|
1
|
|
|
# -*- coding: utf-8 -*- |
|
2
|
|
|
|
|
3
|
|
|
# Copyright 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._sift4_simplest. |
|
20
|
|
|
|
|
21
|
|
|
Sift4 Simplest approximate string distance |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
1 |
|
from __future__ import ( |
|
25
|
|
|
absolute_import, |
|
26
|
|
|
division, |
|
27
|
|
|
print_function, |
|
28
|
|
|
unicode_literals, |
|
29
|
|
|
) |
|
30
|
|
|
|
|
31
|
1 |
|
from six.moves import range |
|
32
|
|
|
|
|
33
|
1 |
|
from ._sift4 import Sift4 |
|
34
|
|
|
|
|
35
|
1 |
|
__all__ = ['Sift4Simplest', 'sift4_simplest'] |
|
36
|
|
|
|
|
37
|
|
|
|
|
38
|
1 |
|
class Sift4Simplest(Sift4): |
|
|
|
|
|
|
39
|
|
|
"""Sift4 Simplest version. |
|
40
|
|
|
|
|
41
|
|
|
This is an approximation of edit distance, described in |
|
42
|
|
|
:cite:`Zackwehdex:2014`. |
|
43
|
|
|
""" |
|
44
|
|
|
|
|
45
|
1 |
|
def dist_abs(self, src, tar, max_offset=5): |
|
|
|
|
|
|
46
|
|
|
"""Return the "simplest" Sift4 distance between two terms. |
|
47
|
|
|
|
|
48
|
|
|
Parameters |
|
49
|
|
|
---------- |
|
50
|
|
|
src : str |
|
51
|
|
|
Source string for comparison |
|
52
|
|
|
tar : str |
|
53
|
|
|
Target string for comparison |
|
54
|
|
|
max_offset : int |
|
55
|
|
|
The number of characters to search for matching letters |
|
56
|
|
|
|
|
57
|
|
|
Returns |
|
58
|
|
|
------- |
|
59
|
|
|
int |
|
60
|
|
|
The Sift4 distance according to the simplest formula |
|
61
|
|
|
|
|
62
|
|
|
Examples |
|
63
|
|
|
-------- |
|
64
|
|
|
>>> cmp = Sift4Simplest() |
|
65
|
|
|
>>> cmp.dist_abs('cat', 'hat') |
|
66
|
|
|
1 |
|
67
|
|
|
>>> cmp.dist_abs('Niall', 'Neil') |
|
68
|
|
|
2 |
|
69
|
|
|
>>> cmp.dist_abs('Colin', 'Cuilen') |
|
70
|
|
|
3 |
|
71
|
|
|
>>> cmp.dist_abs('ATCG', 'TAGC') |
|
72
|
|
|
2 |
|
73
|
|
|
|
|
74
|
|
|
""" |
|
75
|
1 |
|
if not src: |
|
76
|
1 |
|
return len(tar) |
|
77
|
|
|
|
|
78
|
1 |
|
if not tar: |
|
79
|
1 |
|
return len(src) |
|
80
|
|
|
|
|
81
|
1 |
|
src_len = len(src) |
|
82
|
1 |
|
tar_len = len(tar) |
|
83
|
|
|
|
|
84
|
1 |
|
src_cur = 0 |
|
85
|
1 |
|
tar_cur = 0 |
|
86
|
1 |
|
lcss = 0 |
|
87
|
1 |
|
local_cs = 0 |
|
88
|
|
|
|
|
89
|
1 |
|
while (src_cur < src_len) and (tar_cur < tar_len): |
|
90
|
1 |
|
if src[src_cur] == tar[tar_cur]: |
|
91
|
1 |
|
local_cs += 1 |
|
92
|
|
|
else: |
|
93
|
1 |
|
lcss += local_cs |
|
94
|
1 |
|
local_cs = 0 |
|
95
|
1 |
|
if src_cur != tar_cur: |
|
96
|
1 |
|
src_cur = tar_cur = max(src_cur, tar_cur) |
|
97
|
1 |
|
for i in range(max_offset): |
|
98
|
1 |
|
if not ( |
|
99
|
|
|
(src_cur + i < src_len) or (tar_cur + i < tar_len) |
|
|
|
|
|
|
100
|
|
|
): |
|
101
|
1 |
|
break |
|
102
|
1 |
|
if (src_cur + i < src_len) and ( |
|
103
|
|
|
src[src_cur + i] == tar[tar_cur] |
|
|
|
|
|
|
104
|
|
|
): |
|
105
|
1 |
|
src_cur += i |
|
106
|
1 |
|
local_cs += 1 |
|
107
|
1 |
|
break |
|
108
|
1 |
|
if (tar_cur + i < tar_len) and ( |
|
109
|
|
|
src[src_cur] == tar[tar_cur + i] |
|
|
|
|
|
|
110
|
|
|
): |
|
111
|
1 |
|
tar_cur += i |
|
112
|
1 |
|
local_cs += 1 |
|
113
|
1 |
|
break |
|
114
|
|
|
|
|
115
|
1 |
|
src_cur += 1 |
|
116
|
1 |
|
tar_cur += 1 |
|
117
|
|
|
|
|
118
|
1 |
|
lcss += local_cs |
|
119
|
1 |
|
return round(max(src_len, tar_len) - lcss) |
|
120
|
|
|
|
|
121
|
|
|
|
|
122
|
1 |
|
def sift4_simplest(src, tar, max_offset=5): |
|
123
|
|
|
"""Return the "simplest" Sift4 distance between two terms. |
|
124
|
|
|
|
|
125
|
|
|
This is a wrapper for :py:meth:`Sift4Simplest.dist_abs`. |
|
126
|
|
|
|
|
127
|
|
|
Parameters |
|
128
|
|
|
---------- |
|
129
|
|
|
src : str |
|
130
|
|
|
Source string for comparison |
|
131
|
|
|
tar : str |
|
132
|
|
|
Target string for comparison |
|
133
|
|
|
max_offset : int |
|
134
|
|
|
The number of characters to search for matching letters |
|
135
|
|
|
|
|
136
|
|
|
Returns |
|
137
|
|
|
------- |
|
138
|
|
|
int |
|
139
|
|
|
The Sift4 distance according to the simplest formula |
|
140
|
|
|
|
|
141
|
|
|
Examples |
|
142
|
|
|
-------- |
|
143
|
|
|
>>> sift4_simplest('cat', 'hat') |
|
144
|
|
|
1 |
|
145
|
|
|
>>> sift4_simplest('Niall', 'Neil') |
|
146
|
|
|
2 |
|
147
|
|
|
>>> sift4_simplest('Colin', 'Cuilen') |
|
148
|
|
|
3 |
|
149
|
|
|
>>> sift4_simplest('ATCG', 'TAGC') |
|
150
|
|
|
2 |
|
151
|
|
|
|
|
152
|
|
|
""" |
|
153
|
1 |
|
return Sift4Simplest().dist_abs(src, tar, max_offset) |
|
154
|
|
|
|
|
155
|
|
|
|
|
156
|
|
|
if __name__ == '__main__': |
|
157
|
|
|
import doctest |
|
158
|
|
|
|
|
159
|
|
|
doctest.testmod() |
|
160
|
|
|
|
chrislit /
abydos
Loading history...