watcher.py

import os
import re
import time
import json
import shutil
import tempfile
import subprocess
import requests
from datetime import datetime, timezone
from bs4 import BeautifulSoup

# ===== Channel.io 設定 =====
GET_URL = "https://desk-api.channel.io/desk/channels/200605/groups/519217/messages"
POST_URL = GET_URL

PARAMS = {
    "sortOrder": "desc",
    "limit": 36,
    "logFolded": "false",
}

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

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

# ===== ssyoutube =====
SSYOUTUBE_URL = "https://ssyoutube.online/yt-video-detail/"

# ===== Utils =====
def parse_updated_at(value):
    if isinstance(value, (int, float)):
        return datetime.fromtimestamp(value / 1000, tz=timezone.utc)
    elif isinstance(value, str):
        return datetime.fromisoformat(value.replace("Z", "+00:00"))
    return None

def extract_youtube_id(text):
    patterns = [
        r"v=([A-Za-z0-9_-]{11})",
        r"youtu\.be/([A-Za-z0-9_-]{11})",
    ]
    for p in patterns:
        m = re.search(p, text)
        if m:
            return m.group(1)
    return None

# ===== ssyoutube HTML 解析 =====
def fetch_download_links(youtube_url):
    res = requests.post(
        SSYOUTUBE_URL,
        data={"videoURL": youtube_url},
        timeout=30,
        headers={
            "User-Agent": "Mozilla/5.0",
            "Referer": "https://ssyoutube.online/",
        }
    )
    res.raise_for_status()
    
    # nonceを抽出
    soup = BeautifulSoup(res.text, "lxml")
    nonce_match = re.search(r"formData\.append\('nonce', '([^']+)'\);", res.text)
    nonce = nonce_match.group(1) if nonce_match else None
    
    buttons = soup.select("button[data-url]")

    results = []
    for btn in buttons:
        url = btn.get("data-url")
        quality = btn.get("data-quality")
        has_audio = btn.get("data-has-audio")

        if not url:
            continue

        results.append({
            "url": url,
            "quality": quality or "audio",
            "has_audio": has_audio,
        })

    return results, nonce

# ===== 最高画質映像 + 音声を選択 =====
def select_best_video_and_audio(items):
    video_items = []
    audio_item = None

    for item in items:
        if item["quality"] == "audio":
            audio_item = item
        elif item["has_audio"] == "false":
            m = re.match(r"(\d+)p", item["quality"])
            if m:
                video_items.append((int(m.group(1)), item))

    if not video_items or not audio_item:
        return None, None

    video_items.sort(key=lambda x: x[0], reverse=True)
    return video_items[0][1], audio_item

# ===== ssyoutube サーバー側結合 =====
def merge_video_on_server(video_url, audio_url, video_id, quality, video_title, nonce):
    """
    ssyoutubeサーバーで動画と音声を結合する
    """
    # IDと画質からリクエストIDを生成
    request_id = f"{video_id}_{quality}p"
    
    # リクエストデータの構築
    request_data = {
        "id": request_id,
        "ttl": 3600000,
        "inputs": [
            {
                "url": video_url,
                "ext": "mp4",
                "chunkDownload": {
                    "type": "header",
                    "size": 52428800,
                    "concurrency": 3
                }
            },
            {
                "url": audio_url,
                "ext": "m4a"
            }
        ],
        "output": {
            "ext": "mp4",
            "downloadName": f"{video_title}_{quality}p.mp4",
            "chunkUpload": {
                "size": 104857600,
                "concurrency": 3
            }
        },
        "operation": {
            "type": "replace_audio_in_video"
        }
    }
    
    # 結合開始リクエスト
    payload = {
        "action": "process_video_merge",
        "nonce": nonce,
        "request_data": json.dumps(request_data)
    }
    
    headers = {
        "User-Agent": "Mozilla/5.0",
        "Referer": "https://ssyoutube.online/",
        "Origin": "https://ssyoutube.online",
        "Content-Type": "application/x-www-form-urlencoded; charset=UTF-8",
    }
    
    res = requests.post(
        "https://ssyoutube.online/wp-admin/admin-ajax.php",
        data=payload,
        headers=headers,
        timeout=30
    )
    res.raise_for_status()
    
    result = res.json()
    
    if not result.get("success") or not result.get("data", {}).get("success"):
        raise RuntimeError(f"結合開始に失敗しました: {result}")
    
    # ステータス監視用URLを取得
    monitor_url = result["data"]["result"]["monitor"]["http"]
    
    # 完了するまでポーリング
    max_attempts = 60  # 最大60回試行（30秒間隔で約30分）
    attempt = 0
    
    while attempt < max_attempts:
        time.sleep(30)  # 30秒間隔でチェック
        
        status_res = requests.get(monitor_url, timeout=30)
        status_res.raise_for_status()
        status_data = status_res.json()
        
        if not status_data.get("success"):
            raise RuntimeError(f"ステータス取得に失敗しました: {status_data}")
        
        status = status_data["result"]["status"]
        
        if status == "done":
            # 完了したらダウンロードURLを返す
            return status_data["result"]["output"]["url"]
        elif status == "error":
            error_msg = status_data["result"]["error"]
            raise RuntimeError(f"結合処理でエラーが発生しました: {error_msg}")
        # "processing" の場合は継続
        
        attempt += 1
        print(f"結合処理中... 進捗: {status_data['result'].get('progress_in_percent', 0)}%")
    
    raise RuntimeError("結合処理がタイムアウトしました")

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()

# ===== Main =====
def main():
    processed_messages = set()  # 処理済みメッセージを追跡
    
    while True:
        try:
            res = requests.get(
                GET_URL,
                headers=HEADERS_GET,
                params=PARAMS,
                timeout=30,
            )
            res.raise_for_status()

            messages = res.json().get("messages", [])
            latest_msg = None
            latest_time = None

            for msg in messages:
                msg_id = msg.get("id")
                plain_text = msg.get("plainText")
                updated_at = msg.get("updatedAt")

                if not plain_text or updated_at is None:
                    continue
                    
                # 既に処理済みのメッセージはスキップ
                if msg_id in processed_messages:
                    continue

                t = parse_updated_at(updated_at)
                if not t:
                    continue

                if latest_time is None or t > latest_time:
                    latest_time = t
                    latest_msg = msg

            if not latest_msg:
                time.sleep(10)
                continue

            youtube_id = extract_youtube_id(latest_msg["plainText"])
            if not youtube_id:
                time.sleep(10)
                continue

            youtube_url = f"https://www.youtube.com/watch?v={youtube_id}"
            items, nonce = fetch_download_links(youtube_url)
            
            if not nonce:
                print("nonceの取得に失敗しました")
                time.sleep(10)
                continue

            video_item, audio_item = select_best_video_and_audio(items)
            if not video_item or not audio_item:
                print("適切な動画または音声が見つかりません")
                time.sleep(10)
                continue

            # 画質情報を取得
            quality_match = re.match(r"(\d+)p", video_item["quality"])
            if not quality_match:
                print("画質情報の取得に失敗しました")
                time.sleep(10)
                continue
                
            quality = quality_match.group(1)
            
            # 動画タイトル（簡易版 - 実際にはYouTubeから取得する方が良い）
            video_title = f"youtube_video_{youtube_id}"
            
            # サーバー側で結合
            print(f"サーバー側での結合を開始します: {youtube_id} {quality}p")
            merged_url = merge_video_on_server(
                video_item["url"],
                audio_item["url"],
                youtube_id,
                quality,
                video_title,
                nonce
            )
            
            # 結果を送信
            message = f"🎬 結合済み動画（{quality}p）\n{merged_url}"
            send_to_channel(message)
            
            # 処理済みとしてマーク
            processed_messages.add(latest_msg.get("id"))
            
            print("送信完了:", merged_url)

        except Exception as e:
            print("エラー:", e)
            import traceback
            traceback.print_exc()

        time.sleep(15)

if __name__ == "__main__":
    main()