This repo is an implementation of both constant and variable huffman encoding.

make

Run the following for constant length huffman encoding

./huffman_c

Run the following for variable length huffman encoding

./huffman_v

Sample output

input.txt: This is a large text file with random lines of text
comp.hfv: This is the variable length huffman encoding of input.txt
comp.hfc: This is the constant length huffman encoding of input.txt
out_c.txt: This is the uncompressed file generated from comp.hfc
out_v.txt: This is the uncompressed file generated from comp.hfv

It can be observed this is a lossless compression algorithm. The file size of the compressed files can be reduced by using a more space efficient way to store the encoding table/tree.

The encoding table is separated from the encoded text by the following character: #. Every encoding table has one extra character other than those found in the text: \0. This is to ensure that while decoding, there is a known ending to the file. Since \0 is not a printable character, it can never be inputed as a character.