#!/usr/bin/env python3
"""
Crawler for 山东玉皇化工有限公司 - 公司新闻
CMS: 速优(SUYOU) custom, UTF-8
Site: www.yuhuanghuagong.com/news/18/
"""
import requests, re, json, sys, os
from datetime import datetime, timedelta
from bs4 import BeautifulSoup
from urllib.parse import urljoin

BASE_URL = "http://www.yuhuanghuagong.com"
LIST_URL = BASE_URL + "/news/18/"
SITE_NAME = "山东玉皇化工-公司新闻"
INCREMENTAL_DAYS = 7
MAX_PAGES = 5
PER_PAGE = 20

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.main > div.cbox-1 > p.e_text-9.s_title > a + p.e_timeFormat-14"""
    soup = BeautifulSoup(html, "html.parser")
    items = []
    main = soup.find("div", class_="main")
    if not main:
        return items

    # Each article is in a div.cbox-1.p_loopitem
    for item_div in main.find_all("div", class_="p_loopitem"):
        # Title/link
        title_p = item_div.find("p", class_=re.compile(r"e_text-9"))
        if not title_p:
            continue
        a = title_p.find("a", href=True)
        if not a:
            continue
        href = a["href"].strip()
        title = a.get_text(strip=True)
        if not title or not href:
            continue

        # Date from p.e_timeFormat-14
        date_p = item_div.find("p", class_=re.compile(r"timeFormat"))
        date_str = date_p.get_text(strip=True) if date_p else ""

        # Resolve URL
        if href.startswith("http"):
            pass
        elif href.startswith("/"):
            href = BASE_URL + href
        else:
            href = urljoin(LIST_URL, href)

        items.append((title, href, date_str[:7]))
    return items


def get_page_urls():
    urls = [LIST_URL]
    for i in range(2, MAX_PAGES + 1):
        offset = (i - 1) * 10
        urls.append(f"{BASE_URL}/News/1646377329820401664-{offset}-10.html")
    return urls


def parse_detail(html, url):
    """Parse detail page — h1.e_h1-77 title + div.e_richText-63 content."""
    soup = BeautifulSoup(html, "html.parser")

    # Title
    title = ""
    h1 = soup.find("h1", class_=re.compile(r"subtitle"))
    if h1:
        title = h1.get_text(strip=True)
    if not title:
        title_meta = soup.find("meta", attrs={"name": re.compile(r"Title|ArticleTitle", re.I)})
        if title_meta and title_meta.get("content"):
            title = title_meta["content"].strip()

    # Date
    pub_date = ""
    date_p = soup.find("p", class_=re.compile(r"timeFormat"))
    if date_p:
        raw = date_p.get_text(strip=True)
        m = re.search(r"(\d{4}-\d{2}-\d{2})", raw)
        if m:
            pub_date = m.group(1)

    # Content
    content_html = ""
    content_div = soup.find("div", class_=re.compile(r"richText"))
    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:
                            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 == "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", "span"):
                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 (rare on this site)
    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:
                # date_str is YYYY-MM format, compare by year-month
                item_date = datetime.strptime(date_str + "-01", "%Y-%m-%d")
                if item_date >= cutoff.replace(day=1):
                    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 = get_page_urls()

    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()