Replacing the OS to Ubuntu 24.04 and fixing the current results.

Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3.12.10-bookworm
+FROM ubuntu:latest
 
 # ENV PIP_ROOT_USER_ACTION=ignore
 ARG USERNAME=mcp_user
@@ -10,26 +10,30 @@ WORKDIR /tmp/
 COPY requirements.txt packages.txt ./
 
 RUN apt-get -y update && apt-get -y upgrade && xargs apt -y install < packages.txt
-RUN pip install --no-cache-dir --upgrade pip
 
-# Install CUDA 12.8
 
-RUN wget https://developer.download.nvidia.com/compute/cuda/repos/debian12/x86_64/cuda-keyring_1.1-1_all.deb \
+# Install CUDA 12.8
+RUN wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/x86_64/cuda-keyring_1.1-1_all.deb \
     && dpkg -i cuda-keyring_1.1-1_all.deb \
     && apt-get update \
     && apt-get -y install cuda-toolkit-12-8
 
+
+
 # https://docs.nvidia.com/cuda/cuda-installation-guide-linux/#post-installation-actions 
 ENV PATH=/usr/local/cuda-12.8/bin:${PATH}
 ENV LD_LIBRARY_PATH=/usr/local/cuda-12.8/lib64:/usr/lib/x86_64-linux-gnu/
 
 
 # Install cuDNN for CUDA 12
-RUN wget https://developer.download.nvidia.com/compute/cuda/repos/debian12/x86_64/cuda-keyring_1.1-1_all.deb \
+RUN wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/x86_64/cuda-keyring_1.1-1_all.deb \
 	&& dpkg -i cuda-keyring_1.1-1_all.deb \
 	&& apt-get update \
 	&& apt-get -y install cudnn-cuda-12
 
+# Remove EXTERNALLY-MANAGED
+RUN rm -f /usr/lib/python3.12/EXTERNALLY-MANAGED
+
 # Install Pytorch for CUDA 12.8
 RUN pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
 
@@ -41,6 +45,9 @@ RUN pip3 install --no-cache-dir -r requirements.txt
 RUN apt-get clean
 RUN rm -rf /tmp/*
 
+# Delet exist user with UID 1000
+RUN userdel -r ubuntu || true
+
 # https://huggingface.co/docs/hub/spaces-sdks-docker#permissions
 # Set up a new user named "user" with user ID 1000
 RUN useradd -m -u $USER_UID $USERNAME

app_srv/downloader.py

@@ -0,0 +1,120 @@
+import yt_dlp
+import os
+import uuid
+import json
+from pathlib import Path
+from typing import Dict, Any
+from datetime import datetime
+
+def download_youtube_media(url: str, base_dir: str = "./downloads") -> Dict[str, Any]:
+    """
+    Downloads video and audio from YouTube, saving them to a unique GUID folder.
+    Metadata is saved in JSON format including download datetime and timezone.
+    
+    Args:
+        url (str): YouTube video URL
+        base_dir (str): Base download directory (default './downloads')
+    
+    Returns:
+        dict: Dictionary with file paths and information:
+            {
+                'data_path': str,     # Path to download directory
+                'video_path': str,     # Full path to video.mp4
+                'audio_path': str,     # Full path to audio.mp3
+                'metadata_path': str  # Full path to metadata.json
+            }
+    
+    Raises:
+        RuntimeError: If download fails
+    """
+    try:
+        # Generate GUID and create folder
+        guid = str(uuid.uuid4())
+        download_dir = Path(base_dir) / guid
+        os.makedirs(download_dir, exist_ok=True)
+        
+        # File paths
+        video_path = download_dir / "video.mp4"
+        audio_path = download_dir / "audio.mp3"
+        metadata_path = download_dir / "metadata.json"
+
+        # Record exact download start time
+        download_datetime = datetime.now()
+        current_timezone = download_datetime.astimezone().tzinfo
+        
+        # 1. Download video (MP4)
+        video_opts = {
+            'format': (
+                "bestvideo[height=720][ext=mp4]"
+                "/worstvideo[height>720][ext=mp4]"
+                "/bestvideo[height<=720][ext=mp4]"
+                ),
+            'outtmpl': str(video_path),
+            'quiet': True,
+            'no_warnings': False,
+            'restrict_filenames': True,
+        }
+        
+        with yt_dlp.YoutubeDL(video_opts) as ydl:
+            video_info = ydl.extract_info(url, download=True)
+        
+        # 2. Download audio (MP3)
+        audio_opts = {
+            'format': 'bestaudio/best',
+            'outtmpl': str(audio_path),
+            'quiet': True,
+            'postprocessors': [{
+                'key': 'FFmpegExtractAudio',
+                'preferredcodec': 'mp3',
+                'preferredquality': '128',
+            }],
+        }
+        
+        with yt_dlp.YoutubeDL(audio_opts) as ydl:
+            audio_info = ydl.extract_info(url, download=True)
+        
+        # Format date and time for storage
+        formatted_date = download_datetime.strftime('%Y-%m-%d')
+        formatted_time = download_datetime.strftime('%H:%M:%S')
+        
+        # 3. Save metadata to JSON
+        metadata = {
+            'original_url': url,
+            'guid': guid,
+            'download_info': {
+                'date': formatted_date,
+                'time': formatted_time,
+                'timezone': str(current_timezone),
+                'datetime_iso': download_datetime.isoformat(),
+            },
+            'video': {
+                'path': str(video_path),
+                'title': video_info.get('title'),
+                'duration': video_info.get('duration'),
+                'resolution': video_info.get('resolution'),
+                'upload_date': video_info.get('upload_date'),
+            },
+            'audio': {
+                'path': str(audio_path),
+                'bitrate': audio_info.get('abr'),
+                'codec': 'mp3',
+            },
+        }
+        
+        with open(metadata_path, 'w', encoding='utf-8') as f:
+            json.dump(metadata, f, indent=2, ensure_ascii=False)
+        
+        return {
+            'data_path': str(download_dir),
+            'video_path': str(video_path),
+            'audio_path': str(audio_path) + ".mp3",
+            'metadata_path': str(metadata_path),
+        }
+        
+    except Exception as e:
+        raise RuntimeError(f"Media download error: {str(e)}")
+
+if __name__ == "__main__":
+    video_url = "https://www.youtube.com/watch?v=FK3dav4bA4s"
+    result = download_youtube_media(video_url, "./temp")
+    print(result)

app_srv/test.ipynb

@@ -0,0 +1,431 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "7f265e58",
+   "metadata": {},
+   "source": [
+    "# Download video and audio from YouTube"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "69ee0ec3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import yt_dlp\n",
+    "import os\n",
+    "import uuid\n",
+    "import json\n",
+    "from pathlib import Path\n",
+    "from typing import Dict, Any\n",
+    "from datetime import datetime\n",
+    "\n",
+    "def download_youtube_media(url: str, base_dir: str = \"./downloads\") -> Dict[str, Any]:\n",
+    "    \"\"\"\n",
+    "    Downloads video and audio from YouTube, saving them to a unique GUID folder.\n",
+    "    Metadata is saved in JSON format including download datetime and timezone.\n",
+    "    \n",
+    "    Args:\n",
+    "        url (str): YouTube video URL\n",
+    "        base_dir (str): Base download directory (default './downloads')\n",
+    "    \n",
+    "    Returns:\n",
+    "        dict: Dictionary with file paths and information:\n",
+    "            {\n",
+    "                'data_path': str,     # Path to download directory\n",
+    "                'video_path': str,     # Full path to video.mp4\n",
+    "                'audio_path': str,     # Full path to audio.mp3\n",
+    "                'metadata_path': str  # Full path to metadata.json\n",
+    "            }\n",
+    "    \n",
+    "    Raises:\n",
+    "        RuntimeError: If download fails\n",
+    "    \"\"\"\n",
+    "    try:\n",
+    "        # Generate GUID and create folder\n",
+    "        guid = str(uuid.uuid4())\n",
+    "        download_dir = Path(base_dir) / guid\n",
+    "        os.makedirs(download_dir, exist_ok=True)\n",
+    "        \n",
+    "        # File paths\n",
+    "        video_path = download_dir / \"video.mp4\"\n",
+    "        audio_path = download_dir / \"audio.mp3\"\n",
+    "        metadata_path = download_dir / \"metadata.json\"\n",
+    "\n",
+    "        # Record exact download start time\n",
+    "        download_datetime = datetime.now()\n",
+    "        current_timezone = download_datetime.astimezone().tzinfo\n",
+    "        \n",
+    "        # 1. Download video (MP4)\n",
+    "        video_opts = {\n",
+    "            'format': (\n",
+    "                \"bestvideo[height=720][ext=mp4]\"\n",
+    "                \"/worstvideo[height>720][ext=mp4]\"\n",
+    "                \"/bestvideo[height<=720][ext=mp4]\"\n",
+    "                ),\n",
+    "            'outtmpl': str(video_path),\n",
+    "            'quiet': True,\n",
+    "            'no_warnings': False,\n",
+    "            'restrict_filenames': True,\n",
+    "        }\n",
+    "        \n",
+    "        with yt_dlp.YoutubeDL(video_opts) as ydl:\n",
+    "            video_info = ydl.extract_info(url, download=True)\n",
+    "        \n",
+    "        # 2. Download audio (MP3)\n",
+    "        audio_opts = {\n",
+    "            'format': 'bestaudio/best',\n",
+    "            'outtmpl': str(audio_path),\n",
+    "            'quiet': True,\n",
+    "            'postprocessors': [{\n",
+    "                'key': 'FFmpegExtractAudio',\n",
+    "                'preferredcodec': 'mp3',\n",
+    "                'preferredquality': '128',\n",
+    "            }],\n",
+    "        }\n",
+    "        \n",
+    "        with yt_dlp.YoutubeDL(audio_opts) as ydl:\n",
+    "            audio_info = ydl.extract_info(url, download=True)\n",
+    "        \n",
+    "        # Format date and time for storage\n",
+    "        formatted_date = download_datetime.strftime('%Y-%m-%d')\n",
+    "        formatted_time = download_datetime.strftime('%H:%M:%S')\n",
+    "        \n",
+    "        # 3. Save metadata to JSON\n",
+    "        metadata = {\n",
+    "            'original_url': url,\n",
+    "            'guid': guid,\n",
+    "            'download_info': {\n",
+    "                'date': formatted_date,\n",
+    "                'time': formatted_time,\n",
+    "                'timezone': str(current_timezone),\n",
+    "                'datetime_iso': download_datetime.isoformat(),\n",
+    "            },\n",
+    "            'video': {\n",
+    "                'path': str(video_path),\n",
+    "                'title': video_info.get('title'),\n",
+    "                'duration': video_info.get('duration'),\n",
+    "                'resolution': video_info.get('resolution'),\n",
+    "                'upload_date': video_info.get('upload_date'),\n",
+    "            },\n",
+    "            'audio': {\n",
+    "                'path': str(audio_path),\n",
+    "                'bitrate': audio_info.get('abr'),\n",
+    "                'codec': 'mp3',\n",
+    "            },\n",
+    "        }\n",
+    "        \n",
+    "        with open(metadata_path, 'w', encoding='utf-8') as f:\n",
+    "            json.dump(metadata, f, indent=2, ensure_ascii=False)\n",
+    "        \n",
+    "        return {\n",
+    "            'data_path': str(download_dir.absolute()),\n",
+    "            'video_path': str(video_path.absolute()),\n",
+    "            'audio_path': str(audio_path.absolute()) + \".mp3\",\n",
+    "            'metadata_path': str(metadata_path),\n",
+    "        }\n",
+    "        \n",
+    "    except Exception as e:\n",
+    "        raise RuntimeError(f\"Media download error: {str(e)}\")\n",
+    "\n",
+    "if __name__ == \"__main__\":\n",
+    "    video_url = \"https://www.youtube.com/watch?v=FK3dav4bA4s\"\n",
+    "    downloaded_video = download_youtube_media(video_url, \"./temp\")\n",
+    "    print(downloaded_video)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f62e8b83",
+   "metadata": {},
+   "source": [
+    "# Split video to frames in jpg"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "id": "93ab88d9",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{}\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[av1 @ 0x55a7f1349040] Your platform doesn't suppport hardware accelerated AV1 decoding.\n",
+      "[av1 @ 0x55a7f1349040] Failed to get pixel format.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Your platform doesn't suppport hardware accelerated AV1 decoding.\n",
+      "[av1 @ 0x55a7f1349040] Failed to get pixel format.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f1349040] Missing Sequence Header.\n",
+      "[av1 @ 0x55a7f34683c0] Your platform doesn't suppport hardware accelerated AV1 decoding.\n",
+      "[av1 @ 0x55a7f34683c0] Failed to get pixel format.\n"
+     ]
+    }
+   ],
+   "source": [
+    "import os\n",
+    "from pathlib import Path\n",
+    "from typing import Dict\n",
+    "import cv2\n",
+    "\n",
+    "def extract_frames_with_timestamps(\n",
+    "    video_path: str,\n",
+    "    output_dir: str,\n",
+    "    time_step: float = 1.0,\n",
+    "    quality: int = 95,\n",
+    "    frame_prefix: str = \"frame\"\n",
+    ") -> Dict[str, str]:\n",
+    "    \"\"\"\n",
+    "    Extracts frames from video at specified time intervals using OpenCV.\n",
+    "    \n",
+    "    Args:\n",
+    "        video_path: Full path to the video file\n",
+    "        output_dir: Directory where 'frames' subfolder will be created\n",
+    "        time_step: Interval in seconds between captured frames\n",
+    "        quality: JPEG quality (1-100)\n",
+    "        frame_prefix: Prefix for frame filenames\n",
+    "\n",
+    "    Returns:\n",
+    "        Dictionary mapping timestamps (HH:MM:SS) to frame paths\n",
+    "    \"\"\"\n",
+    "    result = {}\n",
+    "    \n",
+    "    try:\n",
+    "        video_path = Path(video_path).absolute()\n",
+    "        output_dir = Path(output_dir).absolute()\n",
+    "        \n",
+    "        if not video_path.exists():\n",
+    "            raise ValueError(f\"Video file not found: {video_path}\")\n",
+    "\n",
+    "        frames_dir = output_dir / \"frames\"\n",
+    "        frames_dir.mkdir(parents=True, exist_ok=True)\n",
+    "\n",
+    "        # Initialize video capture\n",
+    "        cap = cv2.VideoCapture(str(video_path))\n",
+    "        if not cap.isOpened():\n",
+    "            raise RuntimeError(\"Could not open video file\")\n",
+    "\n",
+    "        fps = cap.get(cv2.CAP_PROP_FPS)\n",
+    "        if fps <= 0:\n",
+    "            raise RuntimeError(\"Invalid frame rate\")\n",
+    "\n",
+    "        frame_interval = max(1, int(round(fps * time_step)))\n",
+    "        frame_count = 0\n",
+    "\n",
+    "        while True:\n",
+    "            ret, frame = cap.read()\n",
+    "            if not ret:\n",
+    "                break\n",
+    "\n",
+    "            if frame_count % frame_interval == 0:\n",
+    "                current_time = frame_count / fps\n",
+    "                hh = int(current_time // 3600)\n",
+    "                mm = int((current_time % 3600) // 60)\n",
+    "                ss = current_time % 60\n",
+    "                \n",
+    "                timestamp = f\"{hh:02d}:{mm:02d}:{ss:06.3f}\"\n",
+    "                safe_timestamp = timestamp.replace(':', '_').replace('.', '_')\n",
+    "                frame_path = frames_dir / f\"{frame_prefix}_{safe_timestamp}.jpg\"\n",
+    "                \n",
+    "                if cv2.imwrite(str(frame_path), frame, [cv2.IMWRITE_JPEG_QUALITY, quality]):\n",
+    "                    result[timestamp] = str(frame_path)\n",
+    "\n",
+    "            frame_count += 1\n",
+    "\n",
+    "        cap.release()\n",
+    "        return result\n",
+    "\n",
+    "    except Exception as e:\n",
+    "        for path in result.values():\n",
+    "            try: os.remove(path)\n",
+    "            except: pass\n",
+    "        raise RuntimeError(f\"Frame extraction failed: {str(e)}\")\n",
+    "\n",
+    "\n",
+    "if __name__ == \"__main__\":\n",
+    "    result = extract_frames_with_timestamps(\"/workspaces/Video_Analyser/app_srv/temp/3dee4d5e-b1f7-40f4-b4c6-84e55ad7d7b4/video.mp4\", \"/workspaces/Video_Analyser/app_srv/temp/3dee4d5e-b1f7-40f4-b4c6-84e55ad7d7b4\")\n",
+    "\n",
+    "    print(result)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.10"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

app_srv/video_processor.py

@@ -0,0 +1,92 @@
+import os
+import sys
+from pathlib import Path
+from typing import Optional, Tuple
+import PyNvVideoCodec as nvc
+
+def extract_frames_from_video(
+    video_path: str,
+    output_dir: str,
+    time_step: float = 1.0,
+    quality: int = 95,
+    frame_prefix: str = "frame_"
+) -> Optional[str]:
+    """
+    Extracts frames from MP4 video at specified time intervals using PyNvVideoCodec (NVIDIA GPU acceleration).
+
+    Args:
+        video_path: Full path to the MP4 video file.
+        output_dir: Directory where 'frames' subfolder will be created.
+        time_step: Interval in seconds between captured frames (can be fractional).
+        quality: JPEG quality (1-100).
+        frame_prefix: Prefix for frame filenames.
+
+    Returns:
+        Path to the frames directory if successful, None otherwise.
+
+    Raises:
+        ValueError: If input parameters are invalid.
+        RuntimeError: If frame extraction fails.
+    """
+    try:
+        # Validate input parameters
+        video_path = Path(video_path)
+        if not video_path.exists():
+            raise ValueError(f"Video file not found: {video_path}")
+        if not video_path.suffix.lower() == '.mp4':
+            raise ValueError("Input file must be an MP4 video")
+        
+        if time_step <= 0:
+            raise ValueError("time_step must be positive")
+        if quality < 1 or quality > 100:
+            raise ValueError("quality must be between 1 and 100")
+
+        # Create output directory
+        frames_dir = Path(output_dir) / "frames"
+        frames_dir.mkdir(parents=True, exist_ok=True)
+
+        # Initialize NVDecoder
+        nv_dec = nvc.PyNvDecoder(str(video_path), 0)  # 0 = GPU device ID
+
+        # Get video info
+        width = nv_dec.Width()
+        height = nv_dec.Height()
+        fps = nv_dec.Framerate()
+        total_frames = nv_dec.Framecount()
+        duration = total_frames / fps if fps > 0 else 0
+
+        print(f"Video Info: {width}x{height}, FPS: {fps:.2f}, Duration: {duration:.2f}s")
+
+        # Calculate frame interval based on time_step
+        frame_interval = int(round(fps * time_step))
+        if frame_interval < 1:
+            frame_interval = 1
+
+        frame_count = 0
+        saved_count = 0
+
+        # Process frames
+        while True:
+            # Decode frame
+            raw_frame = nv_dec.DecodeSingleFrame()
+            if raw_frame.empty():
+                break  # End of video
+
+            # Save frame at specified intervals
+            if frame_count % frame_interval == 0:
+                output_path = frames_dir / f"{frame_prefix}{saved_count:06d}.jpg"
+                success = nvc.ConvertToJpeg(raw_frame, str(output_path), quality)
+                if not success:
+                    print(f"Failed to save frame {frame_count}", file=sys.stderr)
+                else:
+                    saved_count += 1
+
+            frame_count += 1
+
+        print(f"Extracted {saved_count} frames (every {frame_interval} frames)")
+        return str(frames_dir)
+
+    except Exception as e:
+        raise RuntimeError(f"Error extracting frames: {str(e)}")
+
+if __name__ == "__main__":

packages.txt

@@ -13,4 +13,5 @@ python-is-python3
 wget
 zlib1g
 net-tools
-lsof
+lsof
+build-essential

requirements.txt

@@ -3,4 +3,7 @@ datasets
 evaluate
 gradio
 gradio[mcp]
-PyNvVideoCodec
+ipython
+ipywidgets
+ffmpeg-python
+opencv-python