#!/usr/bin/env python3
"""Crawler for 来安县人民政府 - 通知公告
CMS: cz_govc platform, UTF-8, 24条/页
Site: www.laian.gov.cn/zwdt/tzgg/
"""
import requests, re, json, sys, os, sqlite3, time
from datetime import datetime, timedelta
from bs4 import BeautifulSoup
from urllib.parse import urljoin

BASE_URL = "http://www.laian.gov.cn"
LIST_PATH = "/zwdt/tzgg"
LIST_URL = BASE_URL + LIST_PATH + "/index.html"
SITE_NAME = "来安县人民政府-通知公告"
GROUP = "安徽滁州"
DB_PATH = os.getenv("SEARCH_DB", "/root/search.db")
MAX_PAGES = 5
INCREMENTAL_DAYS = 7

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, retries=3):
    for attempt in range(retries):
        try:
            resp = session.get(url, timeout=TIMEOUT)
            resp.encoding = "utf-8"
            return resp.text
        except Exception as e:
            if attempt < retries - 1:
                print("  Retry %d/%d: %s" % (attempt + 1, retries, e), file=sys.stderr)
                time.sleep(3 * (attempt + 1))
            else:
                raise e
    return ""


def parse_list(html):
    """Parse list — ul.doc_list > li > span.date + a(title/href)"""
    soup = BeautifulSoup(html, "html.parser")
    items = []
    ul = soup.find("ul", class_=lambda c: c and "doc_list" in c)
    if not ul:
        return items
    for li in ul.find_all("li", recursive=False):
        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
        span = li.find("span", class_="date")
        date = span.get_text(strip=True) if span else ""
        if not href.startswith("http"):
            href = urljoin(BASE_URL, href)
        items.append((title.strip(), href, date[:10]))
    return items


def parse_detail(html, url):
    """Parse detail — div.lmcontent > div.newscontnet > p/table/img + meta"""
    soup = BeautifulSoup(html, "html.parser")

    # Title
    title = ""
    title_el = soup.find("meta", attrs={"name": re.compile(r"ArticleTitle|Title", re.I)})
    if title_el and title_el.get("content"):
        title = title_el["content"].strip()
    if not title:
        h1 = soup.find("h1")
        if h1:
            title = h1.get_text(strip=True)
    if not title:
        title_tag = soup.find("title")
        if title_tag:
            title = title_tag.get_text(strip=True)
            title = re.sub(r"_[^_]+$", "", title).strip()

    # Date from meta
    pub_date = ""
    for meta in soup.find_all("meta"):
        name = (meta.get("name") or "").lower()
        if "pubdate" in name or "publishdate" in name:
            pub_date = (meta.get("content") or "")[:10]
            break
    if not pub_date:
        m = re.search(r"发布时间[：:]\s*(\d{4}-\d{1,2}-\d{1,2})", html)
        if m:
            pub_date = m.group(1)

    # Content from div.lmcontent / div.newscontnet
    content_div = soup.find("div", class_=lambda c: c and ("lmcontent" in c or "newscontnet" in c or c == "content"))
    parts = []
    attach = []

    if content_div:
        # Drill down to innermost content div
        inner = content_div.find("div", class_=lambda c: c and "newscontnet" in c) or content_div

        # Remove noise
        for noise in inner.find_all(["script", "style"]):
            noise.decompose()

        for child in inner.find_all(["p", "table", "img"], recursive=True):
            if child.name == "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 child.name == "p":
                txt = child.get_text(" ", strip=True)
                if txt and len(txt) > 2:
                    parts.append(txt)
            elif child.name == "img":
                src = child.get("src", "")
                alt = child.get("alt", "")
                if src:
                    if not src.startswith("http"):
                        src = urljoin(url, src)
                    parts.append("![%s](%s)" % (alt or "", src))

        # Attachments from content area only
        for a in inner.find_all("a", href=True):
            ahref = a["href"].strip()
            if re.search(r"\.(pdf|doc|docx|xls|xlsx|zip|rar|txt)$", ahref, re.I):
                full_url = ahref if ahref.startswith("http") else urljoin(url, ahref)
                attach.append({"title": a.get_text(strip=True) or full_url.split("/")[-1], "url": full_url})

    content = "\n\n".join(parts)
    if len(content.strip()) < 20:
        content = '<p><a href="%s">%s</a></p>' % (url, title or url.split('/')[-1])

    return title, pub_date, content, attach


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"])

    # Build page URLs
    page_urls = [LIST_URL]
    for i in range(2, MAX_PAGES + 1):
        page_urls.append(LIST_URL + "?page=%d" % i)

    all_items = []
    for idx, url in enumerate(page_urls):
        try:
            html = fetch(url)
            items = parse_list(html)
            print("Page %d: %d items" % (idx + 1, len(items)), file=sys.stderr)
            all_items.extend(items)
        except Exception as e:
            print("Page %d error (%s): %s" % (idx + 1, url, e), file=sys.stderr)

    print("Total: %d" % len(all_items), file=sys.stderr)

    if is_incremental:
        all_items = incremental_filter(all_items)
        print("Incremental: %d" % len(all_items), file=sys.stderr)

    # Dedup by URL
    seen = set()
    unique_items = []
    for item in all_items:
        if item[1] not in seen:
            seen.add(item[1])
            unique_items.append(item)

    print("Unique: %d" % 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("  OK: %s" % final_title[:50], file=sys.stderr)
        except Exception as e:
            print("  ERR %s: %s" % (url, e), file=sys.stderr)

    saved = 0
    skipped = 0
    errors = 0
    for r in results:
        content_text = r.get("content", "")[:50000]
        summary = re.sub(r"\s+", " ", content_text[:200]).strip() or r["title"]
        attachments = r.get("attachments", "")
        if isinstance(attachments, str):
            pass
        else:
            attachments = json.dumps(attachments, ensure_ascii=False)
        try:
            conn = sqlite3.connect(DB_PATH, timeout=60)
            c = conn.cursor()
            c.execute("""INSERT OR IGNORE INTO gov_raw
                (page_url, title, publish_date, site_name, group_name, summary, content, attachments, source_url)
                VALUES (?,?,?,?,?,?,?,?,?)""",
                (r["page_url"], r["title"], r["publish_date"],
                 SITE_NAME, GROUP, summary, content_text,
                 attachments, r["page_url"]))
            if c.rowcount > 0:
                saved += 1
            else:
                skipped += 1
            conn.commit()
            conn.close()
        except Exception as e:
            errors += 1
            print("  DB error: %s" % e, file=sys.stderr)

    print("[%s] Done! Saved: %d, Skipped: %d, Errors: %d" % (SITE_NAME, saved, skipped, errors), file=sys.stderr)

    # JSON output for admin panel
    for r in results:
        print(json.dumps(r, ensure_ascii=False))


if __name__ == "__main__":
    main()