#!/usr/bin/env python3
import os
"""
长寿新闻网 - 城事 (ccs.cn)
华龙网CMS
列表: /web/column/col1812972.html → col1812972_{N}.html (N=2..20)
详情: shareapp.cqliving.com (外部站点)
"""
import re, sys, os, json, time
import urllib.request, urllib.error, ssl
from bs4 import BeautifulSoup
import sqlite3

ssl._create_default_https_context = ssl._create_unverified_context
ctx = ssl.create_default_context()
ctx.check_hostname = False
ctx.verify_mode = ssl.CERT_NONE

BASE_URL = "https://www.ccs.cn"
SITE_NAME = "长寿新闻网-城事"
GROUP_NAME = "新闻资讯"
DB = os.environ.get("SEARCH_DB", "/root/search.db")
HEADERS = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"}

LIST_PATH = "/web/column/col1812972"

_MAX_PG = None
for i, a in enumerate(sys.argv):
    if a == "--pages" and i + 1 < len(sys.argv):
        _MAX_PG = int(sys.argv[i + 1])
        break


def fetch(url, use_ssl=True):
    req = urllib.request.Request(url, headers=HEADERS)
    try:
        if use_ssl:
            return urllib.request.urlopen(req, timeout=30, context=ctx).read().decode("utf-8", errors="replace")
        else:
            return urllib.request.urlopen(req, timeout=30).read().decode("utf-8", errors="replace")
    except Exception as e:
        print(f"[WARN] fetch failed: {url} - {e}", file=sys.stderr)
        return ""


def extract_items(html):
    """Extract items from list page"""
    items = []
    soup = BeautifulSoup(html, "html.parser")
    for a in soup.find_all("a", href=True):
        href = a["href"]
        if "shareapp.cqliving.com" not in href:
            continue
        txt = a.get_text(strip=True)
        if not txt or len(txt) < 10:
            continue

        # Date is appended at end of title like "标题2026-07-14"
        date = ""
        m = re.search(r"(\d{4}-\d{2}-\d{2})$", txt)
        if m:
            date = m.group(1)
            title = txt[:m.start()].strip()
        else:
            title = txt

        if not href.startswith("http"):
            href = "https:" + href if href.startswith("//") else BASE_URL + href

        items.append({"title": title, "url": href, "date": date})
    return items


def parse_detail(html, url):
    """Parse detail page on shareapp.cqliving.com"""
    soup = BeautifulSoup(html, "html.parser")

    # Title
    title = ""
    h1 = soup.select_one("h1")
    if h1:
        title = h1.get_text(strip=True)

    # Date
    date_text = ""
    m = re.search(r"(\d{4}-\d{2}-\d{2})\s+(\d{2}:\d{2})", html)
    if m:
        date_text = m.group(1)

    # Content
    content = ""
    attachments = []

    content_div = soup.select_one("[class*=content]")
    if not content_div:
        return {"title": title, "date": date_text, "content": "", "attachments": []}

    parts = []
    for el in content_div.find_all(["p", "table", "img"], recursive=True):
        if el.name == "p" and el.find_parent("table"):
            continue
        if el.name == "table" and el.find_parent("table"):
            continue

        if el.name == "img":
            src = el.get("src", "")
            if "icon_" in src.lower():
                continue
            if src.startswith("//"):
                src = "https:" + src
            elif src.startswith("/"):
                src = url.split("/")[0] + src
            elif not src.startswith("http"):
                src = url.rsplit("/", 1)[0] + "/" + src
            alt = el.get("alt", "") or ""
            parts.append(f"![{alt}]({src})")
            continue

        if el.name == "table":
            rows = []
            for tr in el.find_all("tr"):
                cells = [c.get_text(" ", strip=True) for c in tr.find_all(["td", "th"])]
                if cells:
                    rows.append("| " + " | ".join(cells) + " |")
            if rows:
                col_count = len(rows[0].split("|")) - 2
                sep = "| " + " | ".join(["---"] * col_count) + " |"
                rows.insert(1, sep)
                parts.append("\n".join(rows))
            continue

        text = el.get_text(" ", strip=True)
        if text:
            parts.append(text)

    content = "\n\n".join(p for p in parts if p.strip())
    content = content.replace("\u00a0", " ").replace("&nbsp;", " ")

    return {"title": title, "date": date_text, "content": content, "attachments": []}


def main():
    conn = sqlite3.connect(DB, timeout=60)
    conn.execute("PRAGMA busy_timeout=60000")
    c = conn.cursor()

# QC20260925 去掉整站清空再重灌(抢锁+中途死掉会清空整站; page_url 有 UNIQUE 索引，插入本就幂等)     c.execute("DELETE FROM gov_search WHERE site_name=?", (SITE_NAME,))
# QC20260925 去掉整站清空再重灌(抢锁+中途死掉会清空整站; page_url 有 UNIQUE 索引，插入本就幂等)     c.execute("DELETE FROM gov_raw WHERE site_name=?", (SITE_NAME,))
    conn.commit()
    print("[清理] 已清理该站旧数据", flush=True)

    all_items = []
    total_pages = 20
    if _MAX_PG:
        total_pages = min(total_pages, _MAX_PG)

    # Page 1
    html = fetch(f"{BASE_URL}{LIST_PATH}.html")
    if not html:
        print("[ERROR] Cannot fetch list page", file=sys.stderr)
        sys.exit(1)
    items = extract_items(html)
    print(f"[INFO] Page 1: {len(items)} items", flush=True)
    all_items.extend(items)

    # Pages 2+
    for pg in range(2, total_pages + 1):
        url = f"{BASE_URL}{LIST_PATH}_{pg}.html"
        html = fetch(url)
        if not html:
            break
        items = extract_items(html)
        if not items:
            break
        print(f"[INFO] Page {pg}: {len(items)} items", flush=True)
        all_items.extend(items)

    print(f"[INFO] Total items from list: {len(all_items)}", flush=True)

    new_count = 0
    skip_count = 0
    empty_count = 0
    total_att = 0

    for idx, item in enumerate(all_items):
        c.execute("SELECT id FROM gov_raw WHERE page_url=?", (item["url"],))
        if c.fetchone():
            skip_count += 1
            continue

        if idx % 5 == 0:
            print(f"[INFO] [{idx}/{len(all_items)}] {item['title'][:40]}...", file=sys.stderr, end=" ", flush=True)

        html = fetch(item["url"])
        if not html:
            print("FETCH FAILED", file=sys.stderr)
            continue

        detail = parse_detail(html, item["url"])
        title = detail["title"] or item["title"]
        date_text = detail["date"] or item["date"]
        content = detail["content"]
        attachments_list = detail["attachments"]

        if not content.strip():
            empty_count += 1
            if idx % 5 == 0:
                print("EMPTY", file=sys.stderr)
            continue

        plain = re.sub(r'\s+', ' ', content).strip()
        summary = plain[:200] if plain else title[:200]
        att_json = json.dumps(attachments_list, ensure_ascii=False) if attachments_list else "[]"

        date_rank = 0
        if date_text:
            m2 = re.search(r"(\d{4})-(\d{2})-(\d{2})", date_text)
            if m2:
                date_rank = int(m2.group(1) + m2.group(2) + m2.group(3))

        try:
            c.execute(
                """INSERT OR IGNORE INTO gov_raw 
                (title, site_name, group_name, page_url, publish_date, content, summary, attachments, date_rank)
                VALUES (?,?,?,?,?,?,?,?,?)""",
                (title, SITE_NAME, GROUP_NAME, item["url"], date_text, content, summary, att_json, date_rank)
            )
            if c.rowcount > 0:
                new_count += 1
                total_att += len(attachments_list)
        except Exception as e:
            if idx % 5 == 0:
                print(f"DB ERROR: {e}", file=sys.stderr)

        time.sleep(0.2)

    conn.commit()
    conn.close()

    print(f"\n[RESULT] {SITE_NAME}: {new_count} new, {skip_count} skip, {empty_count} empty, {total_att} attachments", flush=True)


if __name__ == "__main__":
    main()