COL106: Assignment 4
Due Date : October 13, 2014
In this assignment, you will learn about the famous Lempel-Ziv-Welch (LZW)
compression algorithm. Most compression algorithms including UNIX compress
are based on the LZW algorithm. The LZW algorithm compresses a given text
by figuring out patterns which appear very often in the text. The basic
idea is replace (perhaps very long) patterns which appear many times by a
much shorter code.
A good description of the algorithm can be found here
a working example of how to compress and decompress strings.
Your assignment is to implement the LZW algorithm. You can assume that the
input text contains english characters (capital and small), space, and
full stop. To implement the dictionary of code and key, you will use a
hash table. Use your own hash function, but you must use quadratic probing
to resolve collisions.
- You must not re-use the dictionary from the compression phase during
the decompression. The dictionary must be re-built.
- Initially (during both encoding and decoding), the dictionary contains
the 256 possible ASCII characters. These are given codes 0-255 (ASCII
encoding -- this means 'A' is given the code 65, and 'a' the code 97). You
can start with an initial hash table size of 275 entries (275-256=19
entries are initially unused).
- Use your own hash function.
- Collisions must be resolved using quadratic probing.
- The hash table must be growable. You may encounter scenarios where the
hash table fills up -- these must be handled by suitably increasing the
size of the hash table.
- You may use any of the structures in the Standard Template Library,
other than the hash table.
Two programs -- a compressor and a decompressor -- written in C/C++ is
- Given an input file as command line argument, the compressor must
create a compressed file in the same directory.
- Given a compressed file as a command line argument, the decompressor
must create the uncompressed file in the same directory.
Input String = BABAABAAA
Compressed String = <66><65><256><257><65><260>