Zero-Inversion-Compression (zic) is a non-lossless file compression algorithm with compression rate up to 100% of the original file size.
cargo build
cargo run README.md
> Finished dev [unoptimized + debuginfo] target(s) in 0.00s
> Running `target/debug/zero-inversion-compression README.md`
Compressed file 'README.md' from 1390 bytes to 'README.md.zic' (0 bytes)
(100% compression rate)
zic reduces the file size by discarding any bits of a file that do not contain valuable information, thus reducing the hamming weight of a file.
Files are represented in binary form as a series of 0 and 1. As 0 has no value, these bits can be discarded, leaving only the valuable bits in a file. While this alone might yield good compression results, further reduction of the size is achieved by inverting the compressed data and running the algorithm twice.
- Running the two compression runs in parallel
- Evening out the hamming weight of the file before running the algorithm
- Compress the two loops into one by inverting the bits in place
- Apply the NGGYU-Algorithm for sanitizing the input.
Feel free to fork and put up a pull request for any improvements. A good starting point for various improvements are the @todos in the code.