feat: YouTube downloader cookie support, JS runtime, 720p limit

Dockerfile.complete

@@ -75,6 +75,11 @@ RUN set -ex && \
     apt-get install -y --no-install-recommends ffmpeg nginx supervisor procps && \
     rm -rf /var/lib/apt/lists/*
 
+# 安装 Node.js 22（yt-dlp EJS 挑战求解器需要）并升级 yt-dlp
+RUN curl -fsSL https://deb.nodesource.com/setup_22.x | bash - && \
+    apt-get install -y nodejs && \
+    pip install --upgrade yt-dlp
+
 ENV PATH="/usr/bin:${PATH}"
 ENV HF_ENDPOINT=https://hf-mirror.com
 ENV PYTHONUNBUFFERED=1

backend/app/downloaders/youtube_downloader.py

@@ -12,23 +12,49 @@
 from app.services.proxy_config_manager import ProxyConfigManager
 from app.utils.path_helper import get_data_dir
 from app.utils.url_parser import extract_video_id
+from app.services.cookie_manager import CookieConfigManager
+import tempfile
 
 logger = logging.getLogger(__name__)
 
 
 def _apply_proxy(ydl_opts: dict) -> dict:
-    """YouTube 在国内需要代理。配置了全局代理就塞进 yt-dlp opts。"""
     proxy = ProxyConfigManager().get_proxy_url()
     if proxy:
-        ydl_opts['proxy'] = proxy
-        logger.info(f"yt-dlp 走代理: {proxy}")
+        ydl_opts["proxy"] = proxy
+        logger.info(f"yt-dlp proxy: {proxy}")
     return ydl_opts
 
 
 class YoutubeDownloader(Downloader, ABC):
     def __init__(self):
-
         super().__init__()
+        self._cookie_mgr = CookieConfigManager()
+        self._cookie = self._cookie_mgr.get("youtube")
+        self._cookiefile = self._write_netscape_cookie_file()
+
+    def _write_netscape_cookie_file(self):
+        if not self._cookie:
+            return None
+        tmp = tempfile.NamedTemporaryFile(mode="w", suffix=".txt", delete=False, encoding="utf-8")
+        cookie = self._cookie.strip()
+        # Netscape format from browser extension
+        if cookie.startswith("# Netscape") or cookie.startswith("# HTTP Cookie"):
+            tmp.write(cookie)
+        elif "\t" in cookie:
+            # Netscape format without header
+            tmp.write("# Netscape HTTP Cookie File\n")
+            tmp.write(cookie)
+        else:
+            # Simple key=value; key=value format
+            tmp.write("# Netscape HTTP Cookie File\n")
+            for pair in cookie.split("; "):
+                if "=" in pair:
+                    key, value = pair.split("=", 1)
+                    tmp.write(".youtube.com\tTRUE\t/\tFALSE\t0\t" + key + "\t" + value + "\n")
+        tmp.close()
+        logger.info("YouTube Netscape cookie file: %s", tmp.name)
+        return tmp.name
 
     def download(
         self,
@@ -47,15 +73,17 @@ def download(
         output_path = os.path.join(output_dir, "%(id)s.%(ext)s")
 
         ydl_opts = {
-            'format': 'bestaudio[ext=m4a]/bestaudio/best',
-            'outtmpl': output_path,
-            'noplaylist': True,
-            'quiet': False,
+            "format": "bestaudio[ext=m4a]/bestaudio/best",
+            "outtmpl": output_path,
+            "noplaylist": True,
+            "quiet": False,
         }
 
         if skip_download:
-            ydl_opts['skip_download'] = True
+            ydl_opts["skip_download"] = True
 
+        if self._cookiefile:
+            ydl_opts["cookiefile"] = self._cookiefile
         _apply_proxy(ydl_opts)
         with yt_dlp.YoutubeDL(ydl_opts) as ydl:
             info = ydl.extract_info(video_url, download=not skip_download)
@@ -73,18 +101,16 @@ def download(
             cover_url=cover_url,
             platform="youtube",
             video_id=video_id,
-            raw_info={'tags': info.get('tags')},
+            raw_info={"tags": info.get("tags")},
             video_path=None,
         )
 
     def download_video(
         self,
         video_url: str,
         output_dir: Union[str, None] = None,
+        max_height: int = 720,
     ) -> str:
-        """
-        下载视频，返回视频文件路径
-        """
         if output_dir is None:
             output_dir = get_data_dir()
         video_id = extract_video_id(video_url, "youtube")
@@ -94,42 +120,36 @@ def download_video(
         os.makedirs(output_dir, exist_ok=True)
         output_path = os.path.join(output_dir, "%(id)s.%(ext)s")
 
+        fmt = f"bestvideo[height<={max_height}][ext=mp4]+bestaudio[ext=m4a]/best[height<={max_height}][ext=mp4]"
         ydl_opts = {
-            'format': 'bestvideo[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]',
-            'outtmpl': output_path,
-            'noplaylist': True,
-            'quiet': False,
-            'merge_output_format': 'mp4',  # 确保合并成 mp4
+            "format": fmt,
+            "outtmpl": output_path,
+            "noplaylist": True,
+            "quiet": False,
+            "merge_output_format": "mp4",
         }
 
+        if self._cookiefile:
+            ydl_opts["cookiefile"] = self._cookiefile
         _apply_proxy(ydl_opts)
         with yt_dlp.YoutubeDL(ydl_opts) as ydl:
             info = ydl.extract_info(video_url, download=True)
             video_id = info.get("id")
             video_path = os.path.join(output_dir, f"{video_id}.mp4")
 
         if not os.path.exists(video_path):
-            raise FileNotFoundError(f"视频文件未找到: {video_path}")
+            raise FileNotFoundError(f"video not found: {video_path}")
 
         return video_path
 
     def download_subtitles(self, video_url: str, output_dir: str = None,
                            langs: List[str] = None) -> Optional[TranscriptResult]:
-        """
-        通过 YouTube InnerTube API 直接获取字幕（优先人工字幕，其次自动生成）。
-        比 yt_dlp 方式更轻量，无需写临时文件到磁盘。
-
-        :param video_url: 视频链接
-        :param output_dir: 未使用（保留接口兼容）
-        :param langs: 优先语言列表
-        :return: TranscriptResult 或 None
-        """
         if langs is None:
-            langs = ['zh-Hans', 'zh', 'zh-CN', 'zh-TW', 'en', 'en-US', 'ja']
+            langs = ["zh-Hans", "zh", "zh-CN", "zh-TW", "en", "en-US", "ja"]
 
         video_id = extract_video_id(video_url, "youtube")
         fetcher = YouTubeSubtitleFetcher()
         print(
-            f"尝试获取字幕，video_id={video_id}, langs={langs}"
+            f"fetch subtitles, video_id={video_id}, langs={langs}"
         )
         return fetcher.fetch_subtitles(video_id, langs)