pybalt

Features

pybalt is a powerful and flexible tool for downloading media files from various platforms, including YouTube, X (formerly Twitter), Reddit, Instagram, TikTok, and more. It works using cobalt processing instances and serves both as a CLI and a Python module.

• Download media files to your desktop effortlessly using pybalt as a command-line interface
• Support for playlists, bulk downloads, and processing links from text files
• Runable web api that allows to try multiple instances at once for best latency
• Integrate pybalt into your Python projects with ease with just a few lines of code
• Host your own local cobalt instance for improved privacy and reliability
• Customize configuration with environment variables or the built-in config system

https://github.com/user-attachments/assets/cf5fd9a9-520b-4970-b8c2-42baa80d7523

⚙️ Installation

Install pybalt with pip:

pip install pybalt -U

Or install pybalt on Windows with the bat file included in the repo (if you don't have Python installed):

1. Open PowerShell or Command Prompt with administrator rights (to allow pip to create aliases cobalt and pybalt in the terminal, and install python if needed)
2. Type this command:

powershell -Command "Invoke-WebRequest -Uri https://raw.githubusercontent.com/nichind/pybalt/main/install.bat -OutFile install.bat; .\install.bat"

⚡️ Quickstart

[!NOTE] pybalt will automatically detect if you're running a local cobalt instance and use it first before trying public instances.

[!NOTE] If the cobalt alias isn't working in your terminal, use python -m pybalt <command> instead.

[!CAUTION] Remuxing (-r) requires ffmpeg to be installed on your device and added to your system path.

Command Line Usage

The CLI is intuitive and easy to use! Here are some examples:

1. Download a YouTube video at maximum resolution (-vQ max) and remux it (-remux):

cobalt "https://youtube.com/watch?v=DG2QqcHwNdE" -remux -vQ max

2. Remux an existing video file:

cobalt "C:/Users/username/Videos/video.mp4"

3. Download multiple videos from links in a text file:

cobalt "path/to/links.txt"

4. Download a video and open it immediately:

cobalt "https://youtube.com/watch?v=DG2QqcHwNdE" -o

5. Download a video and show it in File Explorer/Finder:

cobalt "https://youtube.com/watch?v=DG2QqcHwNdE" -s

6. Specify quality, format, and download location:

cobalt "https://youtube.com/watch?v=DG2QqcHwNdE" -vQ 1080 -aF mp3 --audioBitrate 320 -fp "C:/Downloads"

Use cobalt -h to see all available options.

Managing Instances

pybalt can work with multiple cobalt instances, including your own local instance:

1. List configured instances:

cobalt -li

2. Add a new instance:

cobalt -ai "https://example-instance.com" "optional-api-key"

3. Remove an instance:

cobalt -ri 1

Local Instance Management

Run your own cobalt instance with Docker:

1. Set up a local instance:

cobalt -ls

2. Start/stop your local instance:

cobalt -lstart
cobalt -lstop

3. Check local instance status:

cobalt -lstatus

💻 Python Module Integration

pybalt can be easily integrated into your Python projects:

Basic Download

from pybalt import download
from asyncio import run
async def main():
# Simple download with default settings
file_path = await download("https://youtube.com/watch?v=DG2QqcHwNdE")
print(f"Downloaded to: {file_path}")
# Download with custom parameters
file_path = await download(
    "https://youtube.com/watch?v=DG2QqcHwNdE",
    videoQuality="1080",
    audioFormat="mp3",
    audioBitrate="320",
    filenameStyle="pretty",
    remux=True
)
print(f"Downloaded to: {file_path}")

run(main())

Advanced Usage with InstanceManager

from pybalt.core.wrapper import InstanceManager
from asyncio import run
async def main():
# Create an instance manager
manager = InstanceManager(debug=True)
# Get a list of available instances
instances = await manager.get_instances()
print(f"Found {len(instances)} available instances")

# Download a file using the first available instance
file_path = await manager.download(
    url="https://youtube.com/watch?v=DG2QqcHwNdE",
    videoQuality="1080",
    remux=True
)
print(f"Downloaded to: {file_path}")

# Bulk download multiple URLs
urls = [
    "https://youtube.com/watch?v=DG2QqcHwNdE",
    "https://youtube.com/watch?v=anotherVideo"
]

# Download multiple files concurrently
async for path in manager.download_generator(urls=urls, remux=True):
    print(f"Downloaded: {path}")

run(main())

Track Download Progress

from pybalt import download
from asyncio import run
async def main():
# Define a status callback function
async def status_callback(downloaded_size, total_size, download_speed, eta, **kwargs):
percent = (downloaded_size / total_size * 100) if total_size > 0 else 0
print(f"Downloaded: {downloaded_size / 1024 / 1024:.2f}MB / "
f"{total_size / 1024 / 1024:.2f}MB ({percent:.1f}%) at "
f"{download_speed / 1024 / 1024:.2f}MB/s, ETA: {eta:.0f}s")
# Define a completion callback
async def done_callback(file_path, downloaded_size, time_passed, **kwargs):
    print(f"Download completed in {time_passed:.2f}s")
    print(f"File saved to: {file_path}")

# Download with progress tracking
file_path = await download(
    "https://youtube.com/watch?v=DG2QqcHwNdE",
    status_callback=status_callback,
    done_callback=done_callback
)

run(main())

Using Status Parent

from pybalt import download
from asyncio import run, create_task
class StatusParent:
def init(self):
self.downloaded_size = 0
self.total_size = 0
self.download_speed = 0
self.eta = 0
self.completed = False
self.file_path = None
async def main():
# Create status parent object to track progress
status = StatusParent()
# Start download as a background task
task = create_task(download(
    "https://youtube.com/watch?v=DG2QqcHwNdE",
    status_parent=status
))

# Monitor progress while download is running
while not task.done():
    if status.total_size > 0:
        percent = status.downloaded_size / status.total_size * 100
        print(f"Progress: {percent:.1f}% - Speed: {status.download_speed / 1024 / 1024:.2f}MB/s")
    await run(sleep(1))

# Get the result
file_path = await task
print(f"Downloaded to: {file_path}")

run(main())

⚙️ Configuration System

pybalt features a comprehensive configuration system for customizing behavior:

Command-line Configuration Interface

Open the interactive configuration interface:

cobalt -c

Get a specific configuration value:

cobalt -gc "network" "timeout"

Set a configuration value:

cobalt -sc "network" "timeout" "30"

Configuration File Locations

• Windows: %APPDATA%\pybalt\settings.ini
• macOS: ~/Library/Application Support/pybalt/settings.ini
• Linux: ~/.config/pybalt/settings.ini

Main Configuration Sections

• general: General settings like user agent and debug mode
• network: Connection settings like timeouts, proxies, and retries
• instances: Cobalt instance settings
• user_instances: User-defined cobalt instances
• paths: Download paths and file locations
• local: Local instance configuration
• ffmpeg: Remuxing settings
• display: UI/progress display settings

📦 Environment Variables

You can configure pybalt with environment variables:

PYBALT_CONFIG_DIR=path/to/config/dir        # Custom config directory
PYBALT_CONFIG_PATH=path/to/settings.ini     # Custom config file path
Section-specific settings
PYBALT_GENERAL_DEBUG=True                   # Enable debug mode
PYBALT_GENERAL_USER_AGENT=custom-agent      # Custom user agent
PYBALT_NETWORK_TIMEOUT=30                   # Network timeout
PYBALT_NETWORK_USE_SYSTEM_PROXY=True        # Use system proxy
PYBALT_NETWORK_PROXY=http://my-proxy:8080   # Custom proxy
PYBALT_PATHS_DEFAULT_DOWNLOADS_DIR=/path    # Custom download directory
PYBALT_LOCAL_LOCAL_INSTANCE_PORT=9000       # Local instance port

You can also set variables in a .env file in your project directory.

🛠️ Local Instance

pybalt can set up and manage a local cobalt instance using Docker:

Setting Up a Local Instance

cobalt -ls

This interactive wizard will guide you through:

• Setting a port for your instance
• Configuring API key authentication (optional)
• Setting duration limits
• Configuring proxy settings
• Setting up cookies for authenticated services

Managing Your Local Instance

Start your local instance:

cobalt -lstart

Stop your local instance:

cobalt -lstop

Check the status:

cobalt -lstatus

Restart your instance:

cobalt -lrestart

🌐 API Server

pybalt includes a built-in API server that allows you to interact with multiple cobalt instances through a single HTTP request. It also provides a web UI for easy downloading.

Starting the API Server

Start the API server in detached mode:

cobalt --api-start

Specify a custom port (default is 8009):

cobalt --api-port 8080 --api-start

Managing the API Server

Check if the API server is running:

cobalt --api-status

Stop the running API server:

cobalt --api-stop

API Endpoints

The API server exposes the following endpoints:

• GET /: Returns information about the API server, including version and instance count
• POST /: Accepts a JSON payload with a URL and parameters to download content and returns a JSON response with download information
• GET /ui: Serves a web-based user interface for downloading content

Web UI

The API server includes a comprehensive web UI that allows you to:

• Download videos from various platforms through a user-friendly interface
• Configure advanced settings such as video quality, audio format, and file naming
• Track the download progress
• Ignore specific instances that might be causing issues
• Copy direct download links

Access the web UI at http://localhost:8009/ui (or your custom port).

API Usage Examples

Example API Request:

curl -X POST "http://localhost:8009/" \
     -H "Content-Type: application/json" \
     -d '{"url": "https://www.youtube.com/watch?v=DG2QqcHwNdE", "videoQuality": "1080", "audioFormat": "mp3"}'

Example API Response:

{
  "status": "tunnel",
  "url": "http://download-link.mp4",
  "instance_info": {
    "url": "https://instance-url.com",
  },
}

👥 Used by

pybalt is used by the following projects:

• download.nichind.dev - Website for downloading media files from various platforms

👥 Contributors

🌟 That's it!

I spent too much time on this project... please consider leaving a :star: if you like it!