download_cbz_interactive.py

This is a Python script for downloading manga chapters from cloudme.one and packaging them as CBZ files. Supports Windows / macOS / Linux. Current version v1.1.

Features

download_cbz_interactive.py is a manga chapter download tool for cloudme.one. It fetches manga information via API, downloads chapter images, and automatically packages them into CBZ format files for use with various comic readers.

Characteristics

• Interactive or command-line mode for selecting chapter ranges to download
• Automatically fetches manga and chapter information via API
• Downloads high-quality images through img.cloudme.one proxy (requires cf_clearance)
• CDN direct fallback download (no Cookie required, but lower image quality)
• Multi-threaded image downloads, multi-task parallel chapter downloads
• Automatically packages into CBZ files (ZIP format, compatible with comic readers)
• Supports HTTP/SOCKS5 proxy
• Configurable image download threads (1-32) and parallel chapter count (1-5)
• Supports passing cf_clearance via command line, no interactive input needed
• Supports skipping confirmation to start download directly (-y/--yes)
• Auto-detects and installs missing dependencies
• Cross-platform compatible (Windows / macOS / Linux)
• Terminal Unicode auto-adaptation (falls back to ASCII symbols on unsupported terminals)
• Download progress bar display
• Automatically skips existing CBZ files

Dependencies

Dependency	Purpose	Install
Python 3.6+	Runtime environment	—
curl_cffi	Bypass Cloudflare protection, download images	pip install curl_cffi

The script automatically detects dependencies on first run and prompts to install if missing.

Usage

Download Script

CNB:

wget https://cnb.cool/SDCOM/shit/-/git/raw/main/script/download_cbz_interactive.py

GitHub:

wget https://github.com/SDCOM-0415/shit/raw/refs/heads/main/script/download_cbz_interactive.py

Interactive Mode

Run the script directly to enter interactive mode and follow the prompts:

python download_cbz_interactive.py

Non-interactive Mode

Provide all necessary information via command-line arguments, and the script starts downloading directly without interactive input:

# Download chapters 101-120, use proxy, 16 threads, 3 parallel chapters
python download_cbz_interactive.py \
  -o ~/Comics \
  -m 10 \
  -c 101-120 \
  --cf xxxxxxxx \
  -p http://127.0.0.1:7890 \
  -t 16 \
  -j 3 \
  --strip-prefix \
  -y

When -m, -c, and -o are all provided, the script automatically enters non-interactive mode. Parameters not provided will still be prompted in interactive mode.

Specify Download Directory

Specify the save directory for CBZ files via command line arguments:

# Windows
python download_cbz_interactive.py -o D:\Comics

# macOS / Linux
python download_cbz_interactive.py -o ~/Comics

# Can also be combined with other parameters
python download_cbz_interactive.py -o ~/Comics -m 10 -c all -y

The default download directory is the directory where the script is located.

Command Line Arguments

Parameter	Short	Description
--output DIR	-o	Specify the save directory for CBZ files (default: script directory)
--mid ID	-m	Manga ID (e.g., 10 in cloudme.one/refs/10)
--chapters RANGE	-c	Chapter range, supports 101-120, 101,103,105, all
--cf TOKEN		cf_clearance cookie value (for downloading high-quality images)
--proxy URL	-p	HTTP/SOCKS5 proxy address (e.g., http://127.0.0.1:7890)
--threads N	-t	Number of image download threads (1-32, default 8)
--jobs N	-j	Number of parallel chapter downloads (1-5, default 5)
--strip-prefix		Remove prefix before _ in title (e.g., SomeManga_Chapter101 → Chapter101)
--yes	-y	Skip confirmation prompt and start download directly

Execution Flow

Interactive Mode Flow

1. Confirm download directory: Displays the current download directory, enter a new path to change it, or press Enter to confirm
2. (Optional) Set proxy: Enter HTTP/SOCKS5 proxy address, or press Enter to skip
3. (Optional) Set thread count: Set the number of image download threads (1-32), or press Enter to use default 8
4. (Optional) Set parallel tasks: Set the number of chapters to download in parallel (1-5), or press Enter to use default 5
5. Enter manga ID: Enter the manga ID from the cloudme.one website (the number in the URL, e.g., 10 in cloudme.one/refs/10)
6. Fetch manga information: Automatically fetches manga title and full chapter list via API
7. Prefix handling (optional): If chapter titles contain _ separators, you'll be prompted whether to remove the prefix
8. Configure Cloudflare Cookie (optional): Provide cf_clearance cookie to download high-quality images; skip to use CDN direct download for lower quality
9. Select chapters: Enter the chapter range to download, supporting the following formats:
   • Single chapter: 101
   • Range: 101-120
   • Multiple: 101,103,105
   • Mixed: 101-105,108,110-112
   • All: all
10. Confirm and download: Confirm the information and start downloading, with progress and result summary displayed

Non-interactive Mode

When -o, -m, and -c are all provided, the script skips all interactive prompts and starts downloading directly. Other parameters (-p, -t, -j, --cf, --strip-prefix, -y) can be optionally provided as needed.

How to Get cf_clearance

High-quality images require downloading through a Cloudflare-protected proxy, which needs a cf_clearance cookie:

1. Open cloudme.one in your browser
2. Press F12 to open Developer Tools
3. Switch to the Application tab
4. Find Cookies → https://cloudme.one on the left
5. Find the cf_clearance field and copy its value

TIP

If you don't provide cf_clearance, the script will still work, but will download lower quality images via CDN direct connection. You can also pass it via the --cf parameter on the command line, without manually pasting it in interactive mode.

Notes

• The script requires Python 3.6 or higher
• curl_cffi is used to bypass Cloudflare protection and is a core dependency
• When downloading a large number of chapters, the script automatically adds intervals between chapters to avoid too many requests
• Existing CBZ files will be automatically skipped and will not be re-downloaded
• Windows users experiencing garbled text in the terminal will have UTF-8 support automatically enabled by the script
• It is recommended that the image download thread count (-t) does not exceed 32, and the parallel chapter count (-j) does not exceed 5, to avoid being rate-limited

Project Repository

CNB: https://cnb.cool/SDCOM/shit/-/blob/main/script/download_cbz_interactive.py

GitHub: https://github.com/SDCOM-0415/shit/blob/main/script/download_cbz_interactive.py

© Author

SDCOM