|
1
|
|
|
# Copyright 2018-2020 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.fingerprint._count. |
|
18
|
|
|
|
|
19
|
1 |
|
Cisłak & Grabowski's count fingerprint |
|
20
|
|
|
""" |
|
21
|
|
|
|
|
22
|
|
|
from collections import Counter |
|
23
|
|
|
from typing import Tuple |
|
24
|
1 |
|
|
|
25
|
|
|
from ._fingerprint import MOST_COMMON_LETTERS_CG, _Fingerprint |
|
26
|
|
|
|
|
27
|
|
|
__all__ = ['Count'] |
|
28
|
|
|
|
|
29
|
|
|
|
|
30
|
|
|
class Count(_Fingerprint): |
|
31
|
1 |
|
"""Count Fingerprint. |
|
32
|
|
|
|
|
33
|
1 |
|
Based on the count fingerprint from :cite:`Cislak:2017`. |
|
34
|
|
|
|
|
35
|
1 |
|
.. versionadded:: 0.3.6 |
|
36
|
1 |
|
""" |
|
37
|
|
|
|
|
38
|
1 |
|
def __init__( |
|
39
|
|
|
self, |
|
40
|
|
|
n_bits: int = 16, |
|
41
|
1 |
|
most_common: Tuple[str, ...] = MOST_COMMON_LETTERS_CG, |
|
42
|
|
|
) -> None: |
|
43
|
|
|
"""Initialize Count instance. |
|
44
|
|
|
|
|
45
|
|
|
Parameters |
|
46
|
|
|
---------- |
|
47
|
|
|
n_bits : int |
|
48
|
|
|
Number of bits in the fingerprint returned |
|
49
|
1 |
|
most_common : list |
|
50
|
|
|
The most common tokens in the target language, ordered by frequency |
|
51
|
|
|
|
|
52
|
|
|
|
|
53
|
|
|
.. versionadded:: 0.4.0 |
|
54
|
|
|
|
|
55
|
|
|
""" |
|
56
|
|
|
super(Count, self).__init__() |
|
57
|
|
|
self._n_bits = n_bits |
|
58
|
|
|
self._most_common = most_common |
|
59
|
|
|
|
|
60
|
|
|
def fingerprint(self, word: str) -> str: |
|
61
|
|
|
"""Return the count fingerprint. |
|
62
|
|
|
|
|
63
|
1 |
|
Parameters |
|
64
|
1 |
|
---------- |
|
65
|
1 |
|
word : str |
|
66
|
|
|
The word to fingerprint |
|
67
|
1 |
|
|
|
68
|
|
|
Returns |
|
69
|
|
|
------- |
|
70
|
|
|
str |
|
71
|
|
|
The count fingerprint |
|
72
|
|
|
|
|
73
|
|
|
Examples |
|
74
|
|
|
-------- |
|
75
|
|
|
>>> cf = Count() |
|
76
|
|
|
>>> cf.fingerprint('hat') |
|
77
|
|
|
'0001010000000001' |
|
78
|
|
|
>>> cf.fingerprint('niall') |
|
79
|
|
|
'0000010001010000' |
|
80
|
|
|
>>> cf.fingerprint('colin') |
|
81
|
|
|
'0000000101010000' |
|
82
|
|
|
>>> cf.fingerprint('atcg') |
|
83
|
|
|
'0001010000000000' |
|
84
|
|
|
>>> cf.fingerprint('entreatment') |
|
85
|
|
|
'1111010000100000' |
|
86
|
|
|
|
|
87
|
|
|
|
|
88
|
|
|
.. versionadded:: 0.3.0 |
|
89
|
|
|
.. versionchanged:: 0.3.6 |
|
90
|
|
|
Encapsulated in class |
|
91
|
|
|
.. versionchanged:: 0.6.0 |
|
92
|
|
|
Changed to return a str and added fingerprint_int method |
|
93
|
|
|
|
|
94
|
|
|
""" |
|
95
|
|
|
return ('{:0' + str(self._n_bits) + 'b}').format( |
|
96
|
|
|
self.fingerprint_int(word) |
|
97
|
|
|
) |
|
98
|
|
|
|
|
99
|
|
|
def fingerprint_int(self, word: str) -> int: |
|
100
|
1 |
|
"""Return the count fingerprint. |
|
101
|
1 |
|
|
|
102
|
1 |
|
Parameters |
|
103
|
|
|
---------- |
|
104
|
1 |
|
word : str |
|
105
|
1 |
|
The word to fingerprint |
|
106
|
|
|
|
|
107
|
1 |
|
Returns |
|
108
|
1 |
|
------- |
|
109
|
1 |
|
int |
|
110
|
1 |
|
The count fingerprint as an int |
|
111
|
1 |
|
|
|
112
|
|
|
Examples |
|
113
|
1 |
|
-------- |
|
114
|
|
|
>>> cf = Count() |
|
115
|
1 |
|
>>> cf.fingerprint_int('hat') |
|
116
|
1 |
|
5121 |
|
117
|
|
|
>>> cf.fingerprint_int('niall') |
|
118
|
1 |
|
1104 |
|
119
|
|
|
>>> cf.fingerprint_int('colin') |
|
120
|
|
|
336 |
|
121
|
1 |
|
>>> cf.fingerprint_int('atcg') |
|
122
|
|
|
5120 |
|
123
|
|
|
>>> cf.fingerprint_int('entreatment') |
|
124
|
|
|
62496 |
|
125
|
|
|
|
|
126
|
|
|
|
|
127
|
1 |
|
.. versionadded:: 0.6.0 |
|
128
|
|
|
|
|
129
|
|
|
""" |
|
130
|
|
|
n_bits = self._n_bits |
|
131
|
|
|
if n_bits % 2: |
|
132
|
|
|
n_bits += 1 |
|
133
|
|
|
|
|
134
|
|
|
letter_counts = Counter(word) |
|
135
|
|
|
fingerprint = 0 |
|
136
|
|
|
|
|
137
|
|
|
for letter in self._most_common: |
|
138
|
|
|
if n_bits: |
|
139
|
|
|
fingerprint <<= 2 |
|
140
|
|
|
fingerprint += letter_counts[letter] & 3 |
|
141
|
|
|
n_bits -= 2 |
|
142
|
|
|
else: |
|
143
|
|
|
break |
|
144
|
|
|
|
|
145
|
|
|
if n_bits: |
|
146
|
|
|
fingerprint <<= n_bits |
|
147
|
|
|
|
|
148
|
|
|
return fingerprint |
|
149
|
|
|
|
|
150
|
|
|
|
|
151
|
|
|
if __name__ == '__main__': |
|
152
|
|
|
import doctest |
|
153
|
|
|
|
|
154
|
|
|
doctest.testmod() |
|
155
|
|
|
|
chrislit /
abydos
