|
| 1 | +""" |
| 2 | +Move to Front (MTF) is a data compression technique that reorders |
| 3 | +symbols based on their frequency of occurrence. |
| 4 | + - type of context based methods |
| 5 | + - it lays somewhere between coding and transforming data |
| 6 | + - it is core of the Burrows-Wheeler Transform (BWT) |
| 7 | + - for simplicity we do not distinguish between upper and lower case |
| 8 | +
|
| 9 | +
|
| 10 | +Sources: |
| 11 | + - https://en.wikipedia.org/wiki/Move-to-front_transform |
| 12 | + - https://www.mbit.edu.in/wp-content/uploads/2020/05/data_compression.pdf |
| 13 | + (chapter 6.4.1) |
| 14 | +""" |
| 15 | + |
| 16 | +english_alphabet = [ |
| 17 | + "a", |
| 18 | + "b", |
| 19 | + "c", |
| 20 | + "d", |
| 21 | + "e", |
| 22 | + "f", |
| 23 | + "g", |
| 24 | + "h", |
| 25 | + "i", |
| 26 | + "j", |
| 27 | + "k", |
| 28 | + "l", |
| 29 | + "m", |
| 30 | + "n", |
| 31 | + "o", |
| 32 | + "p", |
| 33 | + "q", |
| 34 | + "r", |
| 35 | + "s", |
| 36 | + "t", |
| 37 | + "u", |
| 38 | + "v", |
| 39 | + "w", |
| 40 | + "x", |
| 41 | + "y", |
| 42 | + "z", |
| 43 | +] |
| 44 | + |
| 45 | + |
| 46 | +class MoveToFront: |
| 47 | + """ |
| 48 | + Core class doing encoding and also decoding. |
| 49 | +
|
| 50 | + >>> mtf = MoveToFront() |
| 51 | + >>> list(mtf.encode_text("algorithm")) |
| 52 | + [0, 11, 7, 14, 17, 11, 19, 12, 15] |
| 53 | + >>> mtf.decode_text([0, 11, 7, 14, 17, 11, 19, 12, 15]) |
| 54 | + 'algorithm' |
| 55 | + """ |
| 56 | + |
| 57 | + def __init__(self, source_alphabet: list[str] = english_alphabet) -> None: |
| 58 | + """ |
| 59 | + If the source alphabet is not provided, it will be set to the english. |
| 60 | + The class does not distinguish between UPPER and lower case. |
| 61 | + """ |
| 62 | + self.source_alphabet = list(source_alphabet) |
| 63 | + |
| 64 | + def encode_text(self, plain_text: str) -> list[int]: |
| 65 | + """ |
| 66 | + Encodes given text. The output is list of char positions in the alphabet. |
| 67 | +
|
| 68 | + >>> mtf = MoveToFront() |
| 69 | + >>> list(mtf.encode_text("algorithm")) |
| 70 | + [0, 11, 7, 14, 17, 11, 19, 12, 15] |
| 71 | + """ |
| 72 | + |
| 73 | + if not isinstance(plain_text, str): |
| 74 | + raise TypeError("The parameter plain_text type must be str.") |
| 75 | + |
| 76 | + # making a copy so we do not rotate the original alphabet |
| 77 | + alphabet = list(self.source_alphabet) |
| 78 | + encoded_text = [] |
| 79 | + |
| 80 | + for char in plain_text.lower(): |
| 81 | + # find the position of the char in the alphabet and add it to the result |
| 82 | + char_position_in_alphabet = alphabet.index(char) |
| 83 | + encoded_text.append(char_position_in_alphabet) |
| 84 | + |
| 85 | + # move our char to the front of the alphabet |
| 86 | + alphabet.pop(char_position_in_alphabet) |
| 87 | + alphabet.insert(0, char) |
| 88 | + |
| 89 | + return encoded_text |
| 90 | + |
| 91 | + def decode_text(self, compressed_text: list[int]) -> str: |
| 92 | + """ |
| 93 | + Decodes given text. The input is list of char positions in the alphabet. |
| 94 | +
|
| 95 | + >>> mtf = MoveToFront() |
| 96 | + >>> mtf.decode_text([0, 11, 7, 14, 17, 11, 19, 12, 15]) |
| 97 | + 'algorithm' |
| 98 | + """ |
| 99 | + |
| 100 | + alphabet = list(self.source_alphabet) |
| 101 | + decoded_text = [] |
| 102 | + |
| 103 | + for idx in compressed_text: |
| 104 | + # find corresponding chart to given index |
| 105 | + char = alphabet[idx] |
| 106 | + decoded_text.append(char) |
| 107 | + |
| 108 | + # move found char to the front of the alphabet |
| 109 | + alphabet.pop(idx) |
| 110 | + alphabet.insert(0, char) |
| 111 | + |
| 112 | + return "".join(decoded_text) |
| 113 | + |
| 114 | + |
| 115 | +if __name__ == "__main__": |
| 116 | + mtf = MoveToFront() |
| 117 | + print(mtf.encode_text("algorithm")) |
| 118 | + print(mtf.decode_text([0, 11, 7, 14, 17, 11, 19, 12, 15])) |
0 commit comments