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.phonetic._sfinx_bis. |
20
|
|
|
|
21
|
|
|
SfinxBis |
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__ = ['SfinxBis', 'sfinxbis'] |
38
|
|
|
|
39
|
|
|
|
40
|
1 |
|
class SfinxBis(_Phonetic): |
|
|
|
|
41
|
|
|
"""SfinxBis code. |
42
|
|
|
|
43
|
|
|
SfinxBis is a Soundex-like algorithm defined in :cite:`Axelsson:2009`. |
44
|
|
|
|
45
|
|
|
This implementation follows the reference implementation: |
46
|
|
|
:cite:`Sjoo:2009`. |
47
|
|
|
|
48
|
|
|
SfinxBis is intended chiefly for Swedish names. |
49
|
|
|
""" |
50
|
|
|
|
51
|
1 |
|
_adelstitler = ( |
52
|
|
|
' DE LA ', |
53
|
|
|
' DE LAS ', |
54
|
|
|
' DE LOS ', |
55
|
|
|
' VAN DE ', |
56
|
|
|
' VAN DEN ', |
57
|
|
|
' VAN DER ', |
58
|
|
|
' VON DEM ', |
59
|
|
|
' VON DER ', |
60
|
|
|
' AF ', |
61
|
|
|
' AV ', |
62
|
|
|
' DA ', |
63
|
|
|
' DE ', |
64
|
|
|
' DEL ', |
65
|
|
|
' DEN ', |
66
|
|
|
' DES ', |
67
|
|
|
' DI ', |
68
|
|
|
' DO ', |
69
|
|
|
' DON ', |
70
|
|
|
' DOS ', |
71
|
|
|
' DU ', |
72
|
|
|
' E ', |
73
|
|
|
' IN ', |
74
|
|
|
' LA ', |
75
|
|
|
' LE ', |
76
|
|
|
' MAC ', |
77
|
|
|
' MC ', |
78
|
|
|
' VAN ', |
79
|
|
|
' VON ', |
80
|
|
|
' Y ', |
81
|
|
|
' S:T ', |
82
|
|
|
) |
83
|
|
|
|
84
|
1 |
|
_harde_vokaler = {'A', 'O', 'U', 'Å'} |
85
|
1 |
|
_mjuka_vokaler = {'E', 'I', 'Y', 'Ä', 'Ö'} |
86
|
1 |
|
_uc_c_set = { |
87
|
|
|
'B', |
88
|
|
|
'C', |
89
|
|
|
'D', |
90
|
|
|
'F', |
91
|
|
|
'G', |
92
|
|
|
'H', |
93
|
|
|
'J', |
94
|
|
|
'K', |
95
|
|
|
'L', |
96
|
|
|
'M', |
97
|
|
|
'N', |
98
|
|
|
'P', |
99
|
|
|
'Q', |
100
|
|
|
'R', |
101
|
|
|
'S', |
102
|
|
|
'T', |
103
|
|
|
'V', |
104
|
|
|
'W', |
105
|
|
|
'X', |
106
|
|
|
'Z', |
107
|
|
|
} |
108
|
1 |
|
_uc_set = { |
109
|
|
|
'A', |
110
|
|
|
'B', |
111
|
|
|
'C', |
112
|
|
|
'D', |
113
|
|
|
'E', |
114
|
|
|
'F', |
115
|
|
|
'G', |
116
|
|
|
'H', |
117
|
|
|
'I', |
118
|
|
|
'J', |
119
|
|
|
'K', |
120
|
|
|
'L', |
121
|
|
|
'M', |
122
|
|
|
'N', |
123
|
|
|
'O', |
124
|
|
|
'P', |
125
|
|
|
'Q', |
126
|
|
|
'R', |
127
|
|
|
'S', |
128
|
|
|
'T', |
129
|
|
|
'U', |
130
|
|
|
'V', |
131
|
|
|
'W', |
132
|
|
|
'X', |
133
|
|
|
'Y', |
134
|
|
|
'Z', |
135
|
|
|
'Ä', |
136
|
|
|
'Å', |
137
|
|
|
'Ö', |
138
|
|
|
} |
139
|
|
|
|
140
|
1 |
|
_trans = dict( |
141
|
|
|
zip( |
142
|
|
|
(ord(_) for _ in 'BCDFGHJKLMNPQRSTVZAOUÅEIYÄÖ'), |
|
|
|
|
143
|
|
|
'123729224551268378999999999', |
144
|
|
|
) |
145
|
|
|
) |
146
|
|
|
|
147
|
1 |
|
_substitutions = dict( |
148
|
|
|
zip( |
149
|
|
|
(ord(_) for _ in 'WZÀÁÂÃÆÇÈÉÊËÌÍÎÏÑÒÓÔÕØÙÚÛÜÝ'), |
150
|
|
|
'VSAAAAÄCEEEEIIIINOOOOÖUUUYY', |
151
|
|
|
) |
152
|
|
|
) |
153
|
|
|
|
154
|
1 |
|
def encode(self, word, max_length=-1): |
|
|
|
|
155
|
|
|
"""Return the SfinxBis code for a word. |
156
|
|
|
|
157
|
|
|
Parameters |
158
|
|
|
---------- |
159
|
|
|
word : str |
160
|
|
|
The word to transform |
161
|
|
|
max_length : int |
162
|
|
|
The length of the code returned (defaults to unlimited) |
163
|
|
|
|
164
|
|
|
Returns |
165
|
|
|
------- |
166
|
|
|
tuple |
167
|
|
|
The SfinxBis value |
168
|
|
|
|
169
|
|
|
Examples |
170
|
|
|
-------- |
171
|
|
|
>>> pe = SfinxBis() |
172
|
|
|
>>> pe.encode('Christopher') |
173
|
|
|
('K68376',) |
174
|
|
|
>>> pe.encode('Niall') |
175
|
|
|
('N4',) |
176
|
|
|
>>> pe.encode('Smith') |
177
|
|
|
('S53',) |
178
|
|
|
>>> pe.encode('Schmidt') |
179
|
|
|
('S53',) |
180
|
|
|
|
181
|
|
|
>>> pe.encode('Johansson') |
182
|
|
|
('J585',) |
183
|
|
|
>>> pe.encode('Sjöberg') |
184
|
|
|
('#162',) |
185
|
|
|
|
186
|
|
|
""" |
187
|
|
|
|
188
|
1 |
|
def _foersvensker(lokal_ordet): |
189
|
|
|
"""Return the Swedish-ized form of the word. |
190
|
|
|
|
191
|
|
|
Parameters |
192
|
|
|
---------- |
193
|
|
|
lokal_ordet : str |
194
|
|
|
Word to transform |
195
|
|
|
|
196
|
|
|
Returns |
197
|
|
|
------- |
198
|
|
|
str |
199
|
|
|
Transformed word |
200
|
|
|
|
201
|
|
|
""" |
202
|
1 |
|
lokal_ordet = lokal_ordet.replace('STIERN', 'STJÄRN') |
203
|
1 |
|
lokal_ordet = lokal_ordet.replace('HIE', 'HJ') |
204
|
1 |
|
lokal_ordet = lokal_ordet.replace('SIÖ', 'SJÖ') |
205
|
1 |
|
lokal_ordet = lokal_ordet.replace('SCH', 'SH') |
206
|
1 |
|
lokal_ordet = lokal_ordet.replace('QU', 'KV') |
207
|
1 |
|
lokal_ordet = lokal_ordet.replace('IO', 'JO') |
208
|
1 |
|
lokal_ordet = lokal_ordet.replace('PH', 'F') |
209
|
|
|
|
210
|
1 |
|
for i in self._harde_vokaler: |
211
|
1 |
|
lokal_ordet = lokal_ordet.replace(i + 'Ü', i + 'J') |
212
|
1 |
|
lokal_ordet = lokal_ordet.replace(i + 'Y', i + 'J') |
213
|
1 |
|
lokal_ordet = lokal_ordet.replace(i + 'I', i + 'J') |
214
|
1 |
|
for i in self._mjuka_vokaler: |
215
|
1 |
|
lokal_ordet = lokal_ordet.replace(i + 'Ü', i + 'J') |
216
|
1 |
|
lokal_ordet = lokal_ordet.replace(i + 'Y', i + 'J') |
217
|
1 |
|
lokal_ordet = lokal_ordet.replace(i + 'I', i + 'J') |
218
|
|
|
|
219
|
1 |
|
if 'H' in lokal_ordet: |
220
|
1 |
|
for i in self._uc_c_set: |
221
|
1 |
|
lokal_ordet = lokal_ordet.replace('H' + i, i) |
222
|
|
|
|
223
|
1 |
|
lokal_ordet = lokal_ordet.translate(self._substitutions) |
224
|
|
|
|
225
|
1 |
|
lokal_ordet = lokal_ordet.replace('Ð', 'ETH') |
226
|
1 |
|
lokal_ordet = lokal_ordet.replace('Þ', 'TH') |
227
|
1 |
|
lokal_ordet = lokal_ordet.replace('ß', 'SS') |
228
|
|
|
|
229
|
1 |
|
return lokal_ordet |
230
|
|
|
|
231
|
1 |
|
def _koda_foersta_ljudet(lokal_ordet): |
232
|
|
|
"""Return the word with the first sound coded. |
233
|
|
|
|
234
|
|
|
Parameters |
235
|
|
|
---------- |
236
|
|
|
lokal_ordet : str |
237
|
|
|
Word to transform |
238
|
|
|
|
239
|
|
|
Returns |
240
|
|
|
------- |
241
|
|
|
str |
242
|
|
|
Transformed word |
243
|
|
|
|
244
|
|
|
""" |
245
|
1 |
|
if ( |
246
|
|
|
lokal_ordet[0:1] in self._mjuka_vokaler |
|
|
|
|
247
|
|
|
or lokal_ordet[0:1] in self._harde_vokaler |
|
|
|
|
248
|
|
|
): |
249
|
1 |
|
lokal_ordet = '$' + lokal_ordet[1:] |
250
|
1 |
|
elif lokal_ordet[0:2] in ('DJ', 'GJ', 'HJ', 'LJ'): |
251
|
1 |
|
lokal_ordet = 'J' + lokal_ordet[2:] |
252
|
1 |
|
elif ( |
253
|
|
|
lokal_ordet[0:1] == 'G' |
|
|
|
|
254
|
|
|
and lokal_ordet[1:2] in self._mjuka_vokaler |
|
|
|
|
255
|
|
|
): |
256
|
1 |
|
lokal_ordet = 'J' + lokal_ordet[1:] |
257
|
1 |
|
elif lokal_ordet[0:1] == 'Q': |
258
|
1 |
|
lokal_ordet = 'K' + lokal_ordet[1:] |
259
|
1 |
|
elif lokal_ordet[0:2] == 'CH' and lokal_ordet[2:3] in frozenset( |
260
|
|
|
self._mjuka_vokaler | self._harde_vokaler |
|
|
|
|
261
|
|
|
): |
262
|
1 |
|
lokal_ordet = '#' + lokal_ordet[2:] |
263
|
1 |
|
elif ( |
264
|
|
|
lokal_ordet[0:1] == 'C' |
|
|
|
|
265
|
|
|
and lokal_ordet[1:2] in self._harde_vokaler |
|
|
|
|
266
|
|
|
): |
267
|
1 |
|
lokal_ordet = 'K' + lokal_ordet[1:] |
268
|
1 |
|
elif ( |
269
|
|
|
lokal_ordet[0:1] == 'C' and lokal_ordet[1:2] in self._uc_c_set |
|
|
|
|
270
|
|
|
): |
271
|
1 |
|
lokal_ordet = 'K' + lokal_ordet[1:] |
272
|
1 |
|
elif lokal_ordet[0:1] == 'X': |
273
|
1 |
|
lokal_ordet = 'S' + lokal_ordet[1:] |
274
|
1 |
|
elif ( |
275
|
|
|
lokal_ordet[0:1] == 'C' |
|
|
|
|
276
|
|
|
and lokal_ordet[1:2] in self._mjuka_vokaler |
|
|
|
|
277
|
|
|
): |
278
|
1 |
|
lokal_ordet = 'S' + lokal_ordet[1:] |
279
|
1 |
|
elif lokal_ordet[0:3] in ('SKJ', 'STJ', 'SCH'): |
280
|
1 |
|
lokal_ordet = '#' + lokal_ordet[3:] |
281
|
1 |
|
elif lokal_ordet[0:2] in ('SH', 'KJ', 'TJ', 'SJ'): |
282
|
1 |
|
lokal_ordet = '#' + lokal_ordet[2:] |
283
|
1 |
|
elif ( |
284
|
|
|
lokal_ordet[0:2] == 'SK' |
|
|
|
|
285
|
|
|
and lokal_ordet[2:3] in self._mjuka_vokaler |
|
|
|
|
286
|
|
|
): |
287
|
1 |
|
lokal_ordet = '#' + lokal_ordet[2:] |
288
|
1 |
|
elif ( |
289
|
|
|
lokal_ordet[0:1] == 'K' |
|
|
|
|
290
|
|
|
and lokal_ordet[1:2] in self._mjuka_vokaler |
|
|
|
|
291
|
|
|
): |
292
|
1 |
|
lokal_ordet = '#' + lokal_ordet[1:] |
293
|
1 |
|
return lokal_ordet |
294
|
|
|
|
295
|
|
|
# Steg 1, Versaler |
296
|
1 |
|
word = unicode_normalize('NFC', text_type(word.upper())) |
297
|
1 |
|
word = word.replace('ß', 'SS') |
298
|
1 |
|
word = word.replace('-', ' ') |
299
|
|
|
|
300
|
|
|
# Steg 2, Ta bort adelsprefix |
301
|
1 |
|
for adelstitel in self._adelstitler: |
302
|
1 |
|
while adelstitel in word: |
303
|
1 |
|
word = word.replace(adelstitel, ' ') |
304
|
1 |
|
if word.startswith(adelstitel[1:]): |
305
|
1 |
|
word = word[len(adelstitel) - 1 :] |
306
|
|
|
|
307
|
|
|
# Split word into tokens |
308
|
1 |
|
ordlista = word.split() |
309
|
|
|
|
310
|
|
|
# Steg 3, Ta bort dubbelteckning i början på namnet |
311
|
1 |
|
ordlista = [ |
312
|
|
|
self._delete_consecutive_repeats(ordet) for ordet in ordlista |
313
|
|
|
] |
314
|
1 |
|
if not ordlista: |
315
|
|
|
# noinspection PyRedundantParentheses |
316
|
1 |
|
return ('',) |
317
|
|
|
|
318
|
|
|
# Steg 4, Försvenskning |
319
|
1 |
|
ordlista = [_foersvensker(ordet) for ordet in ordlista] |
320
|
|
|
|
321
|
|
|
# Steg 5, Ta bort alla tecken som inte är A-Ö (65-90,196,197,214) |
322
|
1 |
|
ordlista = [ |
323
|
|
|
''.join(c for c in ordet if c in self._uc_set) |
324
|
|
|
for ordet in ordlista |
325
|
|
|
] |
326
|
|
|
|
327
|
|
|
# Steg 6, Koda första ljudet |
328
|
1 |
|
ordlista = [_koda_foersta_ljudet(ordet) for ordet in ordlista] |
329
|
|
|
|
330
|
|
|
# Steg 7, Dela upp namnet i två delar |
331
|
1 |
|
rest = [ordet[1:] for ordet in ordlista] |
332
|
|
|
|
333
|
|
|
# Steg 8, Utför fonetisk transformation i resten |
334
|
1 |
|
rest = [ordet.replace('DT', 'T') for ordet in rest] |
335
|
1 |
|
rest = [ordet.replace('X', 'KS') for ordet in rest] |
336
|
|
|
|
337
|
|
|
# Steg 9, Koda resten till en sifferkod |
338
|
1 |
|
for vokal in self._mjuka_vokaler: |
339
|
1 |
|
rest = [ordet.replace('C' + vokal, '8' + vokal) for ordet in rest] |
340
|
1 |
|
rest = [ordet.translate(self._trans) for ordet in rest] |
341
|
|
|
|
342
|
|
|
# Steg 10, Ta bort intilliggande dubbletter |
343
|
1 |
|
rest = [self._delete_consecutive_repeats(ordet) for ordet in rest] |
344
|
|
|
|
345
|
|
|
# Steg 11, Ta bort alla "9" |
346
|
1 |
|
rest = [ordet.replace('9', '') for ordet in rest] |
347
|
|
|
|
348
|
|
|
# Steg 12, Sätt ihop delarna igen |
349
|
1 |
|
ordlista = [ |
350
|
|
|
''.join(ordet) for ordet in zip((_[0:1] for _ in ordlista), rest) |
|
|
|
|
351
|
|
|
] |
352
|
|
|
|
353
|
|
|
# truncate, if max_length is set |
354
|
1 |
|
if max_length > 0: |
355
|
1 |
|
ordlista = [ordet[:max_length] for ordet in ordlista] |
356
|
|
|
|
357
|
1 |
|
return tuple(ordlista) |
358
|
|
|
|
359
|
|
|
|
360
|
1 |
|
def sfinxbis(word, max_length=-1): |
361
|
|
|
"""Return the SfinxBis code for a word. |
362
|
|
|
|
363
|
|
|
This is a wrapper for :py:meth:`SfinxBis.encode`. |
364
|
|
|
|
365
|
|
|
Parameters |
366
|
|
|
---------- |
367
|
|
|
word : str |
368
|
|
|
The word to transform |
369
|
|
|
max_length : int |
370
|
|
|
The length of the code returned (defaults to unlimited) |
371
|
|
|
|
372
|
|
|
Returns |
373
|
|
|
------- |
374
|
|
|
tuple |
375
|
|
|
The SfinxBis value |
376
|
|
|
|
377
|
|
|
Examples |
378
|
|
|
-------- |
379
|
|
|
>>> sfinxbis('Christopher') |
380
|
|
|
('K68376',) |
381
|
|
|
>>> sfinxbis('Niall') |
382
|
|
|
('N4',) |
383
|
|
|
>>> sfinxbis('Smith') |
384
|
|
|
('S53',) |
385
|
|
|
>>> sfinxbis('Schmidt') |
386
|
|
|
('S53',) |
387
|
|
|
|
388
|
|
|
>>> sfinxbis('Johansson') |
389
|
|
|
('J585',) |
390
|
|
|
>>> sfinxbis('Sjöberg') |
391
|
|
|
('#162',) |
392
|
|
|
|
393
|
|
|
""" |
394
|
1 |
|
return SfinxBis().encode(word, max_length) |
395
|
|
|
|
396
|
|
|
|
397
|
|
|
if __name__ == '__main__': |
398
|
|
|
import doctest |
399
|
|
|
|
400
|
|
|
doctest.testmod() |
401
|
|
|
|