {page}: some articles before cutoff, stopping")
            break

        page += 1

    log(f"Total articles to process: {len(all_articles)}")

    if not all_articles:
        log("No articles within 3 years")
        return

    # Phase 2: fetch details in parallel
    results = []
    batch_size = MAX_WORKERS * 3
    for batch_start in range(0, len(all_articles), batch_size):
        batch = all_articles[batch_start:batch_start + batch_size]
        batch_urls = [u for u, _, _ in batch]
        log(f"  Details [{batch_start+1}-{batch_start+len(batch)}/{len(all_articles)}]...")
        detail_map = fetch_detail_batch(batch_urls)
        for url, title, date_str in batch:
            content_html, detail_date, detail_title = detail_map.get(url, (None, None, None))
            final_title = detail_title or title
            final_date = detail_date or date_str
            results.append((url, final_title, final_date, content_html or ""))

    # Save
    saved = save_to_db(results)
    log(f"Saved: {saved}/{len(results)}")

    if results:
        dates = sorted([r[2] for r in results if r[2]])
        log(f"Date range: {dates[0] if dates else 'N/A'} ~ {dates[-1] if dates else 'N/A'}")

    rebuild_fts()
    log("Done!")


if __name__ == "__main__":
    main()