|
1
|
|
|
# -*- coding: utf-8 -*- |
|
2
|
|
|
|
|
3
|
|
|
# Copyright 2014-2018 by Christopher C. Little. |
|
4
|
|
|
# This file is part of Abydos. |
|
5
|
|
|
# |
|
6
|
|
|
# Abydos is free software: you can redistribute it and/or modify |
|
7
|
|
|
# it under the terms of the GNU General Public License as published by |
|
8
|
|
|
# the Free Software Foundation, either version 3 of the License, or |
|
9
|
|
|
# (at your option) any later version. |
|
10
|
|
|
# |
|
11
|
|
|
# Abydos is distributed in the hope that it will be useful, |
|
12
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
13
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
14
|
|
|
# GNU General Public License for more details. |
|
15
|
|
|
# |
|
16
|
|
|
# You should have received a copy of the GNU General Public License |
|
17
|
|
|
# along with Abydos. If not, see <http://www.gnu.org/licenses/>. |
|
18
|
|
|
|
|
19
|
1 |
|
"""abydos.distance._ncd_bz2. |
|
20
|
|
|
|
|
21
|
|
|
NCD using bzip2 |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
1 |
|
from __future__ import ( |
|
25
|
|
|
absolute_import, |
|
26
|
|
|
division, |
|
27
|
|
|
print_function, |
|
28
|
|
|
unicode_literals, |
|
29
|
|
|
) |
|
30
|
|
|
|
|
31
|
1 |
|
import bz2 |
|
32
|
|
|
|
|
33
|
1 |
|
from ._distance import _Distance |
|
34
|
|
|
|
|
35
|
1 |
|
__all__ = ['NCDbz2', 'dist_ncd_bz2', 'sim_ncd_bz2'] |
|
36
|
|
|
|
|
37
|
|
|
|
|
38
|
1 |
|
class NCDbz2(_Distance): |
|
|
|
|
|
|
39
|
|
|
"""Normalized Compression Distance using bzip2 compression. |
|
40
|
|
|
|
|
41
|
|
|
Cf. https://en.wikipedia.org/wiki/Bzip2 |
|
42
|
|
|
|
|
43
|
|
|
Normalized compression distance (NCD) :cite:`Cilibrasi:2005`. |
|
44
|
|
|
""" |
|
45
|
|
|
|
|
46
|
1 |
|
_level = 9 |
|
47
|
|
|
|
|
48
|
1 |
|
def __init__(self, level=9): |
|
49
|
|
|
"""Initialize bzip2 compressor. |
|
50
|
|
|
|
|
51
|
|
|
Parameters |
|
52
|
|
|
---------- |
|
53
|
|
|
level : int |
|
54
|
|
|
The compression level (0 to 9) |
|
55
|
|
|
|
|
56
|
|
|
""" |
|
57
|
1 |
|
self._level = level |
|
58
|
|
|
|
|
59
|
1 |
|
def dist(self, src, tar): |
|
|
|
|
|
|
60
|
|
|
"""Return the NCD between two strings using bzip2 compression. |
|
61
|
|
|
|
|
62
|
|
|
Parameters |
|
63
|
|
|
---------- |
|
64
|
|
|
src : str |
|
65
|
|
|
Source string for comparison |
|
66
|
|
|
tar : str |
|
67
|
|
|
Target string for comparison |
|
68
|
|
|
|
|
69
|
|
|
Returns |
|
70
|
|
|
------- |
|
71
|
|
|
float |
|
72
|
|
|
Compression distance |
|
73
|
|
|
|
|
74
|
|
|
Examples |
|
75
|
|
|
-------- |
|
76
|
|
|
>>> cmp = NCDbz2() |
|
77
|
|
|
>>> cmp.dist('cat', 'hat') |
|
78
|
|
|
0.06666666666666667 |
|
79
|
|
|
>>> cmp.dist('Niall', 'Neil') |
|
80
|
|
|
0.03125 |
|
81
|
|
|
>>> cmp.dist('aluminum', 'Catalan') |
|
82
|
|
|
0.17647058823529413 |
|
83
|
|
|
>>> cmp.dist('ATCG', 'TAGC') |
|
84
|
|
|
0.03125 |
|
85
|
|
|
|
|
86
|
|
|
""" |
|
87
|
1 |
|
if src == tar: |
|
88
|
1 |
|
return 0.0 |
|
89
|
|
|
|
|
90
|
1 |
|
src = src.encode('utf-8') |
|
91
|
1 |
|
tar = tar.encode('utf-8') |
|
92
|
|
|
|
|
93
|
1 |
|
src_comp = bz2.compress(src, self._level)[10:] |
|
94
|
1 |
|
tar_comp = bz2.compress(tar, self._level)[10:] |
|
95
|
1 |
|
concat_comp = bz2.compress(src + tar, self._level)[10:] |
|
96
|
1 |
|
concat_comp2 = bz2.compress(tar + src, self._level)[10:] |
|
97
|
|
|
|
|
98
|
1 |
|
return ( |
|
99
|
|
|
min(len(concat_comp), len(concat_comp2)) |
|
100
|
|
|
- min(len(src_comp), len(tar_comp)) |
|
101
|
|
|
) / max(len(src_comp), len(tar_comp)) |
|
102
|
|
|
|
|
103
|
|
|
|
|
104
|
1 |
|
def dist_ncd_bz2(src, tar): |
|
105
|
|
|
"""Return the NCD between two strings using bzip2 compression. |
|
106
|
|
|
|
|
107
|
|
|
This is a wrapper for :py:meth:`NCDbz2.dist`. |
|
108
|
|
|
|
|
109
|
|
|
Parameters |
|
110
|
|
|
---------- |
|
111
|
|
|
src : str |
|
112
|
|
|
Source string for comparison |
|
113
|
|
|
tar : str |
|
114
|
|
|
Target string for comparison |
|
115
|
|
|
|
|
116
|
|
|
Returns |
|
117
|
|
|
------- |
|
118
|
|
|
float |
|
119
|
|
|
Compression distance |
|
120
|
|
|
|
|
121
|
|
|
Examples |
|
122
|
|
|
-------- |
|
123
|
|
|
>>> dist_ncd_bz2('cat', 'hat') |
|
124
|
|
|
0.06666666666666667 |
|
125
|
|
|
>>> dist_ncd_bz2('Niall', 'Neil') |
|
126
|
|
|
0.03125 |
|
127
|
|
|
>>> dist_ncd_bz2('aluminum', 'Catalan') |
|
128
|
|
|
0.17647058823529413 |
|
129
|
|
|
>>> dist_ncd_bz2('ATCG', 'TAGC') |
|
130
|
|
|
0.03125 |
|
131
|
|
|
|
|
132
|
|
|
""" |
|
133
|
1 |
|
return NCDbz2().dist(src, tar) |
|
134
|
|
|
|
|
135
|
|
|
|
|
136
|
1 |
|
def sim_ncd_bz2(src, tar): |
|
137
|
|
|
"""Return the NCD similarity between two strings using bzip2 compression. |
|
138
|
|
|
|
|
139
|
|
|
This is a wrapper for :py:meth:`NCDbz2.sim`. |
|
140
|
|
|
|
|
141
|
|
|
Parameters |
|
142
|
|
|
---------- |
|
143
|
|
|
src : str |
|
144
|
|
|
Source string for comparison |
|
145
|
|
|
tar : str |
|
146
|
|
|
Target string for comparison |
|
147
|
|
|
|
|
148
|
|
|
Returns |
|
149
|
|
|
------- |
|
150
|
|
|
float |
|
151
|
|
|
Compression similarity |
|
152
|
|
|
|
|
153
|
|
|
Examples |
|
154
|
|
|
-------- |
|
155
|
|
|
>>> sim_ncd_bz2('cat', 'hat') |
|
156
|
|
|
0.9333333333333333 |
|
157
|
|
|
>>> sim_ncd_bz2('Niall', 'Neil') |
|
158
|
|
|
0.96875 |
|
159
|
|
|
>>> sim_ncd_bz2('aluminum', 'Catalan') |
|
160
|
|
|
0.8235294117647058 |
|
161
|
|
|
>>> sim_ncd_bz2('ATCG', 'TAGC') |
|
162
|
|
|
0.96875 |
|
163
|
|
|
|
|
164
|
|
|
""" |
|
165
|
1 |
|
return NCDbz2().sim(src, tar) |
|
166
|
|
|
|
|
167
|
|
|
|
|
168
|
|
|
if __name__ == '__main__': |
|
169
|
|
|
import doctest |
|
170
|
|
|
|
|
171
|
|
|
doctest.testmod() |
|
172
|
|
|
|
chrislit /
abydos
Loading history...