|
1
|
|
|
# Copyright 2019-2020 by Christopher C. Little. |
|
2
|
|
|
# This file is part of Abydos. |
|
3
|
|
|
# |
|
4
|
|
|
# Abydos is free software: you can redistribute it and/or modify |
|
5
|
|
|
# it under the terms of the GNU General Public License as published by |
|
6
|
|
|
# the Free Software Foundation, either version 3 of the License, or |
|
7
|
|
|
# (at your option) any later version. |
|
8
|
|
|
# |
|
9
|
|
|
# Abydos is distributed in the hope that it will be useful, |
|
10
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
11
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
12
|
|
|
# GNU General Public License for more details. |
|
13
|
|
|
# |
|
14
|
|
|
# You should have received a copy of the GNU General Public License |
|
15
|
|
|
# along with Abydos. If not, see <http://www.gnu.org/licenses/>. |
|
16
|
|
|
|
|
17
|
|
|
"""abydos.distance._ncd_paq9a. |
|
18
|
|
|
|
|
19
|
1 |
|
NCD using PAQ9A |
|
20
|
|
|
""" |
|
21
|
|
|
|
|
22
|
|
|
from ._distance import _Distance |
|
23
|
|
|
|
|
24
|
1 |
|
try: |
|
25
|
|
|
import paq |
|
26
|
|
|
except ImportError: # pragma: no cover |
|
27
|
|
|
# If the system lacks the paq9a library, that's fine, but PAQ9A compression |
|
28
|
|
|
# similarity won't be supported. |
|
29
|
|
|
paq = None # type: ignore |
|
30
|
|
|
|
|
31
|
1 |
|
__all__ = ['NCDpaq9a'] |
|
32
|
|
|
|
|
33
|
1 |
|
|
|
34
|
1 |
|
class NCDpaq9a(_Distance): |
|
35
|
|
|
"""Normalized Compression Distance using PAQ9A compression. |
|
36
|
|
|
|
|
37
|
|
|
Cf. http://mattmahoney.net/dc/#paq9a |
|
38
|
|
|
|
|
39
|
|
|
Normalized compression distance (NCD) :cite:`Cilibrasi:2005`. |
|
40
|
1 |
|
|
|
41
|
|
|
.. versionadded:: 0.4.0 |
|
42
|
|
|
""" |
|
43
|
1 |
|
|
|
44
|
|
|
def dist(self, src: str, tar: str) -> float: |
|
45
|
|
|
"""Return the NCD between two strings using PAQ9A compression. |
|
46
|
|
|
|
|
47
|
|
|
Parameters |
|
48
|
|
|
---------- |
|
49
|
|
|
src : str |
|
50
|
|
|
Source string for comparison |
|
51
|
|
|
tar : str |
|
52
|
|
|
Target string for comparison |
|
53
|
1 |
|
|
|
54
|
|
|
Returns |
|
55
|
|
|
------- |
|
56
|
|
|
float |
|
57
|
|
|
Compression distance |
|
58
|
|
|
|
|
59
|
|
|
Raises |
|
60
|
|
|
------ |
|
61
|
|
|
ValueError |
|
62
|
|
|
Install the paq module in order to use PAQ9A |
|
63
|
|
|
|
|
64
|
|
|
Examples |
|
65
|
|
|
-------- |
|
66
|
|
|
>>> cmp = NCDpaq9a() |
|
67
|
|
|
>>> cmp.dist('cat', 'hat') |
|
68
|
|
|
0.42857142857142855 |
|
69
|
|
|
>>> cmp.dist('Niall', 'Neil') |
|
70
|
|
|
0.5555555555555556 |
|
71
|
|
|
>>> cmp.dist('aluminum', 'Catalan') |
|
72
|
|
|
0.5833333333333334 |
|
73
|
|
|
>>> cmp.dist('ATCG', 'TAGC') |
|
74
|
|
|
0.5 |
|
75
|
|
|
|
|
76
|
|
|
|
|
77
|
|
|
.. versionadded:: 0.4.0 |
|
78
|
|
|
|
|
79
|
|
|
""" |
|
80
|
|
|
if src == tar: |
|
81
|
|
|
return 0.0 |
|
82
|
|
|
|
|
83
|
|
|
src_b = src.encode('utf-8') |
|
84
|
|
|
tar_b = tar.encode('utf-8') |
|
85
|
|
|
|
|
86
|
|
|
if paq is not None: |
|
87
|
|
|
src_comp = paq.compress(src_b) |
|
88
|
|
|
tar_comp = paq.compress(tar_b) |
|
89
|
|
|
concat_comp = paq.compress(src_b + tar_b) |
|
90
|
|
|
concat_comp2 = paq.compress(tar_b + src_b) |
|
91
|
|
|
else: # pragma: no cover |
|
92
|
|
|
raise ValueError('Install the paq module in order to use PAQ9A') |
|
93
|
|
|
|
|
94
|
|
|
# Each string returned by PAQ9A's compressor has 4 header bytes |
|
95
|
|
|
# followed by a byte of information then 3 null bytes. And it is |
|
96
|
|
|
# concluded with 3 bytes of \xff. So 4+3+3 invariant bytes are |
|
97
|
|
|
# subtracted here. |
|
98
|
|
|
return ( |
|
99
|
|
|
(min(len(concat_comp), len(concat_comp2)) - 10) |
|
100
|
|
|
- (min(len(src_comp), len(tar_comp)) - 10) |
|
101
|
|
|
) / (max(len(src_comp), len(tar_comp)) - 10) |
|
102
|
|
|
|
|
103
|
|
|
|
|
104
|
|
|
if __name__ == '__main__': |
|
105
|
|
|
import doctest |
|
106
|
|
|
|
|
107
|
|
|
doctest.testmod() |
|
108
|
|
|
|
chrislit /
abydos
