ebany video bot

This Telegram bot is designed to easily download media files via links from popular social networks.

The bot automatically downloads videos in the best available quality, adapts them for Telegram's native player, extracts audio tracks, and seamlessly handles TikTok and Instagram photos.

Features

Multi-platform:

• Supports links from YouTube (including Shorts), TikTok, Instagram (Reels/Posts), and SoundCloud.
• Offers a choice: download as Video or Audio. (For SoundCloud, you can also download the Cover Art).

Smart TikTok & Instagram Integration:

• Auto-detection: Automatically detects if a link is a standard video or a photo publication and provides context-aware menus.
• Photo: Download photos as a native Telegram Album (compressed for quick viewing) or as uncompressed Document files (original quality).
• Bypasses captchas and extracts watermark-free media and original MP3 audio via the TikWM API and headless browser scraping.

Video Processing:

• Forced conversion and codec selection to H.264 (MP4) for seamless playback directly in the Telegram chat.
• Supports video streaming (users can start watching before the file is fully downloaded to the cache).
• Original quality preserved: videos are sent as uncompressed documents to avoid Telegram's built-in compression.

Audio & Metadata Extraction:

• Converts audio tracks to MP3 format (320kbps).
• Automatically embeds high-quality cover art and metadata directly into downloaded MP3 files using yt-dlp postprocessors.

Logging:

• Maintains detailed logs with daily rotation in the logs/ directory.

Asynchronous:

• Downloading files does not block the bot's operation for other users, thanks to asyncio.to_thread and aiohttp.

Requirements

To run the bot, you need:

1. Python 3.8+
2. FFmpeg (system-level) — critical for yt-dlp to work properly (merging video and audio, codec conversion).

Installing FFmpeg:

Ubuntu/Debian:

sudo apt update && sudo apt upgrade && sudo apt install ffmpeg

Windows:

• Method 1 (Recommended): Open a terminal (PowerShell or CMD) and run:

winget install Gyan.FFmpeg

⚠️ "winget" command not found? > If you are using an older version of Windows 10, download and install the App Installer from the official GitHub releases (look for the .msixbundle file).

• Method 2 (Manual): Download the archive from the official repository, unzip it, and add the path to the bin folder to your system environment variables (PATH).

C:\ffmpeg\bin

MacOS:

brew install ffmpeg

Installation & Usage

1. Clone the repository

git clone https://github.com/Vovan999222/ebany_video_bot.git

cd ebany_video_bot

2. Create a virtual environment (Recommended)

python -m venv venv

Windows:

venv\Scripts\activate

Linux/Mac:

source venv/bin/activate

3. Install dependencies

You can install the libraries manually:

pip install aiogram yt-dlp Pillow aiohttp playwright

⚠️ Important step for Playwright (Instagram module): After installing the Python package, you must install the Chromium browser binaries and system dependencies:

playwright install chromium

If you are installing on a clean Linux server (e.g., Ubuntu), also run:

playwright install-deps

Or use the requirements.txt file and install it via:

pip install -r requirements.txt

4. Configuration

Open the config.py file and insert your bot token from @BotFather:

TOKEN = ""

Open instagram_photo_downloader.py and insert your Instagram credentials for the scraper to work:

IG_USERNAME = ""
IG_PASSWORD = ""

5. First run & Instagram Login (2FA)

When downloading from Instagram for the first time, the bot will launch a hidden browser and attempt to log in. Watch the terminal! If Instagram asks for a 2FA (Two-Factor Authentication) code, the bot will pause and prompt you in the console:

ИНСТАГРАМ ЗАПРОСИЛ КОД ПОДТВЕРЖДЕНИЯ (2FA)!
Введи код из SMS/WhatsApp и нажми Enter:

Once you enter the code, the session will be saved to ig_browser_state.json. All future downloads will work automatically without requiring passwords or codes.

6. Run the bot

python bot.py

Technical Details

• Limits: The bot processes and sends files up to 50 MB (Telegram Bot API restriction).
• Video: yt-dlp settings are forced to request bestvideo[ext=mp4][vcodec^=avc] format to exclude AV1/VP9 codecs, which are not supported by Telegram's in-app player.
• Audio: Processed via FFmpegExtractAudio with the MP3 codec.
• TikTok Parsing: Uses aiohttp to communicate with the TikWM API for fast, captcha-free data extraction. Pillow is used to process raw WebP image chunks and safely convert them to standard JPEGs.
• Instagram Scraping: Uses playwright in stealth mode to bypass Meta's scraping restrictions. Employs a custom "Visual Radar" algorithm to detect and extract high-quality images based on DOM rendering size, ignoring recommendations and avatars.
• Concurrency & Race Conditions: Implements random dynamic batch ID generation (uuid/random) for downloaded files to prevent race conditions when users send multiple links simultaneously.

License

This project is distributed under the MIT License