unzip

A fast, reliable unzip utility written in Rust. Info-ZIP compatible with modern enhancements.

Features

• Fast: Memory-mapped I/O for large files, optimized release builds with LTO
• Reliable: Memory-safe Rust, comprehensive CRC verification, extensive test coverage
• Compatible: Drop-in replacement for Info-ZIP unzip with matching CLI options
• Modern: Progress bars, glob patterns (**/*.rs), colored output
• Full-featured: All major Info-ZIP options supported

Supported Operations

• Extract archives with progress visualization
• List contents (short and verbose formats)
• Zipinfo mode for detailed archive inspection
• Test archive integrity with CRC verification
• Extract to stdout/pipe
• Selective extraction with glob patterns
• Exclude files with patterns
• Freshen/update modes
• Preserve Unix permissions and timestamps

Compression Support

• Deflate, Deflate64
• LZMA, LZMA2
• Bzip2
• Zstd
• AES encrypted archives (with password via -P option)

Installation

From source

git clone https://github.com/yourusername/unzip.git
cd unzip
cargo build --release

The binary will be at target/release/unzip.

Using Cargo

cargo install --path .

Usage

unzip [OPTIONS] <FILE> [PATTERN]...

Arguments

Argument	Description
<FILE>	Path to the ZIP file to extract
[PATTERN]...	Files to extract (supports glob patterns)

Options

Option	Short	Description
--directory <DIR>	-d	Extract files to specified directory
--list	-l	List contents (short format)
--verbose	-v	List contents (verbose format with compression ratio)
--test	-t	Test archive integrity
--pipe	-p	Extract to stdout (for piping)
--comment	-z	Display archive comment only
--zipinfo [MODE]	-Z	Zipinfo mode: detailed archive information (see modes below)
--overwrite	-o	Overwrite existing files without prompting
--never-overwrite	-n	Never overwrite existing files
--freshen	-f	Only update existing files (don't create new)
--update	-u	Update files (freshen + create new if needed)
--junk-paths	-j	Extract without directory structure
--case-insensitive	-C	Match filenames case-insensitively
--lowercase	-L	Convert filenames to lowercase
--no-timestamps	-D	Skip restoring file and directory timestamps
--quiet	-q	Quiet mode (-q less output, -qq minimal)
--threads <NUM>	-T	Number of threads (default: auto)
--exclude <PATTERN>	-x	Exclude files matching pattern
--password <PASSWORD>	-P	Password for encrypted files (insecure, visible in process list)
--help	-h	Print help
--version	-V	Print version

Zipinfo Modes

Mode	Description
-Z or -Z s	Short format (default): Unix ls -l style with compression method
-Z m	Medium format: Short format + compression percentage
-Z l	Long format: Short format + compressed size in bytes
-Z v	Verbose format: Detailed multi-line information per file
-Z 1	Filenames only: One filename per line (no headers)
-Z 2	Filenames with headers: Filenames one per line with headers/trailers

Examples

Basic Operations

# Extract all files to current directory
unzip archive.zip

# Extract to a specific directory
unzip -d /tmp/output archive.zip

# List archive contents
unzip -l archive.zip

# List with compression details
unzip -v archive.zip

# Test archive integrity
unzip -t archive.zip

Zipinfo Mode

# Default zipinfo output (short format)
unzip -Z archive.zip

# Medium format with compression percentage
unzip -Z m archive.zip

# Long format with compressed sizes
unzip -Z l archive.zip

# Verbose detailed information
unzip -Z v archive.zip

# Just filenames (for scripting)
unzip -Z 1 archive.zip

Selective Extraction

# Extract only .txt files
unzip archive.zip '*.txt'

# Extract all Rust source files
unzip archive.zip '**/*.rs'

# Extract everything except logs
unzip archive.zip -x '*.log' -x '*.tmp'

# Extract specific file to stdout
unzip -p archive.zip config.json | jq .

Overwrite Control

# Overwrite all existing files
unzip -o archive.zip

# Never overwrite (skip existing)
unzip -n archive.zip

# Only update files that are older
unzip -u archive.zip

# Only freshen existing files (don't create new)
unzip -f archive.zip

Other Options

# Extract without directory structure (flatten)
unzip -j archive.zip

# Case-insensitive pattern matching
unzip -C archive.zip '*.TXT'

# Convert filenames to lowercase
unzip -L archive.zip

# Skip timestamp restoration (use current time)
unzip -D archive.zip

# Quiet extraction
unzip -q archive.zip

# Very quiet (errors only)
unzip -qq archive.zip

# Extract password-protected archive (WARNING: insecure!)
unzip -P mypassword encrypted.zip

Note on Passwords: The -P option exposes your password in the process list and command history. This is insecure and should only be used in scripts with controlled access. For interactive use, consider using environment variables or secure password managers.

Glob Pattern Syntax

Pattern	Description
*	Matches any characters except /
**	Matches any characters including / (recursive)
?	Matches any single character

Pattern Examples

• *.txt - All .txt files in root
• src/*.rs - All .rs files directly in src/
• **/*.rs - All .rs files in any directory
• src/** - Everything under src/
• file?.txt - file1.txt, fileA.txt, etc.

Performance

This implementation is optimized for speed:

• Memory-mapped I/O: Files >1MB use mmap for faster reading
• Buffered writing: 256KB write buffers for efficient disk I/O
• LTO builds: Link-time optimization in release builds
• Minimal allocations: Reuses buffers where possible
• Linux kernel optimizations (on Linux):
  • madvise(MADV_SEQUENTIAL) - Hint for sequential access patterns
  • madvise(MADV_WILLNEED) - Pre-fault pages for faster access
  • fallocate() - Pre-allocate disk space to avoid fragmentation
  • fadvise(POSIX_FADV_SEQUENTIAL) - Hint for file access patterns

Benchmark Results

Tested on Linux 6.18 with a 1GB ZIP archive containing 100 files (10MB each, stored/uncompressed):

Tool	Time	Speedup
Info-ZIP unzip 6.00	4.92s	1x (baseline)
This (Rust unzip)	0.95s	~5x faster

# Reproduce the benchmark
$ time /usr/bin/unzip -q -o -d out_system test.zip
4.92s

$ time ./target/release/unzip -q -o -d out_rust test.zip
0.95s

Additional metadata benchmarks (5-run average) on a 44MiB deflated ZIP (1000 x 4KB + 20 x 1MB + 2 x 10MB, binary):

Operation	Info-ZIP unzip	Rust unzip
-l list	0.0010s	0.0008s
-Z zipinfo	0.0015s	0.0010s
-p pipe (single 4KB file)	0.0004s	0.0015s

Notes:

• Listing/zipinfo use metadata-only reads (no decompressor setup).
• Pipe is still slower on tiny binary outputs; it is faster on compressible text.

Comparison with Info-ZIP

Feature	Info-ZIP	This
Memory safety	C (manual)	Rust (guaranteed)
Progress bar	No	Yes
Glob patterns	Basic	Extended (**)
Memory-mapped I/O	No	Yes (>1MB files)
Last updated	2009	Active
Performance	Baseline	~5x faster

Compatibility

This tool aims for CLI compatibility with Info-ZIP unzip. Most common options work identically:

# These work the same way
unzip -l archive.zip
unzip -d /tmp archive.zip
unzip -o archive.zip
unzip archive.zip '*.txt' -x '*.log'

Dependencies

• zip - ZIP archive handling
• clap - CLI argument parsing
• anyhow - Error handling
• indicatif - Progress bars
• memmap2 - Memory-mapped files
• filetime - File timestamp handling
• crc32fast - Fast CRC verification
• rustix - Linux syscalls for kernel optimizations (Linux only)

Contributing

Contributions are welcome! See CONTRIBUTING.md for guidelines.

For development information including architecture, project structure, and coding standards, see CLAUDE.md.

License

MIT