https://github.com/amicks/huffman-coding

Utilizes Huffman's lossless data compression algorithm to encode/decode files. Final project in CMPS12B.
https://github.com/amicks/huffman-coding

Last synced: about 1 month ago
JSON representation

Utilizes Huffman's lossless data compression algorithm to encode/decode files. Final project in CMPS12B.

• Host: GitHub
• URL: https://github.com/amicks/huffman-coding
• Owner: amicks
• Created: 2017-06-12T20:55:33.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-09-04T00:07:23.000Z (about 7 years ago)
• Last Synced: 2024-10-09T09:50:15.831Z (about 1 month ago)
• Language: C
• Homepage:
• Size: 713 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
###### Author: Allston Mickey

## Files:

* bv.h

* code.h

* decode.c

* encode.c

* heap.h

* huffman.c, huffman.h

* queue.c, queue.h

* stack.c, stack.h

* Makefile

## How to compile & run:

Command | Description

--- | ---

make | Compiles the encode and decode programs. 

make all | Compiles the encode and decode programs. 

make encode | Compiles the encode program.

make decode | Compiles the decode program.

make clean | Remove the object files and encode/decode binaries

./encode _[flags]_ | Run the encode program.

./decode _[flags]_ | Run the decode program.

## Flags:

Argument(s) | Description | Coding Type

--- | --- | ---

-i _[path]_ | Full path to the input file | Both **_[REQUIRED]_**

-o _[path]_ | Full path to the output file | Both

-A | Sets all flags (-vp, -ch if available) | Both

-v | Enable verbose mode - prints statistics about the program | Both

-p | Prints the Huffman Tree | Both

-c | Prints the codes/bit paths to each leaf in the Huffman Tree | Encoding

-h | (./encode only) Prints the histogram of bytes | Encoding

## Description:

These programs encode/decode any file by using the greedy Huffman Algorithm.

#### [encoding]

1. Construct a histogram of the bytes in a file.

2. Enqueue each histogram entry as a node in the Huffman Tree.

3. Repeatedly dequeue 2X and enqueue until the root of the tree is remaining.

4. Load the codes for each leaf by a post-order traversal of the tree, assigning a 0 when traversing left and a 1 when traversing right.  Store the codes in a bit vector for maximum efficiency.

5. Write the tree and the paths to each leaf to an encoded output file.

#### [decode]

1. Read the encoded file.

2. Reconstruct the Huffman Tree with a stack.

3. For each byte in the original, uncompressed file, look up its code and step through the tree.

4. Write the bytes back to a decoded output file.

---

* Using both programs, any file can be encoded and then decoded back to an exact copy of the original.

* Note that the default permissions to a file are 0744 (-rwxr--r--).

  You may change these on your own if they do not suit what you intended.