Comic Compressor

A high-performance Rust application for compressing comic book files (CBR/CBZ/PDF) with parallel processing. Converts images to WebP format for optimal file size reduction while maintaining visual quality.

Features

• ✅ Cross-platform compatibility - Works on Mac, Windows, and Linux
• ✅ Parallel processing - Processes multiple files and images simultaneously
• ✅ Multiple format support - Handles CBR (RAR), CBZ (ZIP), and PDF files with automatic format detection
• ✅ Advanced PDF support - Direct image extraction from PDFs (JPEG, PNG, CMYK, Grayscale)
• ✅ Automatic folder processing - Processes all comic files in a directory by default
• ✅ Glob pattern support - Process selective files using patterns (e.g., "ABC*.cbr")
• ✅ Progress visualization - Docker-style layered progress display
• ✅ Smart compression - Skips images that don't benefit from compression
• ✅ Intelligent file preservation - Keeps already well-compressed files unchanged (especially RAR archives)
• ✅ Robust error handling - Continues processing even with corrupt images
• ✅ CBR output format - Always outputs .cbr files regardless of input format
• ✅ Standalone binary - No external dependencies required

Installation

Download Pre-built Binaries (Recommended)

Download the latest release for your platform from GitHub Releases:

• Linux (x86_64): compress_comics-x86_64-unknown-linux-gnu.tar.gz
• Windows (x86_64): compress_comics-x86_64-pc-windows-msvc.zip
• macOS (Intel): compress_comics-x86_64-apple-darwin.tar.gz
• macOS (Apple Silicon): compress_comics-aarch64-apple-darwin.tar.gz

Linux/macOS Installation:

# Download and extract (replace URL with latest release)
tar -xzf compress_comics-x86_64-unknown-linux-gnu.tar.gz
chmod +x compress_comics
sudo mv compress_comics /usr/local/bin/  # Optional: add to PATH

Windows Installation:

1. Download the .zip file for Windows
2. Extract compress_comics.exe
3. Place in a folder that's in your PATH or run directly

From Source

git clone https://github.com/erikvullings/compress_comics.git
cd compress_comics
cargo build --release

The compiled binary will be available at target/release/compress_comics

From crates.io

cargo install compress_comics

Usage

Process a single file

compress_comics comic.cbz --quality 85
compress_comics comic.cbr --quality 85
compress_comics comic.pdf --quality 85

Process all comic files in current directory (default behavior)

compress_comics

Process all comic files in a specific directory

compress_comics /path/to/comics/

Process files using glob patterns

# Simple patterns (automatically searches recursively)
compress_comics --glob-pattern "ABC*.cbr"        # Files starting with "ABC" anywhere
compress_comics --glob-pattern "*Killer*.cbr"    # Files containing "Killer" anywhere

# Explicit recursive patterns
compress_comics --glob-pattern "**/De Killer*.cbr"  # Recursive search for "De Killer"
compress_comics --glob-pattern "**/*Volume*/*.cbz"  # Complex nested patterns

# Absolute path patterns
compress_comics --glob-pattern "/full/path/**/Killer*.cbr"  # Full path search

# Current directory only
compress_comics --glob-pattern "*.pdf"              # PDF files in current directory

# Debug your patterns
compress_comics --glob-pattern "pattern" --verbose  # Shows found files before processing

Custom settings

compress_comics comics/ --quality 75 --target-height 1600

Rename original files (convenient workflow)

compress_comics comics/ --rename-original --quality 85
# Result: Original files become *_original.ext, compressed files get clean names

Skip already well-compressed files

compress_comics comics/ --min-savings 10.0  # Only compress if >10% savings possible
# Files with less potential savings are left unchanged (especially useful for RAR archives)

Options

• --quality / -q: WebP quality (1-100, default: 90)

  • 85-95: High quality, moderate compression
  • 65-80: Balanced quality and size
  • 40-60: Small files, lower quality

• --target-height / -H: Target height for images in pixels (default: 1800)

• --max-dimension / -m: Maximum dimension fallback (default: 1200)

• --rename-original / -r: Rename original file to <name>_original.<ext> and give compressed file the original name

• --glob-pattern / -g: Process only files matching the glob pattern (e.g., "ABC*.cbr", "*.pdf")

• --min-savings: Minimum compression savings percentage required to keep compressed file (default: 5.0)

• --verbose / -v: Enable detailed output with warnings for debugging (disabled by default for clean output)

Glob Pattern Tips

Glob patterns use wildcards to match file paths:

• * matches any characters within a directory name
• ** matches any number of directories (recursive)
• ? matches a single character
• [abc] matches any character in brackets

Common Scenarios

Find files by series name anywhere in directory tree:

compress_comics --glob-pattern "**/De Killer*.cbr"

Find files in specific nested structure:

compress_comics --glob-pattern "**/Striparchief*/**/De Killer*.cbr"

Find files with specific volume numbers:

compress_comics --glob-pattern "**/*Volume 1*.cbr"
compress_comics --glob-pattern "**/*S0[1-3]*.cbr"  # Seasons 1-3

Use verbose mode to debug patterns:

compress_comics --glob-pattern "**/Killer*.cbr" --verbose

This shows exactly which files were found before processing.

Output

Default Behavior

The tool creates new files with the suffix optimized_webp_q{quality}.cbr:

• Input: MyComic.cbz → Output: MyComic optimized_webp_q90.cbr
• Input: MyComic.cbr → Output: MyComic optimized_webp_q90.cbr
• Input: MyComic.pdf → Output: MyComic optimized_webp_q90.cbr

With --rename-original Option

When using --rename-original, the compressed file takes the original name:

• MyComic.cbz → MyComic_original.cbz (backup) + MyComic.cbr (compressed)
• MyComic.cbr → MyComic_original.cbr (backup) + MyComic.cbr (compressed)
• MyComic.pdf → MyComic_original.pdf (backup) + MyComic.cbr (compressed)

Performance Features

Parallel Processing

• Files are processed in parallel using all available CPU cores
• Images within each file are also processed in parallel
• Progress is displayed for each file simultaneously

Smart Compression

• Only compresses images when WebP provides size benefits
• Automatically detects two-page spreads and adjusts processing
• Skips already well-compressed images

Memory Efficient

• Uses temporary directories for processing
• Automatic cleanup after completion
• Streaming archive processing

Progress Display

The tool shows progress similar to Docker image downloads:

🚀 Found 3 comic file(s) to process
Settings: Quality=90, Target Height=1800px
-----------------------------------------------------
⠋ [00:01:23] [████████████████████████████████████████] 2/3 files (00:00:45)
  📖 Comic1.cbz [████████████████████████████████] 100%
  📖 Comic2.cbz [████████████████░░░░░░░░░░░░░░░░] 65%
  📖 Comic3.cbz [████░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 15%

Summary Report

After processing, the tool provides a detailed summary:

📊 Processing Summary:
-----------------------------------------------------
📖 Comic1.cbz: 45.2% savings (23 images processed, 2 skipped)
📖 Comic2.cbz: 38.7% savings (18 images processed, 1 skipped)

🎯 Overall Results:
   Total files processed: 2
   Total images processed: 41
   Total images skipped: 3
   Overall size reduction: 42.1%
   Original size: 125.43 MB
   Compressed size: 72.65 MB

💡 1 file(s) were already well-compressed and showed minimal improvement.

Real-World Results

CBR/CBZ Files

📖 Amber Blake - 01.cbr: 61.1% savings (77.9 MB saved, 104 images processed, 0 skipped)
📖 Auschwitz - 01.cbr: 67.9% savings (74.9 MB saved, 84 images processed, 0 skipped)

Overall size reduction: 64.3% (152.8 MB saved)
Original size: 237.8 MB → Final size: 85.0 MB

PDF Files

📖 Brocéliande - Tome 67.pdf: 76.3% savings (91.1 MB saved, 55 images processed, 0 skipped)

Overall size reduction: 76.3% (91.1 MB saved)
Original size: 119.4 MB → Final size: 28.3 MB

With --rename-original Option

📖 comic1.cbr: 76.4% savings (84 images processed, 0 skipped)
📖 comic2.pdf: 81.1% savings (55 images processed, 0 skipped)

Before: comic1.cbr (115.75 MB), comic2.pdf (125.21 MB)
After:  comic1_original.cbr (backup), comic2_original.pdf (backup)
        comic1.cbr (27.35 MB), comic2.cbr (23.71 MB)

Total: 229.80 MB → 48.69 MB (78.8% savings)

Why These Results?

• PDF files often have the highest compression ratios because they typically contain uncompressed or lightly compressed images
• CBR/CBZ files vary depending on original compression - some modern files are already well-optimized
• WebP format provides excellent quality-to-size ratio, especially for comic book artwork
• --rename-original makes workflow seamless - no manual file management needed

Technical Details

• Language: Rust (standalone binary, no runtime dependencies)
• Image Processing: High-quality Lanczos3 resampling
• Compression: WebP lossy compression with configurable quality
• Archive Format: ZIP-based CBR files (universal comic reader compatibility)
• Extraction:
  • CBR files: Native RAR support with ZIP fallback for compatibility
  • CBZ files: Native ZIP extraction
  • PDF files: Direct embedded image extraction (JPEG, PNG, CMYK, Grayscale)
• Threading: Rayon for work-stealing parallelism

PDF Support Details

The tool provides comprehensive PDF support for comic books:

✅ Supported PDF Image Formats

• JPEG (DCTDecode): Direct extraction with no quality loss
• PNG/Compressed (FlateDecode): Decompression and reconstruction
• Raw RGB/Grayscale: Uncompressed pixel data extraction
• CMYK Images: Automatic conversion to RGB color space

⚠️ Unsupported PDF Formats

• CCITT Fax compression: Skipped with informative message
• Complex vector graphics: Only embedded raster images are extracted
• Text-only PDFs: No images to extract

Limitations

• Output uses ZIP compression for CBR files (not RAR compression, but maintains .cbr extension for compatibility)
• WebP format may not be supported by very old comic readers
• PDF vector graphics are not rasterized (only embedded images are extracted)

Building for Distribution

To build optimized binaries for distribution:

cargo build --release
strip target/release/compress_comics  # Optional: reduce binary size

The resulting binary is self-contained and can be distributed without any dependencies.