• Three basic types of data compression.
  • encodeFile needs to call encodeString
  • decodeFile needs to call decodeString
import os

import json

# Encodes as a list of (char, count) tuples
    def encodeString(stringVal):
        encodedList = []
        prevChar = None
        count = 0
        for char in stringVal:
            if prevChar != char and prevChar is not None:
                encodedList.append((prevChar, count))
                count = 0
            prevChar = char
            count = count + 1
        encodedList.append((prevChar, count))
        return encodedList
    
    def decodeString(encodedList):
        decodedStr = ''
        for item in encodedList:
            try:
                decodedStr = decodedStr + item[0] * item[1]
            except:
                print(item)
        return decodedStr

def encodeFile(filename, newFilename):
        with open(filename) as f:
            data = encodeString(f.read())
    		# We grab the JSON blob from above and is then written to a file like below
        with open(newFilename, 'w') as f:
            f.write(json.dumps(data))
    
    
    def decodeFile(filename):
        with open(filename) as f:
            data = f.read()
        return decodeString(json.loads(data))
		
print(f'Original file size: {os.path.getsize(\10_04_challenge_art.txt\)}'),

# This takes in a new file name
encodeFile('10_04_challenge_art.txt', '10_04_challenge_art_encoded.txt')

print(f'New file size: {os.path.getsize(\10_04_challenge_art_encoded.txt\)}')
    print(decodeFile('10_04_challenge_art_encoded.txt'))
  • Running the above will then output: 
    Original file size: 2757
New file size: 2441
Outputs the ASCII art below.
  • Keep in mind the characters that are in a JSON file, commas, apostrophes, brackets etc all take up space.
  • To further improve the compression, we can do: 
    # [('A', 1), ('B', 80), ('C', 10)]
# This outputs | ~ | ~ in that pattern instead
  # becomes A|1~B|80~C|10
  def encodeFile(filename, newFilename):
      with open(filename) as f:
          data = encodeString(f.read())
    
      data = [f'{char}|{count}' for char, count in data]
        
      with open(newFilename, 'w') as f:
          f.write('~'.join(data))
    
  # The decoder then splits the data and puts it into pairs
  def decodeFile(filename):
      with open(filename) as f:
          data = f.read()
            
      pairs = data.split('~')
      pairs = [p.split('|') for p in pairs]
               # This is the stage where it is placed into pairs
      pairs = [(p[0], int(p[1])) for p in pairs]
      return decodeString(pairs)
  • Running that, we then get: 
    Original file size: 2757
New file size: 1007
Outputs the ASCII art below.
  • Even more improved solution than the last.
  • You can store any integer up to 255 in a single byte (or single character’s worth of data) of data. ``` def encodeFile(filename, newFilename): with open(filename) as f: data = encodeString(f.read()) output = bytearray() for item in data: # Character byte output.extend(bytes(item[0], ‘utf-8’)) # Integer count byte output.extend(item[1].to_bytes(1, ‘big’)) with open(newFilename, ‘wb’) as binary_file: # Write bytes to file binary_file.write(output)

def decodeFile(filename): with open(filename, ‘rb’) as f: data = f.read() # Split data into pairs bytePairs = [data[i:i+2] for i in range(0, len(data), 2)] encodedList = [] for bytePair in bytePairs: encodedList.append((bytePair[:1].decode(‘utf-8’), int.from_bytes(bytePair[1:], ‘big’))) return decodeString(encodedList)

print(f’Original file size: {os.path.getsize(\10_04_challenge_art.txt)}’), encodeFile(‘10_04_challenge_art.txt’, ‘10_04_challenge_art_encoded.aa’)

print(f’New file size: {os.path.getsize(\10_04_challenge_art_encoded.aa)}’) print(decodeFile(‘10_04_challenge_art_encoded.aa’)) ```

  • The above outputs the file at 466 bytes.

Updated: