|
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.phonetic._soundex_br. |
|
20
|
|
|
|
|
21
|
|
|
SoundexBR |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
1 |
|
from __future__ import ( |
|
25
|
|
|
absolute_import, |
|
26
|
|
|
division, |
|
27
|
|
|
print_function, |
|
28
|
|
|
unicode_literals, |
|
29
|
|
|
) |
|
30
|
|
|
|
|
31
|
1 |
|
from unicodedata import normalize as unicode_normalize |
|
32
|
|
|
|
|
33
|
1 |
|
from six import text_type |
|
34
|
|
|
|
|
35
|
1 |
|
from ._phonetic import _Phonetic |
|
36
|
|
|
|
|
37
|
1 |
|
__all__ = ['SoundexBR', 'soundex_br'] |
|
38
|
|
|
|
|
39
|
|
|
|
|
40
|
1 |
|
class SoundexBR(_Phonetic): |
|
|
|
|
|
|
41
|
|
|
"""SoundexBR. |
|
42
|
|
|
|
|
43
|
|
|
This is based on :cite:`Marcelino:2015`. |
|
44
|
|
|
""" |
|
45
|
|
|
|
|
46
|
1 |
|
_trans = dict( |
|
47
|
|
|
zip( |
|
48
|
|
|
(ord(_) for _ in 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'), |
|
|
|
|
|
|
49
|
|
|
'01230120022455012623010202', |
|
50
|
|
|
) |
|
51
|
|
|
) |
|
52
|
|
|
|
|
53
|
1 |
|
def encode(self, word, max_length=4, zero_pad=True): |
|
|
|
|
|
|
54
|
|
|
"""Return the SoundexBR encoding of a word. |
|
55
|
|
|
|
|
56
|
|
|
Parameters |
|
57
|
|
|
---------- |
|
58
|
|
|
word : str |
|
59
|
|
|
The word to transform |
|
60
|
|
|
max_length : int |
|
61
|
|
|
The length of the code returned (defaults to 4) |
|
62
|
|
|
zero_pad : bool |
|
63
|
|
|
Pad the end of the return value with 0s to achieve a max_length |
|
64
|
|
|
string |
|
65
|
|
|
|
|
66
|
|
|
Returns |
|
67
|
|
|
------- |
|
68
|
|
|
str |
|
69
|
|
|
The SoundexBR code |
|
70
|
|
|
|
|
71
|
|
|
Examples |
|
72
|
|
|
-------- |
|
73
|
|
|
>>> soundex_br('Oliveira') |
|
74
|
|
|
'O416' |
|
75
|
|
|
>>> soundex_br('Almeida') |
|
76
|
|
|
'A453' |
|
77
|
|
|
>>> soundex_br('Barbosa') |
|
78
|
|
|
'B612' |
|
79
|
|
|
>>> soundex_br('Araújo') |
|
80
|
|
|
'A620' |
|
81
|
|
|
>>> soundex_br('Gonçalves') |
|
82
|
|
|
'G524' |
|
83
|
|
|
>>> soundex_br('Goncalves') |
|
84
|
|
|
'G524' |
|
85
|
|
|
|
|
86
|
|
|
""" |
|
87
|
1 |
|
word = unicode_normalize('NFKD', text_type(word.upper())) |
|
88
|
1 |
|
word = ''.join(c for c in word if c in self._uc_set) |
|
89
|
|
|
|
|
90
|
1 |
|
if word[:2] == 'WA': |
|
91
|
1 |
|
first = 'V' |
|
92
|
1 |
|
elif word[:1] == 'K' and word[1:2] in {'A', 'O', 'U'}: |
|
93
|
1 |
|
first = 'C' |
|
94
|
1 |
|
elif word[:1] == 'C' and word[1:2] in {'I', 'E'}: |
|
95
|
1 |
|
first = 'S' |
|
96
|
1 |
|
elif word[:1] == 'G' and word[1:2] in {'E', 'I'}: |
|
97
|
1 |
|
first = 'J' |
|
98
|
1 |
|
elif word[:1] == 'Y': |
|
99
|
1 |
|
first = 'I' |
|
100
|
1 |
|
elif word[:1] == 'H': |
|
101
|
1 |
|
first = word[1:2] |
|
102
|
1 |
|
word = word[1:] |
|
103
|
|
|
else: |
|
104
|
1 |
|
first = word[:1] |
|
105
|
|
|
|
|
106
|
1 |
|
sdx = first + word[1:].translate(self._trans) |
|
107
|
1 |
|
sdx = self._delete_consecutive_repeats(sdx) |
|
108
|
1 |
|
sdx = sdx.replace('0', '') |
|
109
|
|
|
|
|
110
|
1 |
|
if zero_pad: |
|
111
|
1 |
|
sdx += '0' * max_length |
|
112
|
|
|
|
|
113
|
1 |
|
return sdx[:max_length] |
|
114
|
|
|
|
|
115
|
|
|
|
|
116
|
1 |
|
def soundex_br(word, max_length=4, zero_pad=True): |
|
117
|
|
|
"""Return the SoundexBR encoding of a word. |
|
118
|
|
|
|
|
119
|
|
|
This is a wrapper for :py:meth:`SoundexBR.encode`. |
|
120
|
|
|
|
|
121
|
|
|
Parameters |
|
122
|
|
|
---------- |
|
123
|
|
|
word : str |
|
124
|
|
|
The word to transform |
|
125
|
|
|
max_length : int |
|
126
|
|
|
The length of the code returned (defaults to 4) |
|
127
|
|
|
zero_pad : bool |
|
128
|
|
|
Pad the end of the return value with 0s to achieve a max_length string |
|
129
|
|
|
|
|
130
|
|
|
Returns |
|
131
|
|
|
------- |
|
132
|
|
|
str |
|
133
|
|
|
The SoundexBR code |
|
134
|
|
|
|
|
135
|
|
|
Examples |
|
136
|
|
|
-------- |
|
137
|
|
|
>>> soundex_br('Oliveira') |
|
138
|
|
|
'O416' |
|
139
|
|
|
>>> soundex_br('Almeida') |
|
140
|
|
|
'A453' |
|
141
|
|
|
>>> soundex_br('Barbosa') |
|
142
|
|
|
'B612' |
|
143
|
|
|
>>> soundex_br('Araújo') |
|
144
|
|
|
'A620' |
|
145
|
|
|
>>> soundex_br('Gonçalves') |
|
146
|
|
|
'G524' |
|
147
|
|
|
>>> soundex_br('Goncalves') |
|
148
|
|
|
'G524' |
|
149
|
|
|
|
|
150
|
|
|
""" |
|
151
|
1 |
|
return SoundexBR().encode(word, max_length, zero_pad) |
|
152
|
|
|
|
|
153
|
|
|
|
|
154
|
|
|
if __name__ == '__main__': |
|
155
|
|
|
import doctest |
|
156
|
|
|
|
|
157
|
|
|
doctest.testmod() |
|
158
|
|
|
|
chrislit /
abydos
Loading history...