The Lempel–Ziv–Welch (LZW) is a data compression algorithm. I did my own version of the LZW compressor and decompressor. I used 12-bit fixed with codes.

There are three examples of my code running:

• file1.txt <-> file1.z
• file2.txt <-> file2.z
• file3.txt <-> file3.z

If you use g++ to execute your program, you should go to the folder where is the compressor.cpp file. Then, you should write the line g++ compressor.cpp –o compressor -std=c++11. Then, you should write compressor name_of_file_you_want_to_compress. For instance, compressor /LZWInputData/file1.txt. If you use a program like Eclipe, Notepad, … you should add the name of the file to compress as a parameter in the running options. In order to compress a file named file1.txt you should execute the program with the full filename path as an argument.

There are :

1. Private attributes:

• a dictionary (map)
• a position (integer - indicates the position in the file)
• an OldCode (char)

2. Public attributes:

• a constructor
• a method compressedToStream, which does the compression
• a method initializeDictionary, which creates the dictionary at the beginning of the program

If you use g++ to execute your program, you should just like you did before with the compressor but instead of writting compressor, you should write decompressor.

There are :

1. Private attributes:

• a dictionary (map)
• a position (integer - the position in the file to decompress, that is to say the number of codes we have already read)
• an OldByte (unsigned char - it memorizes the last byte we read in the file)

2. Public attributes:

• A constructor which uses the ifstream constructor
• readCode : read the next code in the file to decompress
• DecompressToStream: do the decompression, using the readCode method
• initializeDictionary: initialize (or reinitialize) the dictionary when needed

• the code is written for a 12-bit fixed with codes. I would like to improve it and make it possible to use with any number of bits.