#!/usr/bin/env python3
"""
Crawler for 轮台县人民政府（金色轮台）- 生态环境
CMS: TRS WCM / 自定义 (same as xjkel)
Site: www.xjlt.gov.cn/xjltx/c114898/
15 items/page. Pagination: zfxxgk.shtml (p1) → zfxxgk_{N}.shtml (p2-p5)
"""
import requests, re, json, sys, os
from datetime import datetime, timedelta
from bs4 import BeautifulSoup
from urllib.parse import urljoin

BASE_URL = "http://www.xjlt.gov.cn"
LIST_PATH = "/xjltx/c114898"
LIST_URL = BASE_URL + LIST_PATH + "/zfxxgk.shtml"
SITE_NAME = "轮台县人民政府-生态环境"
INCREMENTAL_DAYS = 7
MAX_PAGES = 5
PER_PAGE = 15

HEADERS = {
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36",
    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
    "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
}
TIMEOUT = 20

session = requests.Session()
session.headers.update(HEADERS)


def fetch(url):
    resp = session.get(url, timeout=TIMEOUT)
    resp.encoding = "utf-8"
    return resp.text


def parse_list(html):
    soup = BeautifulSoup(html, "html.parser")
    items = []
    lb = soup.find("div", class_="listBox")
    if not lb:
        return items
    for li in lb.find_all("li"):
        a = li.find("a", href=True)
        if not a:
            continue
        title = a.get("title") or a.get_text(strip=True)
        href = a["href"].strip()
        if not title or not href:
            continue
        date_span = li.find("span")
        date = date_span.get_text(strip=True) if date_span else ""
        if href.startswith("/"):
            href = BASE_URL + href
        elif not href.startswith("http"):
            href = BASE_URL + LIST_PATH + "/" + href.lstrip("./")
        items.append((title.strip(), href, date[:10]))
    return items


def parse_detail(html, url):
    soup = BeautifulSoup(html, "html.parser")
    title = ""
    meta_title = soup.find("meta", attrs={"name": "ArticleTitle"})
    if meta_title and meta_title.get("content"):
        title = meta_title["content"].strip()
    if not title:
        h1 = soup.find("h1")
        if h1:
            title = h1.get_text(strip=True)
    pub_date = ""
    meta_date = soup.find("meta", attrs={"name": "PubDate"})
    if meta_date and meta_date.get("content"):
        m = re.search(r"(\d{4}-\d{2}-\d{2})", meta_date["content"])
        if m:
            pub_date = m.group(1)
    content_html = ""
    content_div = soup.find("div", class_="content")
    if content_div:
        parts = []
        for child in content_div.children:
            if not getattr(child, "name", None):
                continue
            tag = child.name.lower()
            if tag == "p":
                # Preserve <a> links inside <p> as markdown [text](url)
                p_parts = []
                for elem in child.contents:
                    if isinstance(elem, str):
                        txt = elem.strip()
                        if txt:
                            p_parts.append(txt)
                    elif elem.name == "a":
                        ahref = elem.get("href", "").strip()
                        atxt = elem.get_text(strip=True)
                        if ahref and atxt:
                            full_href = urljoin(url, ahref)
                            p_parts.append(f'<p><a href="{full_href}">{atxt}</a></p>')
                        elif atxt:
                            p_parts.append(atxt)
                    elif elem.name == "img":
                        src = elem.get("src", "")
                        if src:
                            alt = elem.get("alt", "")
                            full_src = urljoin(url, src)
                            p_parts.append(f'<p><a href="{full_src}">查看图片</a></p>' if alt else f'<p><a href="{full_src}">查看图片</a></p>')
                    elif hasattr(elem, "get_text"):
                        txt = elem.get_text(" ", strip=True)
                        if txt:
                            p_parts.append(txt)
                txt = " ".join(p_parts)
                if txt:
                    parts.append(txt)
            elif tag == "table":
                table_rows = []
                for tr in child.find_all("tr"):
                    cells = [td.get_text(" ", strip=True) for td in tr.find_all(["td", "th"])]
                    if cells:
                        table_rows.append("| " + " | ".join(cells) + " |")
                if table_rows:
                    parts.append("\n".join(table_rows))
            elif tag == "img":
                src = child.get("src", "")
                if src:
                    alt = child.get("alt", "")
                    full_src = urljoin(url, src)
                    parts.append(f'<p><a href="{full_src}">查看图片</a></p>' if alt else f'<p><a href="{full_src}">查看图片</a></p>')
        content_html = "\n\n".join(parts)
    # Attachments — scan content div only
    scope = content_div or soup
    attachments = []
    for a in scope.find_all("a", href=True):
        ahref = a["href"].strip().lower()
        if re.search(r"\.(pdf|doc|docx|xls|xlsx|zip|rar|txt)$", ahref):
            full_url = urljoin(url, a["href"].strip())
            attachments.append({"title": a.get_text(strip=True) or full_url.split("/")[-1], "url": full_url})
    if len(content_html.strip()) < 20 and attachments:
        content_html = f'<p><a href="{url}">{title}</a></p>\n\n附件列表：\n'
        for att in attachments:
            content_html += f"  - [{att['title']}]({att['url']})\n"
    return title, pub_date, content_html, attachments


def incremental_filter(items):
    cutoff = datetime.now() - timedelta(days=INCREMENTAL_DAYS)
    filtered = []
    for title, url, date_str in items:
        try:
            item_date = datetime.strptime(date_str, "%Y-%m-%d")
            if item_date >= cutoff:
                filtered.append((title, url, date_str))
        except (ValueError, IndexError):
            filtered.append((title, url, date_str))
    return filtered


def main():
    is_incremental = any(arg in sys.argv for arg in ["--incremental", "incremental", "inc"])
    page_urls = [LIST_URL]
    for i in range(2, MAX_PAGES + 1):
        page_urls.append(BASE_URL + LIST_PATH + f"/zfxxgk_{i}.shtml")
    all_items = []
    for idx, url in enumerate(page_urls):
        try:
            html = fetch(url)
            items = parse_list(html)
            print(f"Page {idx+1}: {len(items)} items", file=sys.stderr)
            all_items.extend(items)
        except Exception as e:
            print(f"Page {idx+1} error ({url}): {e}", file=sys.stderr)
    print(f"Total items from list: {len(all_items)}", file=sys.stderr)
    if is_incremental:
        all_items = incremental_filter(all_items)
        print(f"After incremental filter ({INCREMENTAL_DAYS}d): {len(all_items)} items", file=sys.stderr)
    seen = set()
    unique_items = []
    for item in all_items:
        if item[1] not in seen:
            seen.add(item[1])
            unique_items.append(item)
    print(f"Unique items: {len(unique_items)}", file=sys.stderr)
    results = []
    for title, url, list_date in unique_items:
        try:
            html = fetch(url)
            det_title, det_date, content, attachments = parse_detail(html, url)
            final_title = det_title or title
            final_date = det_date or list_date
            results.append({
                "title": final_title,
                "page_url": url,
                "publish_date": final_date,
                "content": content,
                "attachments": json.dumps(attachments, ensure_ascii=False) if attachments else "",
                "site_name": SITE_NAME,
            })
            print(f"  OK: {final_title[:50]}", file=sys.stderr)
        except Exception as e:
            print(f"  ERR {url}: {e}", file=sys.stderr)
    for r in results:
        print(json.dumps(r, ensure_ascii=False))


if __name__ == "__main__":
    main()