A program to compress files using Huffman Coding.
This program takes a training file to learn the frequency of different characters in the language of your text files.
It uses the frequency of characters in your training file to create a huffman tree which acts as a cipher when encoding or decoding your text files.
The example files given show how the test_source.txt can be compressed with the program from 440KB to 246KB.
The training_file.txt given is a free Frankenstein ebook from http://www.gutenberg.org. The frequency of characters in this text file will make a huffman tree that will compress any English text file quite efficiently.
To encode a file:
./compressor huffcode <training filename> <input filename> <output filename>
To decode a file:
./compressor huffdecode <training filename> <input filename> <output filename>