#!/usr/bin/env python3
"""
Crawler for 新疆生产建设兵团 - 民意征集
CMS: ZCMS, div.list + div.article + div.content
Site: www.xjbt.gov.cn/hd/myzj/
"""
import requests, re, json, sys, os
from datetime import datetime, timedelta
from bs4 import BeautifulSoup
from urllib.parse import urljoin

BASE_URL = "http://sthjj.xjbt.gov.cn"
LIST_PATH = "/xxgk/tzgg"
LIST_URL = BASE_URL + LIST_PATH + "/index.shtml"
SITE_NAME = "兵团生态环境局-通知公示公告"
INCREMENTAL_DAYS = 7
MAX_PAGES = 5

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):
    """Parse list page — div.list > div.article > div.name > a + div.time"""
    soup = BeautifulSoup(html, "html.parser")
    items = []
    list_div = soup.find("div", class_="list")
    if not list_div:
        return items
    for article in list_div.find_all("div", class_="article", recursive=False):
        a = article.find("a", href=True)
        if not a:
            continue
        title = a.get_text(strip=True)
        href = a["href"].strip()
        if not title or not href:
            continue
        time_div = article.find("div", class_="time")
        date = time_div.get_text(strip=True) if time_div 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):
    """Parse detail page — div.content with <p> preserving <a>/<img>."""
    soup = BeautifulSoup(html, "html.parser")

    # Title from h2
    title = ""
    h2 = soup.find("h2", class_="h2")
    if h2:
        title = h2.get_text(strip=True)
    if not title:
        meta_title = soup.find("meta", attrs={"name": "ArticleTitle"})
        if meta_title and meta_title.get("content"):
            title = meta_title["content"].strip()

    # Date from meta
    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 from div.content
    content_html = ""
    content_div = soup.find("div", class_="content")
    if content_div:
        parts = []
        processed = set()
        for child in content_div.find_all(["p", "table", "img"], recursive=True):
            if id(child) in processed:
                continue
            processed.add(id(child))
            tag = child.name.lower()
            if tag == "p":
                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"[{atxt}]({full_href})")
                        elif atxt:
                            p_parts.append(atxt)
                    elif elem.name == "img":
                        processed.add(id(elem))
                        src = elem.get("src", "")
                        if src:
                            alt = elem.get("alt", "")
                            full_src = urljoin(url, src)
                            p_parts.append(f"![{alt}]({full_src})" if alt else f"![]({full_src})")
                    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"![{alt}]({full_src})" if alt else f"![]({full_src})")
        content_html = "\n\n".join(parts)

    # Attachments
    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"/index_{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: {len(all_items)}", file=sys.stderr)
    if is_incremental:
        all_items = incremental_filter(all_items)
        print(f"Incremental: {len(all_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: {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()