#!/usr/bin/env python3
"""
Crawler for 远安县人民政府 - 通知公告
CMS: 远安政府CMS, UTF-8
Site: www.yuanan.gov.cn/list-206-1.html
"""
import requests, re, json, sys, os
from datetime import datetime, timedelta
from bs4 import BeautifulSoup
from urllib.parse import urljoin

BASE_URL = "http://www.yuanan.gov.cn"
LIST_URL = BASE_URL + "/list-206-1.html"
SITE_NAME = "远安县-通知公告"
INCREMENTAL_DAYS = 7
MAX_PAGES = 5
PER_PAGE = 10

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 — div.txtlist-con > ul > li > div.txtlisty1 > a + div.txtlisty2"""
    soup = BeautifulSoup(html, "html.parser")
    items = []
    # Find the article list ul (the one inside txtlist-con with 10 links)
    for ul in soup.find_all("ul"):
        links = ul.find_all("a", href=True)
        if len(links) >= 5:
            # Check if this is the article list by looking for content- links
            content_links = [a for a in links if "/content-" in a.get("href", "")]
            if len(content_links) >= 3:
                for li in ul.find_all("li", recursive=False):
                    a = li.find("a", href=True)
                    if not a:
                        continue
                    href = a["href"].strip()
                    title = a.get("title") or a.get_text(strip=True)
                    if not title or not href or "/content-" not in href:
                        continue
                    date_div = li.find("div", class_="txtlisty2")
                    date_str = date_div.get_text(strip=True) if date_div else ""
                    # Resolve
                    if href.startswith("http"):
                        pass
                    elif href.startswith("/"):
                        href = BASE_URL + href
                    else:
                        href = urljoin(LIST_URL, href)
                    items.append((title.strip(), href, date_str[:10]))
                break
    return items


def parse_detail(html, url):
    """Parse detail — <title> title + div.nry-info date + div.txtcontent-div content."""
    soup = BeautifulSoup(html, "html.parser")

    # Title from <title>
    title = ""
    title_tag = soup.find("title")
    if title_tag:
        raw = title_tag.get_text(strip=True)
        title = re.sub(r"\s*[-–—].*$", "", raw).strip()

    # Date
    pub_date = ""
    nry_div = soup.find("div", class_="nry-info")
    if nry_div:
        spans = nry_div.find_all("span")
        for sp in spans:
            txt = sp.get_text(strip=True)
            m = re.search(r"(\d{4}-\d{2}-\d{2})", txt)
            if m:
                pub_date = m.group(1)
                break

    # Content
    content_html = ""
    content_div = soup.find("div", class_="txtcontent-div")
    if not content_div:
        content_div = soup.find("div", class_="yaxveiw100")
    if content_div:
        parts = []
        for child in content_div.children:
            if child.name is None:
                txt = str(child).strip()
                if txt:
                    parts.append(txt)
                continue
            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 and not ahref.startswith("javascript"):
                            full_href = urljoin(url, ahref)
                            p_parts.append(f"[{atxt}]({full_href})")
                        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"![{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)
                joined = " ".join(p_parts)
                if joined:
                    parts.append(joined)
            elif tag == "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:
                        rows.append("| " + " | ".join(cells) + " |")
                if rows:
                    parts.append("\n".join(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})")
            elif tag in ("div", "section"):
                txt = child.get_text(" ", strip=True)
                if txt:
                    parts.append(txt)
            elif tag == "br":
                pass
            else:
                txt = child.get_text(" ", strip=True)
                if txt:
                    parts.append(txt)
        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:
        if date_str:
            try:
                item_date = datetime.strptime(date_str, "%Y-%m-%d")
                if item_date >= cutoff:
                    filtered.append((title, url, date_str))
            except ValueError:
                filtered.append((title, url, date_str))
        else:
            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 + f"/list-206-{i}.html")

    all_items = []
    for idx, url in enumerate(page_urls):
        try:
            html = fetch(url)
            items = parse_list(html)
            if not items:
                break
            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()