HZip

The goal of this project is to implement a lossless data compression algorithm in Haskell.

LZ77

The standard LZ77 implementation uses two bytes for each pair. The first byte is either a special flag which indicates the following byte is a literal or a <distance, length> pair which tells us how to decode the string.

To compile

./build.sh

To encode

./lz77.hs -e <filepath>

This will produce a .hzip file with the same name as the original file.

To decode

./lz77.hs -d <filepath>

LZSS

The LZSS algorithm is very similar to the LZ77 implementation but further reduces file size at the cost of compression times. Rather than using a reserved byte to indicate a literal, a single bit is used where 1 is a literal and 0 is a <distance, length> pair.

To compile

./build.sh

To encode

./lzss.hs -e <filepath>

This will produce a .hzip file with the same name as the original file.

To decode

./lzss.hs -d <filepath>

Dependencies

Provides several functions to quickly search for substrings in strict or lazy ByteStrings.

cabal install stringsearch