get_youtube_comments

A lightweight, multi-threaded script that downloads (and continues) YouTube video comments for one or more channels using yt-dlp as the backend.

This repository provides a small command-line tool that:

• Lists videos for given channel IDs (in the order provided),
• Spawns multiple workers to download comments in parallel,
• Supports continuing partial downloads (useful for long-running collection jobs),
• Accepts a cookies file when needed to access age-gated or region-restricted content.

---

Features

• Parallel comment collection (configurable worker count).
• Works with multiple YouTube channels in a single run.
• Uses yt-dlp for robust video listing and comment extraction.
• Simple filesystem-based output (one folder per channel/video).
• Optional cookies support for authenticated or region-restricted access.
• NEW: Accepts channel IDs directly from a file (one ID per line).

---

Prerequisites

• Python 3.8+
• yt-dlp in your PATH (install with pip or follow official instructions):

pip install yt-dlp

• tqdm for progress bars:

pip install tqdm

---

Installation

Clone this repo (or copy the script) and install the Python dependency:

git clone <repo-url>
cd get_youtube_comments
pip install -r requirements.txt   # if you prefer a requirements file
# or
pip install yt-dlp tqdm

If you plan to access content behind login/age gate, export cookies using your browser (cookies.txt) and pass it with --cookies.

---

Usage

From command-line arguments

python get_comments.py --channel_ids <CHANNEL_ID_1> <CHANNEL_ID_2> ... --output <path/to/output> [--num_threads N] [--cookies cookies.txt]

From a file of channel IDs

Create a text file with one channel ID per line (ignore empty lines and # comments). Example channels.txt:

UC_x5XG1OV2P6uZZ5FSM9Ttw
# This is a comment line
UCYO_jab_esuFRV4b17AJtAw

Run with:

python get_comments.py --channels_file channels.txt --output <path/to/output>

Mix both sources

You can provide both inline IDs and a file. Duplicates are automatically removed:

python get_comments.py --channels_file channels.txt --channel_ids UC123... UC456... --output <path/to/output>

---

Command-line options

Option	Required	Description
--channel_ids	No	One or more YouTube channel IDs to collect from.
--channels_file	No	Path to a text file containing one channel ID per line.
--output	Yes	Root output folder where results will be written.
--num_threads	No	Number of worker threads (default: 4).
--cookies	No	Path to a cookies file (e.g., cookies.txt) for authenticated access.
-h, --help	No	Show usage and exit.

---

How it works

1. For each channel ID from CLI or file, the script asks yt-dlp to list the channel's videos.
2. The videos are queued (preserving channel order).
3. The script spawns N worker threads (--num_threads) that pull the next video from the queue and run yt-dlp --write-comments to fetch comments.
4. Each worker writes a file (or files) under the provided --output directory. Partially downloaded channels are preserved so the process can be continued later without re-downloading everything.

---

Contributing

Contributions, bug reports and improvements are welcome.