JellyDiscover

Automated, Personalized Recommendations for Jellyfin.

JellyDiscover is a standalone recommendation engine that analyzes watch history to create tailored "Recommended" libraries for every user on your server. It runs in the background, calculates scores based on Genres, Actors, and Directors, and seamlessly integrates these lists back into Jellyfin.

Features

• Personalized: Analyzes user history to recommend unwatched content they will actually like.
• Zero-Copy: Uses .strm files and Symlinks to link to your media. No extra storage space is consumed.
• Self-Healing: Automatically refreshes libraries and cleans up stale entries to keep Jellyfin in sync.
• Universal: Works on Windows, Linux, and Docker.
• Dashboard: A web-based GUI to manage settings, schedules, and path mappings.

📦 Installation

Option 1: Windows (Installer)

Best for: Standard Windows users.

1. Download the latest JellyDiscover_Setup_v1.1.0.exe from [Releases].
2. Run the installer.
3. Drive Configuration Step:
   • Local Drives: Select this if your media is on internal drives (C:\, D:\).
   • Network Drives: Select this if your media is on a NAS (Z:\, \\NAS\Share).
   • Note: If you select Network Drives, you will be prompted to enter your Windows Username and Password. This allows the background service to access your network shares securely.
4. Once installed, open http://localhost:5000 to finish the setup.

Option 2: Docker (Compose)

Best for: Unraid, Synology, TrueNAS Scale.

1. Download JellyDiscover_Docker_v1.1.0.zip from [Releases].
2. Extract it to a folder on your server.
3. Open docker-compose.yml and edit the volume mounts to match your media folders:

   volumes:
     - ./data:/app/data
     # UPDATE THESE LINES TO MATCH YOUR FOLDERS:
     - /mnt/media/movies:/media/movies:ro
     - /mnt/media/tv:/media/tv:ro

4. Run docker-compose up -d --build.
5. Open http://localhost:5000. You must configure the Path Substitutions table (see below).

Option 3: Linux (Manual)

Best for: Ubuntu, Debian, Fedora (Running natively).

1. Download & Extract: Download JellyDiscover_Linux_v1.1.0.zip from [Releases].

   unzip JellyDiscover_Linux_v1.1.0.zip -d /opt/
   cd /opt/JellyDiscover

2. Install Dependencies:

   pip3 install -r requirements.txt

3. Setup Dashboard Service (Systemd):

   # The zip includes the service file pre-configured for /opt/JellyDiscover
   sudo cp linux/jellydiscover.service /etc/systemd/system/
   sudo systemctl enable jellydiscover
   sudo systemctl start jellydiscover

4. Setup Daily Schedule (Cron): Open your crontab (crontab -e) and add the following line to run the engine daily at 4:00 AM:

   0 4 * * * /usr/bin/python3 /opt/JellyDiscover/src/engine.py >> /var/log/jellydiscover.log 2>&1

⚙️ The Dashboard

Access the web interface at http://localhost:5000.

1. Connection

• Jellyfin URL: The address of your server (e.g., http://192.168.1.5:8096).
• API Key: Generate this in Jellyfin Dashboard > API Keys.

2. Path Substitutions (Crucial for Docker)

This table maps the file paths Jellyfin sees to the file paths JellyDiscover sees.

• Windows Users: Usually NOT needed. The installer handles permissions for you.
• Linux Users: Usually NOT needed. Paths usually match (e.g., /mnt/media).
• Docker Users: REQUIRED.
  • Example: Jellyfin sees /data/movies. Docker sees /media/movies.
  • Action: Add a rule mapping Remote /data/movies -> Local /media/movies.

3. Scoring Bias

Adjust how the recommendation algorithm weighs different factors.

• Seen Penalty: How much to penalize content the user has already watched (Prevents repeats).
• Collection Boost: Bonus points if the item belongs to a collection the user likes (e.g., suggesting "Harry Potter 2" if they watched "Harry Potter 1").

4. Maintenance Tab

• Logs: View live logs to troubleshoot connection or scanning issues.
• Cleaner Utility: If your libraries ever look "glitched" (e.g., duplicate entries or items that won't play), run this tool to wipe the database and start fresh.

🛠 Troubleshooting & Errors

Common Log Errors

Logs are located in C:\ProgramData\JellyDiscover\logs\JellyDiscover.log (Windows) or ./data/logs/JellyDiscover.log (Linux/Docker).

How to Uninstall

• Windows: Use the Uninstaller in Control Panel. It will ask if you want to run the Cleanup Tool to remove the Virtual Libraries from Jellyfin.

• Linux/Docker: Run python3 src/cleaner.py manually, then delete the folder.