news.py

import os
import time
import json
import requests
import hashlib
from datetime import datetime, timezone
from bs4 import BeautifulSoup
import xml.etree.ElementTree as ET

# ===== Channel.io 設定 =====
CHANNEL_ID = "200605"
GROUP_ID = "534457"

POST_URL = f"https://desk-api.channel.io/desk/channels/{CHANNEL_ID}/groups/{GROUP_ID}/messages"

X_ACCOUNT = os.getenv("channeliotokenbot2")
if not X_ACCOUNT:
    raise RuntimeError("環境変数 channeliotokenbot2 が設定されていません")

HEADERS_POST = {
    "accept": "application/json",
    "accept-language": "ja",
    "content-type": "application/json",
    "x-account": X_ACCOUNT,
}

# ===== RSS =====
RSS_URL = "https://www.nippon.com/ja/rss-all/"

# ===== 設定 =====
INTERVAL_SECONDS = 60 * 60 * 12  # 12時間
SENT_LOG_FILE = "sent_nippon_news.json"


# ===== Utils =====
def load_sent_log():
    if not os.path.exists(SENT_LOG_FILE):
        return set()
    with open(SENT_LOG_FILE, "r", encoding="utf-8") as f:
        return set(json.load(f))


def save_sent_log(sent_set):
    with open(SENT_LOG_FILE, "w", encoding="utf-8") as f:
        json.dump(list(sent_set), f, ensure_ascii=False, indent=2)


def hash_link(link: str) -> str:
    """リンクをハッシュ化（重複防止用）"""
    return hashlib.sha256(link.encode("utf-8")).hexdigest()


def send_to_channel(text):
    payload = {
        "requestId": f"desk-web-{int(time.time() * 1000)}",
        "blocks": [
            {"type": "text", "value": text}
        ],
    }

    res = requests.post(
        POST_URL,
        headers=HEADERS_POST,
        data=json.dumps(payload),
        timeout=30
    )
    res.raise_for_status()


def fetch_rss_items():
    res = requests.get(RSS_URL, timeout=30)
    res.raise_for_status()

    root = ET.fromstring(res.content)
    channel = root.find("channel")
    if channel is None:
        return []

    items = []
    for item in channel.findall("item"):
        title = item.findtext("title", "").strip()
        link = item.findtext("link", "").strip()
        description_raw = item.findtext("description", "").strip()

        # CDATA 内のHTMLをテキスト化
        soup = BeautifulSoup(description_raw, "lxml")
        description = soup.get_text(strip=True)

        if title and link:
            items.append({
                "title": title,
                "link": link,
                "description": description,
            })

    return items


# ===== Main =====
def main():
    sent_log = load_sent_log()

    while True:
        try:
            items = fetch_rss_items()
            new_count = 0

            for item in items:
                link_hash = hash_link(item["link"])
                if link_hash in sent_log:
                    continue

                message = (
                    f"<link type=\"url\" value=\"{item['link']}\">"
                    f"{item['title']}"
                    f"</link>\n\n"
                    f"{item['description']}"
                )

                send_to_channel(message)
                sent_log.add(link_hash)
                new_count += 1

                time.sleep(1)  # 連投防止

            if new_count > 0:
                save_sent_log(sent_log)
                print(f"{new_count} 件のニュースを送信しました")
            else:
                print("新しいニュースはありません")

        except Exception as e:
            print("エラー:", e)

        time.sleep(INTERVAL_SECONDS)


if __name__ == "__main__":
    main()