README.md

Implementation of archivator using Huffman coding

How to use

compress("compress-me.txt", "compressed.bin")

decompress("compressed.bin", "decompressed.txt")

Result

encoded huffman tree code: 0010010000001011001010001011100001001011000101100011...
encoded text code: 000000001111011011010010001111010011101101111110111001110100...
before: 7528bytes, after: 4109bytes, compression 45.4%

compression can be negative if the text is too short, as the binary file with encoded text also contains a Huffman tree data, which needed to decode the text.

average result is 45%, depends on frequency can be up to 90%

!not working with cyrillic symbols, it's a simple archivator made with learning purposes

how it's implemented

compression:

1. read text from the input file
2. create frequencies table using collections.Counter
3. create Huffman nodes using queue.PriorityQueue
4. merge the nodes to get Huffman Tree
5. create code table (dictionary) with chars and corresponding codes
6. encode the text using the code table
7. encode Huffman tree (each node is 0, each leaf is 1 + encoded ascii char)
8. place the encoded tree before encoded text
9. save it to a new binary file

decompression

1. read encoded data from binary file
2. decode a Huffman tree
3. using the tree decode the text
4. save it to output file