|
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._metaphone. |
|
20
|
|
|
|
|
21
|
|
|
The phonetic._metaphone module implements Metaphone and Double Metaphone. |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
1 |
|
from __future__ import unicode_literals |
|
25
|
|
|
|
|
26
|
1 |
|
from six.moves import range |
|
27
|
|
|
|
|
28
|
1 |
|
from ._phonetic import Phonetic |
|
29
|
|
|
|
|
30
|
1 |
|
__all__ = ['DoubleMetaphone', 'Metaphone', 'double_metaphone', 'metaphone'] |
|
31
|
|
|
|
|
32
|
|
|
|
|
33
|
1 |
|
class Metaphone(Phonetic): |
|
|
|
|
|
|
34
|
|
|
"""Metaphone. |
|
35
|
|
|
|
|
36
|
|
|
Based on Lawrence Philips' Pick BASIC code from 1990 :cite:`Philips:1990`, |
|
37
|
|
|
as described in :cite:`Philips:1990b`. |
|
38
|
|
|
This incorporates some corrections to the above code, particularly |
|
39
|
|
|
some of those suggested by Michael Kuhn in :cite:`Kuhn:1995`. |
|
40
|
|
|
""" |
|
41
|
|
|
|
|
42
|
1 |
|
_frontv = {'E', 'I', 'Y'} |
|
43
|
1 |
|
_varson = {'C', 'G', 'P', 'S', 'T'} |
|
44
|
|
|
|
|
45
|
1 |
|
def encode(self, word, max_length=-1): |
|
|
|
|
|
|
46
|
|
|
"""Return the Metaphone code for a word. |
|
47
|
|
|
|
|
48
|
|
|
Based on Lawrence Philips' Pick BASIC code from 1990 |
|
49
|
|
|
:cite:`Philips:1990`, as described in :cite:`Philips:1990b`. |
|
50
|
|
|
This incorporates some corrections to the above code, particularly |
|
51
|
|
|
some of those suggested by Michael Kuhn in :cite:`Kuhn:1995`. |
|
52
|
|
|
|
|
53
|
|
|
:param str word: the word to transform |
|
54
|
|
|
:param int max_length: the maximum length of the returned Metaphone |
|
55
|
|
|
code (defaults to 64, but in Philips' original implementation this |
|
56
|
|
|
was 4) |
|
57
|
|
|
:returns: the Metaphone value |
|
58
|
|
|
:rtype: str |
|
59
|
|
|
|
|
60
|
|
|
>>> pe = Metaphone() |
|
61
|
|
|
>>> pe.encode('Christopher') |
|
62
|
|
|
'KRSTFR' |
|
63
|
|
|
>>> pe.encode('Niall') |
|
64
|
|
|
'NL' |
|
65
|
|
|
>>> pe.encode('Smith') |
|
66
|
|
|
'SM0' |
|
67
|
|
|
>>> pe.encode('Schmidt') |
|
68
|
|
|
'SKMTT' |
|
69
|
|
|
""" |
|
70
|
|
|
# Require a max_length of at least 4 |
|
71
|
1 |
|
if max_length != -1: |
|
72
|
1 |
|
max_length = max(4, max_length) |
|
73
|
|
|
else: |
|
74
|
1 |
|
max_length = 64 |
|
75
|
|
|
|
|
76
|
|
|
# As in variable sound--those modified by adding an "h" |
|
77
|
1 |
|
ename = ''.join(c for c in word.upper() if c.isalnum()) |
|
78
|
1 |
|
ename = ename.replace('ß', 'SS') |
|
79
|
|
|
|
|
80
|
|
|
# Delete non-alphanumeric characters and make all caps |
|
81
|
1 |
|
if not ename: |
|
82
|
1 |
|
return '' |
|
83
|
1 |
|
if ename[0:2] in {'PN', 'AE', 'KN', 'GN', 'WR'}: |
|
84
|
1 |
|
ename = ename[1:] |
|
85
|
1 |
|
elif ename[0] == 'X': |
|
86
|
1 |
|
ename = 'S' + ename[1:] |
|
87
|
1 |
|
elif ename[0:2] == 'WH': |
|
88
|
1 |
|
ename = 'W' + ename[2:] |
|
89
|
|
|
|
|
90
|
|
|
# Convert to metaphone |
|
91
|
1 |
|
elen = len(ename) - 1 |
|
92
|
1 |
|
metaph = '' |
|
93
|
1 |
|
for i in range(len(ename)): |
|
|
|
|
|
|
94
|
1 |
|
if len(metaph) >= max_length: |
|
95
|
1 |
|
break |
|
96
|
1 |
|
if ( |
|
97
|
|
|
ename[i] not in {'G', 'T'} |
|
|
|
|
|
|
98
|
|
|
and i > 0 |
|
|
|
|
|
|
99
|
|
|
and ename[i - 1] == ename[i] |
|
|
|
|
|
|
100
|
|
|
): |
|
101
|
1 |
|
continue |
|
102
|
|
|
|
|
103
|
1 |
|
if ename[i] in self._uc_v_set and i == 0: |
|
104
|
1 |
|
metaph = ename[i] |
|
105
|
|
|
|
|
106
|
1 |
|
elif ename[i] == 'B': |
|
107
|
1 |
|
if i != elen or ename[i - 1] != 'M': |
|
108
|
1 |
|
metaph += ename[i] |
|
109
|
|
|
|
|
110
|
1 |
|
elif ename[i] == 'C': |
|
111
|
1 |
|
if not ( |
|
112
|
|
|
i > 0 |
|
|
|
|
|
|
113
|
|
|
and ename[i - 1] == 'S' |
|
|
|
|
|
|
114
|
|
|
and ename[i + 1 : i + 2] in self._frontv |
|
|
|
|
|
|
115
|
|
|
): |
|
116
|
1 |
|
if ename[i + 1 : i + 3] == 'IA': |
|
117
|
1 |
|
metaph += 'X' |
|
118
|
1 |
|
elif ename[i + 1 : i + 2] in self._frontv: |
|
119
|
1 |
|
metaph += 'S' |
|
120
|
1 |
|
elif i > 0 and ename[i - 1 : i + 2] == 'SCH': |
|
121
|
1 |
|
metaph += 'K' |
|
122
|
1 |
|
elif ename[i + 1 : i + 2] == 'H': |
|
123
|
1 |
|
if ( |
|
124
|
|
|
i == 0 |
|
|
|
|
|
|
125
|
|
|
and i + 1 < elen |
|
|
|
|
|
|
126
|
|
|
and ename[i + 2 : i + 3] not in self._uc_v_set |
|
|
|
|
|
|
127
|
|
|
): |
|
128
|
1 |
|
metaph += 'K' |
|
129
|
|
|
else: |
|
130
|
1 |
|
metaph += 'X' |
|
131
|
|
|
else: |
|
132
|
1 |
|
metaph += 'K' |
|
133
|
|
|
|
|
134
|
1 |
|
elif ename[i] == 'D': |
|
135
|
1 |
|
if ( |
|
136
|
|
|
ename[i + 1 : i + 2] == 'G' |
|
|
|
|
|
|
137
|
|
|
and ename[i + 2 : i + 3] in self._frontv |
|
|
|
|
|
|
138
|
|
|
): |
|
139
|
1 |
|
metaph += 'J' |
|
140
|
|
|
else: |
|
141
|
1 |
|
metaph += 'T' |
|
142
|
|
|
|
|
143
|
1 |
|
elif ename[i] == 'G': |
|
144
|
1 |
|
if ename[i + 1 : i + 2] == 'H' and not ( |
|
145
|
|
|
i + 1 == elen or ename[i + 2 : i + 3] not in self._uc_v_set |
|
|
|
|
|
|
146
|
|
|
): |
|
147
|
1 |
|
continue |
|
148
|
1 |
|
elif i > 0 and ( |
|
149
|
|
|
(i + 1 == elen and ename[i + 1] == 'N') |
|
|
|
|
|
|
150
|
|
|
or (i + 3 == elen and ename[i + 1 : i + 4] == 'NED') |
|
|
|
|
|
|
151
|
|
|
): |
|
152
|
1 |
|
continue |
|
153
|
1 |
|
elif ( |
|
154
|
|
|
i - 1 > 0 |
|
|
|
|
|
|
155
|
|
|
and i + 1 <= elen |
|
|
|
|
|
|
156
|
|
|
and ename[i - 1] == 'D' |
|
|
|
|
|
|
157
|
|
|
and ename[i + 1] in self._frontv |
|
|
|
|
|
|
158
|
|
|
): |
|
159
|
1 |
|
continue |
|
160
|
1 |
|
elif ename[i + 1 : i + 2] == 'G': |
|
161
|
1 |
|
continue |
|
162
|
1 |
|
elif ename[i + 1 : i + 2] in self._frontv: |
|
163
|
1 |
|
if i == 0 or ename[i - 1] != 'G': |
|
164
|
1 |
|
metaph += 'J' |
|
165
|
|
|
else: |
|
166
|
1 |
|
metaph += 'K' |
|
167
|
|
|
else: |
|
168
|
1 |
|
metaph += 'K' |
|
169
|
|
|
|
|
170
|
1 |
|
elif ename[i] == 'H': |
|
171
|
1 |
|
if ( |
|
172
|
|
|
i > 0 |
|
|
|
|
|
|
173
|
|
|
and ename[i - 1] in self._uc_v_set |
|
|
|
|
|
|
174
|
|
|
and ename[i + 1 : i + 2] not in self._uc_v_set |
|
|
|
|
|
|
175
|
|
|
): |
|
176
|
1 |
|
continue |
|
177
|
1 |
|
elif i > 0 and ename[i - 1] in self._varson: |
|
178
|
1 |
|
continue |
|
179
|
|
|
else: |
|
180
|
1 |
|
metaph += 'H' |
|
181
|
|
|
|
|
182
|
1 |
|
elif ename[i] in {'F', 'J', 'L', 'M', 'N', 'R'}: |
|
183
|
1 |
|
metaph += ename[i] |
|
184
|
|
|
|
|
185
|
1 |
|
elif ename[i] == 'K': |
|
186
|
1 |
|
if i > 0 and ename[i - 1] == 'C': |
|
187
|
1 |
|
continue |
|
188
|
|
|
else: |
|
189
|
1 |
|
metaph += 'K' |
|
190
|
|
|
|
|
191
|
1 |
|
elif ename[i] == 'P': |
|
192
|
1 |
|
if ename[i + 1 : i + 2] == 'H': |
|
193
|
1 |
|
metaph += 'F' |
|
194
|
|
|
else: |
|
195
|
1 |
|
metaph += 'P' |
|
196
|
|
|
|
|
197
|
1 |
|
elif ename[i] == 'Q': |
|
198
|
1 |
|
metaph += 'K' |
|
199
|
|
|
|
|
200
|
1 |
|
elif ename[i] == 'S': |
|
201
|
1 |
|
if ( |
|
202
|
|
|
i > 0 |
|
|
|
|
|
|
203
|
|
|
and i + 2 <= elen |
|
|
|
|
|
|
204
|
|
|
and ename[i + 1] == 'I' |
|
|
|
|
|
|
205
|
|
|
and ename[i + 2] in 'OA' |
|
|
|
|
|
|
206
|
|
|
): |
|
207
|
1 |
|
metaph += 'X' |
|
208
|
1 |
|
elif ename[i + 1 : i + 2] == 'H': |
|
209
|
1 |
|
metaph += 'X' |
|
210
|
|
|
else: |
|
211
|
1 |
|
metaph += 'S' |
|
212
|
|
|
|
|
213
|
1 |
|
elif ename[i] == 'T': |
|
214
|
1 |
|
if ( |
|
215
|
|
|
i > 0 |
|
|
|
|
|
|
216
|
|
|
and i + 2 <= elen |
|
|
|
|
|
|
217
|
|
|
and ename[i + 1] == 'I' |
|
|
|
|
|
|
218
|
|
|
and ename[i + 2] in {'A', 'O'} |
|
|
|
|
|
|
219
|
|
|
): |
|
220
|
1 |
|
metaph += 'X' |
|
221
|
1 |
|
elif ename[i + 1 : i + 2] == 'H': |
|
222
|
1 |
|
metaph += '0' |
|
223
|
1 |
|
elif ename[i + 1 : i + 3] != 'CH': |
|
224
|
1 |
|
if ename[i - 1 : i] != 'T': |
|
225
|
1 |
|
metaph += 'T' |
|
226
|
|
|
|
|
227
|
1 |
|
elif ename[i] == 'V': |
|
228
|
1 |
|
metaph += 'F' |
|
229
|
|
|
|
|
230
|
1 |
|
elif ename[i] in 'WY': |
|
231
|
1 |
|
if ename[i + 1 : i + 2] in self._uc_v_set: |
|
232
|
1 |
|
metaph += ename[i] |
|
233
|
|
|
|
|
234
|
1 |
|
elif ename[i] == 'X': |
|
235
|
1 |
|
metaph += 'KS' |
|
236
|
|
|
|
|
237
|
1 |
|
elif ename[i] == 'Z': |
|
238
|
1 |
|
metaph += 'S' |
|
239
|
|
|
|
|
240
|
1 |
|
return metaph |
|
241
|
|
|
|
|
242
|
|
|
|
|
243
|
1 |
|
def metaphone(word, max_length=-1): |
|
244
|
|
|
"""Return the Metaphone code for a word. |
|
245
|
|
|
|
|
246
|
|
|
This is a wrapper for :py:meth:`Metaphone.encode`. |
|
247
|
|
|
|
|
248
|
|
|
:param str word: the word to transform |
|
249
|
|
|
:param int max_length: the maximum length of the returned Metaphone code |
|
250
|
|
|
(defaults to 64, but in Philips' original implementation this was 4) |
|
251
|
|
|
:returns: the Metaphone value |
|
252
|
|
|
:rtype: str |
|
253
|
|
|
|
|
254
|
|
|
>>> metaphone('Christopher') |
|
255
|
|
|
'KRSTFR' |
|
256
|
|
|
>>> metaphone('Niall') |
|
257
|
|
|
'NL' |
|
258
|
|
|
>>> metaphone('Smith') |
|
259
|
|
|
'SM0' |
|
260
|
|
|
>>> metaphone('Schmidt') |
|
261
|
|
|
'SKMTT' |
|
262
|
|
|
""" |
|
263
|
1 |
|
return Metaphone().encode(word, max_length) |
|
264
|
|
|
|
|
265
|
|
|
|
|
266
|
1 |
|
class DoubleMetaphone(Phonetic): |
|
|
|
|
|
|
267
|
|
|
"""Double Metaphone. |
|
268
|
|
|
|
|
269
|
|
|
Based on Lawrence Philips' (Visual) C++ code from 1999 |
|
270
|
|
|
:cite:`Philips:2000`. |
|
271
|
|
|
""" |
|
272
|
|
|
|
|
273
|
1 |
|
def encode(self, word, max_length=-1): |
|
|
|
|
|
|
274
|
|
|
"""Return the Double Metaphone code for a word. |
|
275
|
|
|
|
|
276
|
|
|
:param word: the word to transform |
|
277
|
|
|
:param max_length: the maximum length of the returned Double Metaphone |
|
278
|
|
|
codes (defaults to 64, but in Philips' original implementation this |
|
279
|
|
|
was 4) |
|
280
|
|
|
:returns: the Double Metaphone value(s) |
|
281
|
|
|
:rtype: tuple |
|
282
|
|
|
|
|
283
|
|
|
>>> pe = DoubleMetaphone() |
|
284
|
|
|
>>> pe.encode('Christopher') |
|
285
|
|
|
('KRSTFR', '') |
|
286
|
|
|
>>> pe.encode('Niall') |
|
287
|
|
|
('NL', '') |
|
288
|
|
|
>>> pe.encode('Smith') |
|
289
|
|
|
('SM0', 'XMT') |
|
290
|
|
|
>>> pe.encode('Schmidt') |
|
291
|
|
|
('XMT', 'SMT') |
|
292
|
|
|
""" |
|
293
|
|
|
# Require a max_length of at least 4 |
|
294
|
1 |
|
if max_length != -1: |
|
295
|
1 |
|
max_length = max(4, max_length) |
|
296
|
|
|
else: |
|
297
|
1 |
|
max_length = 64 |
|
298
|
|
|
|
|
299
|
1 |
|
primary = '' |
|
300
|
1 |
|
secondary = '' |
|
301
|
|
|
|
|
302
|
1 |
|
def _slavo_germanic(): |
|
303
|
|
|
"""Return True if the word appears to be Slavic or Germanic.""" |
|
304
|
1 |
|
if 'W' in word or 'K' in word or 'CZ' in word: |
|
305
|
1 |
|
return True |
|
306
|
1 |
|
return False |
|
307
|
|
|
|
|
308
|
1 |
|
def _metaph_add(pri, sec=''): |
|
309
|
|
|
"""Return a new metaphone tuple with the supplied elements.""" |
|
310
|
1 |
|
newpri = primary |
|
311
|
1 |
|
newsec = secondary |
|
312
|
1 |
|
if pri: |
|
313
|
1 |
|
newpri += pri |
|
314
|
1 |
|
if sec: |
|
315
|
1 |
|
if sec != ' ': |
|
316
|
1 |
|
newsec += sec |
|
317
|
|
|
else: |
|
318
|
1 |
|
newsec += pri |
|
319
|
1 |
|
return newpri, newsec |
|
320
|
|
|
|
|
321
|
1 |
|
def _is_vowel(pos): |
|
322
|
|
|
"""Return True if the character at word[pos] is a vowel.""" |
|
323
|
1 |
|
if pos >= 0 and word[pos] in {'A', 'E', 'I', 'O', 'U', 'Y'}: |
|
324
|
1 |
|
return True |
|
325
|
1 |
|
return False |
|
326
|
|
|
|
|
327
|
1 |
|
def _get_at(pos): |
|
328
|
|
|
"""Return the character at word[pos].""" |
|
329
|
1 |
|
return word[pos] |
|
330
|
|
|
|
|
331
|
1 |
|
def _string_at(pos, slen, substrings): |
|
332
|
|
|
"""Return True if word[pos:pos+slen] is in substrings.""" |
|
333
|
1 |
|
if pos < 0: |
|
334
|
1 |
|
return False |
|
335
|
1 |
|
return word[pos : pos + slen] in substrings |
|
336
|
|
|
|
|
337
|
1 |
|
current = 0 |
|
338
|
1 |
|
length = len(word) |
|
339
|
1 |
|
if length < 1: |
|
340
|
1 |
|
return '', '' |
|
341
|
1 |
|
last = length - 1 |
|
342
|
|
|
|
|
343
|
1 |
|
word = word.upper() |
|
344
|
1 |
|
word = word.replace('ß', 'SS') |
|
345
|
|
|
|
|
346
|
|
|
# Pad the original string so that we can index beyond the edge of the |
|
347
|
|
|
# world |
|
348
|
1 |
|
word += ' ' |
|
349
|
|
|
|
|
350
|
|
|
# Skip these when at start of word |
|
351
|
1 |
|
if word[0:2] in {'GN', 'KN', 'PN', 'WR', 'PS'}: |
|
352
|
1 |
|
current += 1 |
|
353
|
|
|
|
|
354
|
|
|
# Initial 'X' is pronounced 'Z' e.g. 'Xavier' |
|
355
|
1 |
|
if _get_at(0) == 'X': |
|
356
|
1 |
|
primary, secondary = _metaph_add('S') # 'Z' maps to 'S' |
|
357
|
1 |
|
current += 1 |
|
358
|
|
|
|
|
359
|
|
|
# Main loop |
|
360
|
1 |
|
while True: |
|
|
|
|
|
|
361
|
1 |
|
if current >= length: |
|
362
|
1 |
|
break |
|
363
|
|
|
|
|
364
|
1 |
|
if _get_at(current) in {'A', 'E', 'I', 'O', 'U', 'Y'}: |
|
365
|
1 |
|
if current == 0: |
|
366
|
|
|
# All init vowels now map to 'A' |
|
367
|
1 |
|
primary, secondary = _metaph_add('A') |
|
368
|
1 |
|
current += 1 |
|
369
|
1 |
|
continue |
|
370
|
|
|
|
|
371
|
1 |
|
elif _get_at(current) == 'B': |
|
372
|
|
|
# "-mb", e.g", "dumb", already skipped over... |
|
373
|
1 |
|
primary, secondary = _metaph_add('P') |
|
374
|
1 |
|
if _get_at(current + 1) == 'B': |
|
375
|
1 |
|
current += 2 |
|
376
|
|
|
else: |
|
377
|
1 |
|
current += 1 |
|
378
|
1 |
|
continue |
|
379
|
|
|
|
|
380
|
1 |
|
elif _get_at(current) == 'Ç': |
|
381
|
1 |
|
primary, secondary = _metaph_add('S') |
|
382
|
1 |
|
current += 1 |
|
383
|
1 |
|
continue |
|
384
|
|
|
|
|
385
|
1 |
|
elif _get_at(current) == 'C': |
|
386
|
|
|
# Various Germanic |
|
387
|
1 |
|
if ( |
|
388
|
|
|
current > 1 |
|
|
|
|
|
|
389
|
|
|
and not _is_vowel(current - 2) |
|
|
|
|
|
|
390
|
|
|
and _string_at((current - 1), 3, {'ACH'}) |
|
|
|
|
|
|
391
|
|
|
and ( |
|
|
|
|
|
|
392
|
|
|
(_get_at(current + 2) != 'I') |
|
393
|
|
|
and ( |
|
394
|
|
|
(_get_at(current + 2) != 'E') |
|
395
|
|
|
or _string_at( |
|
396
|
|
|
(current - 2), 6, {'BACHER', 'MACHER'} |
|
397
|
|
|
) |
|
398
|
|
|
) |
|
399
|
|
|
) |
|
400
|
|
|
): |
|
401
|
1 |
|
primary, secondary = _metaph_add('K') |
|
402
|
1 |
|
current += 2 |
|
403
|
1 |
|
continue |
|
404
|
|
|
|
|
405
|
|
|
# Special case 'caesar' |
|
406
|
1 |
|
elif current == 0 and _string_at(current, 6, {'CAESAR'}): |
|
407
|
1 |
|
primary, secondary = _metaph_add('S') |
|
408
|
1 |
|
current += 2 |
|
409
|
1 |
|
continue |
|
410
|
|
|
|
|
411
|
|
|
# Italian 'chianti' |
|
412
|
1 |
|
elif _string_at(current, 4, {'CHIA'}): |
|
413
|
1 |
|
primary, secondary = _metaph_add('K') |
|
414
|
1 |
|
current += 2 |
|
415
|
1 |
|
continue |
|
416
|
|
|
|
|
417
|
1 |
|
elif _string_at(current, 2, {'CH'}): |
|
418
|
|
|
# Find 'Michael' |
|
419
|
1 |
|
if current > 0 and _string_at(current, 4, {'CHAE'}): |
|
420
|
1 |
|
primary, secondary = _metaph_add('K', 'X') |
|
421
|
1 |
|
current += 2 |
|
422
|
1 |
|
continue |
|
423
|
|
|
|
|
424
|
|
|
# Greek roots e.g. 'chemistry', 'chorus' |
|
425
|
1 |
|
elif ( |
|
426
|
|
|
current == 0 |
|
|
|
|
|
|
427
|
|
|
and ( |
|
|
|
|
|
|
428
|
|
|
_string_at((current + 1), 5, {'HARAC', 'HARIS'}) |
|
429
|
|
|
or _string_at( |
|
430
|
|
|
(current + 1), 3, {'HOR', 'HYM', 'HIA', 'HEM'} |
|
431
|
|
|
) |
|
432
|
|
|
) |
|
433
|
|
|
and not _string_at(0, 5, {'CHORE'}) |
|
|
|
|
|
|
434
|
|
|
): |
|
435
|
1 |
|
primary, secondary = _metaph_add('K') |
|
436
|
1 |
|
current += 2 |
|
437
|
1 |
|
continue |
|
438
|
|
|
|
|
439
|
|
|
# Germanic, Greek, or otherwise 'ch' for 'kh' sound |
|
440
|
1 |
|
elif ( |
|
441
|
|
|
( |
|
|
|
|
|
|
442
|
|
|
_string_at(0, 4, {'VAN ', 'VON '}) |
|
443
|
|
|
or _string_at(0, 3, {'SCH'}) |
|
444
|
|
|
) |
|
445
|
|
|
or |
|
|
|
|
|
|
446
|
|
|
# 'architect but not 'arch', 'orchestra', 'orchid' |
|
|
|
|
|
|
447
|
|
|
_string_at( |
|
|
|
|
|
|
448
|
|
|
(current - 2), 6, {'ORCHES', 'ARCHIT', 'ORCHID'} |
|
449
|
|
|
) |
|
450
|
|
|
or _string_at((current + 2), 1, {'T', 'S'}) |
|
|
|
|
|
|
451
|
|
|
or ( |
|
|
|
|
|
|
452
|
|
|
( |
|
453
|
|
|
_string_at( |
|
454
|
|
|
(current - 1), 1, {'A', 'O', 'U', 'E'} |
|
455
|
|
|
) |
|
456
|
|
|
or (current == 0) |
|
457
|
|
|
) |
|
458
|
|
|
and |
|
459
|
|
|
# e.g., 'wachtler', 'wechsler', but not 'tichner' |
|
460
|
|
|
_string_at( |
|
461
|
|
|
(current + 2), |
|
462
|
|
|
1, |
|
463
|
|
|
{ |
|
464
|
|
|
'L', |
|
465
|
|
|
'R', |
|
466
|
|
|
'N', |
|
467
|
|
|
'M', |
|
468
|
|
|
'B', |
|
469
|
|
|
'H', |
|
470
|
|
|
'F', |
|
471
|
|
|
'V', |
|
472
|
|
|
'W', |
|
473
|
|
|
' ', |
|
474
|
|
|
}, |
|
475
|
|
|
) |
|
476
|
|
|
) |
|
477
|
|
|
): |
|
478
|
1 |
|
primary, secondary = _metaph_add('K') |
|
479
|
|
|
|
|
480
|
|
|
else: |
|
481
|
1 |
|
if current > 0: |
|
482
|
1 |
|
if _string_at(0, 2, {'MC'}): |
|
483
|
|
|
# e.g., "McHugh" |
|
484
|
1 |
|
primary, secondary = _metaph_add('K') |
|
485
|
|
|
else: |
|
486
|
1 |
|
primary, secondary = _metaph_add('X', 'K') |
|
487
|
|
|
else: |
|
488
|
1 |
|
primary, secondary = _metaph_add('X') |
|
489
|
|
|
|
|
490
|
1 |
|
current += 2 |
|
491
|
1 |
|
continue |
|
492
|
|
|
|
|
493
|
|
|
# e.g, 'czerny' |
|
494
|
1 |
|
elif _string_at(current, 2, {'CZ'}) and not _string_at( |
|
495
|
|
|
(current - 2), 4, {'WICZ'} |
|
|
|
|
|
|
496
|
|
|
): |
|
497
|
1 |
|
primary, secondary = _metaph_add('S', 'X') |
|
498
|
1 |
|
current += 2 |
|
499
|
1 |
|
continue |
|
500
|
|
|
|
|
501
|
|
|
# e.g., 'focaccia' |
|
502
|
1 |
|
elif _string_at((current + 1), 3, {'CIA'}): |
|
503
|
1 |
|
primary, secondary = _metaph_add('X') |
|
504
|
1 |
|
current += 3 |
|
505
|
|
|
|
|
506
|
|
|
# double 'C', but not if e.g. 'McClellan' |
|
507
|
1 |
|
elif _string_at(current, 2, {'CC'}) and not ( |
|
508
|
|
|
(current == 1) and (_get_at(0) == 'M') |
|
|
|
|
|
|
509
|
|
|
): |
|
510
|
|
|
# 'bellocchio' but not 'bacchus' |
|
511
|
1 |
|
if _string_at( |
|
512
|
|
|
(current + 2), 1, {'I', 'E', 'H'} |
|
|
|
|
|
|
513
|
|
|
) and not _string_at((current + 2), 2, ['HU']): |
|
514
|
|
|
# 'accident', 'accede' 'succeed' |
|
515
|
1 |
|
if ( |
|
516
|
|
|
(current == 1) and _get_at(current - 1) == 'A' |
|
|
|
|
|
|
517
|
|
|
) or _string_at((current - 1), 5, {'UCCEE', 'UCCES'}): |
|
518
|
1 |
|
primary, secondary = _metaph_add('KS') |
|
519
|
|
|
# 'bacci', 'bertucci', other italian |
|
520
|
|
|
else: |
|
521
|
1 |
|
primary, secondary = _metaph_add('X') |
|
522
|
1 |
|
current += 3 |
|
523
|
1 |
|
continue |
|
524
|
|
|
else: # Pierce's rule |
|
525
|
1 |
|
primary, secondary = _metaph_add('K') |
|
526
|
1 |
|
current += 2 |
|
527
|
1 |
|
continue |
|
528
|
|
|
|
|
529
|
1 |
|
elif _string_at(current, 2, {'CK', 'CG', 'CQ'}): |
|
530
|
1 |
|
primary, secondary = _metaph_add('K') |
|
531
|
1 |
|
current += 2 |
|
532
|
1 |
|
continue |
|
533
|
|
|
|
|
534
|
1 |
|
elif _string_at(current, 2, {'CI', 'CE', 'CY'}): |
|
535
|
|
|
# Italian vs. English |
|
536
|
1 |
|
if _string_at(current, 3, {'CIO', 'CIE', 'CIA'}): |
|
537
|
1 |
|
primary, secondary = _metaph_add('S', 'X') |
|
538
|
|
|
else: |
|
539
|
1 |
|
primary, secondary = _metaph_add('S') |
|
540
|
1 |
|
current += 2 |
|
541
|
1 |
|
continue |
|
542
|
|
|
|
|
543
|
|
|
# else |
|
544
|
|
|
else: |
|
545
|
1 |
|
primary, secondary = _metaph_add('K') |
|
546
|
|
|
|
|
547
|
|
|
# name sent in 'mac caffrey', 'mac gregor |
|
548
|
1 |
|
if _string_at((current + 1), 2, {' C', ' Q', ' G'}): |
|
549
|
1 |
|
current += 3 |
|
550
|
1 |
|
elif _string_at( |
|
551
|
|
|
(current + 1), 1, {'C', 'K', 'Q'} |
|
|
|
|
|
|
552
|
|
|
) and not _string_at((current + 1), 2, {'CE', 'CI'}): |
|
553
|
1 |
|
current += 2 |
|
554
|
|
|
else: |
|
555
|
1 |
|
current += 1 |
|
556
|
1 |
|
continue |
|
557
|
|
|
|
|
558
|
1 |
|
elif _get_at(current) == 'D': |
|
559
|
1 |
|
if _string_at(current, 2, {'DG'}): |
|
560
|
1 |
|
if _string_at((current + 2), 1, {'I', 'E', 'Y'}): |
|
561
|
|
|
# e.g. 'edge' |
|
562
|
1 |
|
primary, secondary = _metaph_add('J') |
|
563
|
1 |
|
current += 3 |
|
564
|
1 |
|
continue |
|
565
|
|
|
else: |
|
566
|
|
|
# e.g. 'edgar' |
|
567
|
1 |
|
primary, secondary = _metaph_add('TK') |
|
568
|
1 |
|
current += 2 |
|
569
|
1 |
|
continue |
|
570
|
|
|
|
|
571
|
1 |
|
elif _string_at(current, 2, {'DT', 'DD'}): |
|
572
|
1 |
|
primary, secondary = _metaph_add('T') |
|
573
|
1 |
|
current += 2 |
|
574
|
1 |
|
continue |
|
575
|
|
|
|
|
576
|
|
|
# else |
|
577
|
|
|
else: |
|
578
|
1 |
|
primary, secondary = _metaph_add('T') |
|
579
|
1 |
|
current += 1 |
|
580
|
1 |
|
continue |
|
581
|
|
|
|
|
582
|
1 |
|
elif _get_at(current) == 'F': |
|
583
|
1 |
|
if _get_at(current + 1) == 'F': |
|
584
|
1 |
|
current += 2 |
|
585
|
|
|
else: |
|
586
|
1 |
|
current += 1 |
|
587
|
1 |
|
primary, secondary = _metaph_add('F') |
|
588
|
1 |
|
continue |
|
589
|
|
|
|
|
590
|
1 |
|
elif _get_at(current) == 'G': |
|
591
|
1 |
|
if _get_at(current + 1) == 'H': |
|
592
|
1 |
|
if (current > 0) and not _is_vowel(current - 1): |
|
593
|
1 |
|
primary, secondary = _metaph_add('K') |
|
594
|
1 |
|
current += 2 |
|
595
|
1 |
|
continue |
|
596
|
|
|
|
|
597
|
|
|
# 'ghislane', ghiradelli |
|
598
|
1 |
|
elif current == 0: |
|
599
|
1 |
|
if _get_at(current + 2) == 'I': |
|
600
|
1 |
|
primary, secondary = _metaph_add('J') |
|
601
|
|
|
else: |
|
602
|
1 |
|
primary, secondary = _metaph_add('K') |
|
603
|
1 |
|
current += 2 |
|
604
|
1 |
|
continue |
|
605
|
|
|
|
|
606
|
|
|
# Parker's rule (with some further refinements) - |
|
607
|
|
|
# e.g., 'hugh' |
|
608
|
1 |
|
elif ( |
|
609
|
|
|
( |
|
|
|
|
|
|
610
|
|
|
(current > 1) |
|
611
|
|
|
and _string_at((current - 2), 1, {'B', 'H', 'D'}) |
|
612
|
|
|
) |
|
613
|
|
|
or |
|
|
|
|
|
|
614
|
|
|
# e.g., 'bough' |
|
|
|
|
|
|
615
|
|
|
( |
|
|
|
|
|
|
616
|
|
|
(current > 2) |
|
617
|
|
|
and _string_at((current - 3), 1, {'B', 'H', 'D'}) |
|
618
|
|
|
) |
|
619
|
|
|
or |
|
|
|
|
|
|
620
|
|
|
# e.g., 'broughton' |
|
|
|
|
|
|
621
|
|
|
( |
|
|
|
|
|
|
622
|
|
|
(current > 3) |
|
623
|
|
|
and _string_at((current - 4), 1, {'B', 'H'}) |
|
624
|
|
|
) |
|
625
|
|
|
): |
|
626
|
1 |
|
current += 2 |
|
627
|
1 |
|
continue |
|
628
|
|
|
else: |
|
629
|
|
|
# e.g. 'laugh', 'McLaughlin', 'cough', |
|
630
|
|
|
# 'gough', 'rough', 'tough' |
|
631
|
1 |
|
if ( |
|
632
|
|
|
(current > 2) |
|
|
|
|
|
|
633
|
|
|
and (_get_at(current - 1) == 'U') |
|
|
|
|
|
|
634
|
|
|
and ( |
|
|
|
|
|
|
635
|
|
|
_string_at( |
|
636
|
|
|
(current - 3), 1, {'C', 'G', 'L', 'R', 'T'} |
|
637
|
|
|
) |
|
638
|
|
|
) |
|
639
|
|
|
): |
|
640
|
1 |
|
primary, secondary = _metaph_add('F') |
|
641
|
1 |
|
elif (current > 0) and _get_at(current - 1) != 'I': |
|
642
|
1 |
|
primary, secondary = _metaph_add('K') |
|
643
|
1 |
|
current += 2 |
|
644
|
1 |
|
continue |
|
645
|
|
|
|
|
646
|
1 |
|
elif _get_at(current + 1) == 'N': |
|
647
|
1 |
|
if ( |
|
648
|
|
|
(current == 1) |
|
|
|
|
|
|
649
|
|
|
and _is_vowel(0) |
|
|
|
|
|
|
650
|
|
|
and not _slavo_germanic() |
|
|
|
|
|
|
651
|
|
|
): |
|
652
|
1 |
|
primary, secondary = _metaph_add('KN', 'N') |
|
653
|
|
|
# not e.g. 'cagney' |
|
654
|
1 |
|
elif ( |
|
655
|
|
|
not _string_at((current + 2), 2, {'EY'}) |
|
|
|
|
|
|
656
|
|
|
and (_get_at(current + 1) != 'Y') |
|
|
|
|
|
|
657
|
|
|
and not _slavo_germanic() |
|
|
|
|
|
|
658
|
|
|
): |
|
659
|
1 |
|
primary, secondary = _metaph_add('N', 'KN') |
|
660
|
|
|
else: |
|
661
|
1 |
|
primary, secondary = _metaph_add('KN') |
|
662
|
1 |
|
current += 2 |
|
663
|
1 |
|
continue |
|
664
|
|
|
|
|
665
|
|
|
# 'tagliaro' |
|
666
|
1 |
|
elif ( |
|
667
|
|
|
_string_at((current + 1), 2, {'LI'}) |
|
|
|
|
|
|
668
|
|
|
and not _slavo_germanic() |
|
|
|
|
|
|
669
|
|
|
): |
|
670
|
1 |
|
primary, secondary = _metaph_add('KL', 'L') |
|
671
|
1 |
|
current += 2 |
|
672
|
1 |
|
continue |
|
673
|
|
|
|
|
674
|
|
|
# -ges-, -gep-, -gel-, -gie- at beginning |
|
675
|
1 |
|
elif (current == 0) and ( |
|
676
|
|
|
(_get_at(current + 1) == 'Y') |
|
|
|
|
|
|
677
|
|
|
or _string_at( |
|
|
|
|
|
|
678
|
|
|
(current + 1), |
|
679
|
|
|
2, |
|
680
|
|
|
{ |
|
681
|
|
|
'ES', |
|
682
|
|
|
'EP', |
|
683
|
|
|
'EB', |
|
684
|
|
|
'EL', |
|
685
|
|
|
'EY', |
|
686
|
|
|
'IB', |
|
687
|
|
|
'IL', |
|
688
|
|
|
'IN', |
|
689
|
|
|
'IE', |
|
690
|
|
|
'EI', |
|
691
|
|
|
'ER', |
|
692
|
|
|
}, |
|
693
|
|
|
) |
|
694
|
|
|
): |
|
695
|
1 |
|
primary, secondary = _metaph_add('K', 'J') |
|
696
|
1 |
|
current += 2 |
|
697
|
1 |
|
continue |
|
698
|
|
|
|
|
699
|
|
|
# -ger-, -gy- |
|
700
|
1 |
|
elif ( |
|
701
|
|
|
( |
|
|
|
|
|
|
702
|
|
|
_string_at((current + 1), 2, {'ER'}) |
|
703
|
|
|
or (_get_at(current + 1) == 'Y') |
|
704
|
|
|
) |
|
705
|
|
|
and not _string_at(0, 6, {'DANGER', 'RANGER', 'MANGER'}) |
|
|
|
|
|
|
706
|
|
|
and not _string_at((current - 1), 1, {'E', 'I'}) |
|
|
|
|
|
|
707
|
|
|
and not _string_at((current - 1), 3, {'RGY', 'OGY'}) |
|
|
|
|
|
|
708
|
|
|
): |
|
709
|
1 |
|
primary, secondary = _metaph_add('K', 'J') |
|
710
|
1 |
|
current += 2 |
|
711
|
1 |
|
continue |
|
712
|
|
|
|
|
713
|
|
|
# italian e.g, 'biaggi' |
|
714
|
1 |
|
elif _string_at( |
|
715
|
|
|
(current + 1), 1, {'E', 'I', 'Y'} |
|
|
|
|
|
|
716
|
|
|
) or _string_at((current - 1), 4, {'AGGI', 'OGGI'}): |
|
717
|
|
|
# obvious germanic |
|
718
|
1 |
|
if ( |
|
719
|
|
|
_string_at(0, 4, {'VAN ', 'VON '}) |
|
|
|
|
|
|
720
|
|
|
or _string_at(0, 3, {'SCH'}) |
|
|
|
|
|
|
721
|
|
|
) or _string_at((current + 1), 2, {'ET'}): |
|
722
|
1 |
|
primary, secondary = _metaph_add('K') |
|
723
|
1 |
|
elif _string_at((current + 1), 4, {'IER '}): |
|
724
|
1 |
|
primary, secondary = _metaph_add('J') |
|
725
|
|
|
else: |
|
726
|
1 |
|
primary, secondary = _metaph_add('J', 'K') |
|
727
|
1 |
|
current += 2 |
|
728
|
1 |
|
continue |
|
729
|
|
|
|
|
730
|
|
|
else: |
|
731
|
1 |
|
if _get_at(current + 1) == 'G': |
|
732
|
1 |
|
current += 2 |
|
733
|
|
|
else: |
|
734
|
1 |
|
current += 1 |
|
735
|
1 |
|
primary, secondary = _metaph_add('K') |
|
736
|
1 |
|
continue |
|
737
|
|
|
|
|
738
|
1 |
|
elif _get_at(current) == 'H': |
|
739
|
|
|
# only keep if first & before vowel or btw. 2 vowels |
|
740
|
1 |
|
if ((current == 0) or _is_vowel(current - 1)) and _is_vowel( |
|
741
|
|
|
current + 1 |
|
|
|
|
|
|
742
|
|
|
): |
|
743
|
1 |
|
primary, secondary = _metaph_add('H') |
|
744
|
1 |
|
current += 2 |
|
745
|
|
|
else: # also takes care of 'HH' |
|
746
|
1 |
|
current += 1 |
|
747
|
1 |
|
continue |
|
748
|
|
|
|
|
749
|
1 |
|
elif _get_at(current) == 'J': |
|
750
|
|
|
# obvious spanish, 'jose', 'san jacinto' |
|
751
|
1 |
|
if _string_at(current, 4, ['JOSE']) or _string_at( |
|
752
|
|
|
0, 4, {'SAN '} |
|
|
|
|
|
|
753
|
|
|
): |
|
754
|
1 |
|
if ( |
|
755
|
|
|
(current == 0) and (_get_at(current + 4) == ' ') |
|
|
|
|
|
|
756
|
|
|
) or _string_at(0, 4, ['SAN ']): |
|
757
|
1 |
|
primary, secondary = _metaph_add('H') |
|
758
|
|
|
else: |
|
759
|
1 |
|
primary, secondary = _metaph_add('J', 'H') |
|
760
|
1 |
|
current += 1 |
|
761
|
1 |
|
continue |
|
762
|
|
|
|
|
763
|
1 |
|
elif (current == 0) and not _string_at(current, 4, {'JOSE'}): |
|
764
|
|
|
# Yankelovich/Jankelowicz |
|
765
|
1 |
|
primary, secondary = _metaph_add('J', 'A') |
|
766
|
|
|
# Spanish pron. of e.g. 'bajador' |
|
767
|
1 |
|
elif ( |
|
768
|
|
|
_is_vowel(current - 1) |
|
|
|
|
|
|
769
|
|
|
and not _slavo_germanic() |
|
|
|
|
|
|
770
|
|
|
and ( |
|
|
|
|
|
|
771
|
|
|
(_get_at(current + 1) == 'A') |
|
772
|
|
|
or (_get_at(current + 1) == 'O') |
|
773
|
|
|
) |
|
774
|
|
|
): |
|
775
|
1 |
|
primary, secondary = _metaph_add('J', 'H') |
|
776
|
1 |
|
elif current == last: |
|
777
|
1 |
|
primary, secondary = _metaph_add('J', ' ') |
|
778
|
1 |
|
elif not _string_at( |
|
779
|
|
|
(current + 1), 1, {'L', 'T', 'K', 'S', 'N', 'M', 'B', 'Z'} |
|
|
|
|
|
|
780
|
|
|
) and not _string_at((current - 1), 1, {'S', 'K', 'L'}): |
|
781
|
1 |
|
primary, secondary = _metaph_add('J') |
|
782
|
|
|
|
|
783
|
1 |
|
if _get_at(current + 1) == 'J': # it could happen! |
|
784
|
1 |
|
current += 2 |
|
785
|
|
|
else: |
|
786
|
1 |
|
current += 1 |
|
787
|
1 |
|
continue |
|
788
|
|
|
|
|
789
|
1 |
|
elif _get_at(current) == 'K': |
|
790
|
1 |
|
if _get_at(current + 1) == 'K': |
|
791
|
1 |
|
current += 2 |
|
792
|
|
|
else: |
|
793
|
1 |
|
current += 1 |
|
794
|
1 |
|
primary, secondary = _metaph_add('K') |
|
795
|
1 |
|
continue |
|
796
|
|
|
|
|
797
|
1 |
|
elif _get_at(current) == 'L': |
|
798
|
1 |
|
if _get_at(current + 1) == 'L': |
|
799
|
|
|
# Spanish e.g. 'cabrillo', 'gallegos' |
|
800
|
1 |
|
if ( |
|
801
|
|
|
(current == (length - 3)) |
|
|
|
|
|
|
802
|
|
|
and _string_at( |
|
|
|
|
|
|
803
|
|
|
(current - 1), 4, {'ILLO', 'ILLA', 'ALLE'} |
|
804
|
|
|
) |
|
805
|
|
|
) or ( |
|
806
|
|
|
( |
|
807
|
|
|
_string_at((last - 1), 2, {'AS', 'OS'}) |
|
808
|
|
|
or _string_at(last, 1, {'A', 'O'}) |
|
809
|
|
|
) |
|
810
|
|
|
and _string_at((current - 1), 4, {'ALLE'}) |
|
811
|
|
|
): |
|
812
|
1 |
|
primary, secondary = _metaph_add('L', ' ') |
|
813
|
1 |
|
current += 2 |
|
814
|
1 |
|
continue |
|
815
|
1 |
|
current += 2 |
|
816
|
|
|
else: |
|
817
|
1 |
|
current += 1 |
|
818
|
1 |
|
primary, secondary = _metaph_add('L') |
|
819
|
1 |
|
continue |
|
820
|
|
|
|
|
821
|
1 |
|
elif _get_at(current) == 'M': |
|
822
|
1 |
|
if ( |
|
823
|
|
|
( |
|
|
|
|
|
|
824
|
|
|
_string_at((current - 1), 3, {'UMB'}) |
|
825
|
|
|
and ( |
|
826
|
|
|
((current + 1) == last) |
|
827
|
|
|
or _string_at((current + 2), 2, {'ER'}) |
|
828
|
|
|
) |
|
829
|
|
|
) |
|
830
|
|
|
or |
|
|
|
|
|
|
831
|
|
|
# 'dumb', 'thumb' |
|
|
|
|
|
|
832
|
|
|
(_get_at(current + 1) == 'M') |
|
|
|
|
|
|
833
|
|
|
): |
|
834
|
1 |
|
current += 2 |
|
835
|
|
|
else: |
|
836
|
1 |
|
current += 1 |
|
837
|
1 |
|
primary, secondary = _metaph_add('M') |
|
838
|
1 |
|
continue |
|
839
|
|
|
|
|
840
|
1 |
|
elif _get_at(current) == 'N': |
|
841
|
1 |
|
if _get_at(current + 1) == 'N': |
|
842
|
1 |
|
current += 2 |
|
843
|
|
|
else: |
|
844
|
1 |
|
current += 1 |
|
845
|
1 |
|
primary, secondary = _metaph_add('N') |
|
846
|
1 |
|
continue |
|
847
|
|
|
|
|
848
|
1 |
|
elif _get_at(current) == 'Ñ': |
|
849
|
1 |
|
current += 1 |
|
850
|
1 |
|
primary, secondary = _metaph_add('N') |
|
851
|
1 |
|
continue |
|
852
|
|
|
|
|
853
|
1 |
|
elif _get_at(current) == 'P': |
|
854
|
1 |
|
if _get_at(current + 1) == 'H': |
|
855
|
1 |
|
primary, secondary = _metaph_add('F') |
|
856
|
1 |
|
current += 2 |
|
857
|
1 |
|
continue |
|
858
|
|
|
|
|
859
|
|
|
# also account for "campbell", "raspberry" |
|
860
|
1 |
|
elif _string_at((current + 1), 1, {'P', 'B'}): |
|
861
|
1 |
|
current += 2 |
|
862
|
|
|
else: |
|
863
|
1 |
|
current += 1 |
|
864
|
1 |
|
primary, secondary = _metaph_add('P') |
|
865
|
1 |
|
continue |
|
866
|
|
|
|
|
867
|
1 |
|
elif _get_at(current) == 'Q': |
|
868
|
1 |
|
if _get_at(current + 1) == 'Q': |
|
869
|
1 |
|
current += 2 |
|
870
|
|
|
else: |
|
871
|
1 |
|
current += 1 |
|
872
|
1 |
|
primary, secondary = _metaph_add('K') |
|
873
|
1 |
|
continue |
|
874
|
|
|
|
|
875
|
1 |
|
elif _get_at(current) == 'R': |
|
876
|
|
|
# french e.g. 'rogier', but exclude 'hochmeier' |
|
877
|
1 |
|
if ( |
|
878
|
|
|
(current == last) |
|
|
|
|
|
|
879
|
|
|
and not _slavo_germanic() |
|
|
|
|
|
|
880
|
|
|
and _string_at((current - 2), 2, {'IE'}) |
|
|
|
|
|
|
881
|
|
|
and not _string_at((current - 4), 2, {'ME', 'MA'}) |
|
|
|
|
|
|
882
|
|
|
): |
|
883
|
1 |
|
primary, secondary = _metaph_add('', 'R') |
|
884
|
|
|
else: |
|
885
|
1 |
|
primary, secondary = _metaph_add('R') |
|
886
|
|
|
|
|
887
|
1 |
|
if _get_at(current + 1) == 'R': |
|
888
|
1 |
|
current += 2 |
|
889
|
|
|
else: |
|
890
|
1 |
|
current += 1 |
|
891
|
1 |
|
continue |
|
892
|
|
|
|
|
893
|
1 |
|
elif _get_at(current) == 'S': |
|
894
|
|
|
# special cases 'island', 'isle', 'carlisle', 'carlysle' |
|
895
|
1 |
|
if _string_at((current - 1), 3, {'ISL', 'YSL'}): |
|
896
|
1 |
|
current += 1 |
|
897
|
1 |
|
continue |
|
898
|
|
|
|
|
899
|
|
|
# special case 'sugar-' |
|
900
|
1 |
|
elif (current == 0) and _string_at(current, 5, {'SUGAR'}): |
|
901
|
1 |
|
primary, secondary = _metaph_add('X', 'S') |
|
902
|
1 |
|
current += 1 |
|
903
|
1 |
|
continue |
|
904
|
|
|
|
|
905
|
1 |
|
elif _string_at(current, 2, {'SH'}): |
|
906
|
|
|
# Germanic |
|
907
|
1 |
|
if _string_at( |
|
908
|
|
|
(current + 1), 4, {'HEIM', 'HOEK', 'HOLM', 'HOLZ'} |
|
|
|
|
|
|
909
|
|
|
): |
|
910
|
1 |
|
primary, secondary = _metaph_add('S') |
|
911
|
|
|
else: |
|
912
|
1 |
|
primary, secondary = _metaph_add('X') |
|
913
|
1 |
|
current += 2 |
|
914
|
1 |
|
continue |
|
915
|
|
|
|
|
916
|
|
|
# Italian & Armenian |
|
917
|
1 |
|
elif _string_at(current, 3, {'SIO', 'SIA'}) or _string_at( |
|
918
|
|
|
current, 4, {'SIAN'} |
|
|
|
|
|
|
919
|
|
|
): |
|
920
|
1 |
|
if not _slavo_germanic(): |
|
921
|
1 |
|
primary, secondary = _metaph_add('S', 'X') |
|
922
|
|
|
else: |
|
923
|
1 |
|
primary, secondary = _metaph_add('S') |
|
924
|
1 |
|
current += 3 |
|
925
|
1 |
|
continue |
|
926
|
|
|
|
|
927
|
|
|
# German & anglicisations, e.g. 'smith' match 'schmidt', |
|
928
|
|
|
# 'snider' match 'schneider' |
|
929
|
|
|
# also, -sz- in Slavic language although in Hungarian it is |
|
930
|
|
|
# pronounced 's' |
|
931
|
1 |
|
elif ( |
|
932
|
|
|
(current == 0) |
|
|
|
|
|
|
933
|
|
|
and _string_at((current + 1), 1, {'M', 'N', 'L', 'W'}) |
|
|
|
|
|
|
934
|
|
|
) or _string_at((current + 1), 1, {'Z'}): |
|
935
|
1 |
|
primary, secondary = _metaph_add('S', 'X') |
|
936
|
1 |
|
if _string_at((current + 1), 1, {'Z'}): |
|
937
|
1 |
|
current += 2 |
|
938
|
|
|
else: |
|
939
|
1 |
|
current += 1 |
|
940
|
1 |
|
continue |
|
941
|
|
|
|
|
942
|
1 |
|
elif _string_at(current, 2, {'SC'}): |
|
943
|
|
|
# Schlesinger's rule |
|
944
|
1 |
|
if _get_at(current + 2) == 'H': |
|
945
|
|
|
# dutch origin, e.g. 'school', 'schooner' |
|
946
|
1 |
|
if _string_at( |
|
947
|
|
|
(current + 3), |
|
|
|
|
|
|
948
|
|
|
2, |
|
|
|
|
|
|
949
|
|
|
{'OO', 'ER', 'EN', 'UY', 'ED', 'EM'}, |
|
|
|
|
|
|
950
|
|
|
): |
|
951
|
|
|
# 'schermerhorn', 'schenker' |
|
952
|
1 |
|
if _string_at((current + 3), 2, {'ER', 'EN'}): |
|
953
|
1 |
|
primary, secondary = _metaph_add('X', 'SK') |
|
954
|
|
|
else: |
|
955
|
1 |
|
primary, secondary = _metaph_add('SK') |
|
956
|
1 |
|
current += 3 |
|
957
|
1 |
|
continue |
|
958
|
|
|
else: |
|
959
|
1 |
|
if ( |
|
960
|
|
|
(current == 0) |
|
|
|
|
|
|
961
|
|
|
and not _is_vowel(3) |
|
|
|
|
|
|
962
|
|
|
and (_get_at(3) != 'W') |
|
|
|
|
|
|
963
|
|
|
): |
|
964
|
1 |
|
primary, secondary = _metaph_add('X', 'S') |
|
965
|
|
|
else: |
|
966
|
1 |
|
primary, secondary = _metaph_add('X') |
|
967
|
1 |
|
current += 3 |
|
968
|
1 |
|
continue |
|
969
|
|
|
|
|
970
|
1 |
|
elif _string_at((current + 2), 1, {'I', 'E', 'Y'}): |
|
971
|
1 |
|
primary, secondary = _metaph_add('S') |
|
972
|
1 |
|
current += 3 |
|
973
|
1 |
|
continue |
|
974
|
|
|
|
|
975
|
|
|
# else |
|
976
|
|
|
else: |
|
977
|
1 |
|
primary, secondary = _metaph_add('SK') |
|
978
|
1 |
|
current += 3 |
|
979
|
1 |
|
continue |
|
980
|
|
|
|
|
981
|
|
|
else: |
|
982
|
|
|
# french e.g. 'resnais', 'artois' |
|
983
|
1 |
|
if (current == last) and _string_at( |
|
984
|
|
|
(current - 2), 2, {'AI', 'OI'} |
|
|
|
|
|
|
985
|
|
|
): |
|
986
|
1 |
|
primary, secondary = _metaph_add('', 'S') |
|
987
|
|
|
else: |
|
988
|
1 |
|
primary, secondary = _metaph_add('S') |
|
989
|
|
|
|
|
990
|
1 |
|
if _string_at((current + 1), 1, {'S', 'Z'}): |
|
991
|
1 |
|
current += 2 |
|
992
|
|
|
else: |
|
993
|
1 |
|
current += 1 |
|
994
|
1 |
|
continue |
|
995
|
|
|
|
|
996
|
1 |
|
elif _get_at(current) == 'T': |
|
997
|
1 |
|
if _string_at(current, 4, {'TION'}): |
|
998
|
1 |
|
primary, secondary = _metaph_add('X') |
|
999
|
1 |
|
current += 3 |
|
1000
|
1 |
|
continue |
|
1001
|
|
|
|
|
1002
|
1 |
|
elif _string_at(current, 3, {'TIA', 'TCH'}): |
|
1003
|
1 |
|
primary, secondary = _metaph_add('X') |
|
1004
|
1 |
|
current += 3 |
|
1005
|
1 |
|
continue |
|
1006
|
|
|
|
|
1007
|
1 |
|
elif _string_at(current, 2, {'TH'}) or _string_at( |
|
1008
|
|
|
current, 3, {'TTH'} |
|
|
|
|
|
|
1009
|
|
|
): |
|
1010
|
|
|
# special case 'thomas', 'thames' or germanic |
|
1011
|
1 |
|
if ( |
|
1012
|
|
|
_string_at((current + 2), 2, {'OM', 'AM'}) |
|
|
|
|
|
|
1013
|
|
|
or _string_at(0, 4, {'VAN ', 'VON '}) |
|
|
|
|
|
|
1014
|
|
|
or _string_at(0, 3, {'SCH'}) |
|
|
|
|
|
|
1015
|
|
|
): |
|
1016
|
1 |
|
primary, secondary = _metaph_add('T') |
|
1017
|
|
|
else: |
|
1018
|
1 |
|
primary, secondary = _metaph_add('0', 'T') |
|
1019
|
1 |
|
current += 2 |
|
1020
|
1 |
|
continue |
|
1021
|
|
|
|
|
1022
|
1 |
|
elif _string_at((current + 1), 1, {'T', 'D'}): |
|
1023
|
1 |
|
current += 2 |
|
1024
|
|
|
else: |
|
1025
|
1 |
|
current += 1 |
|
1026
|
1 |
|
primary, secondary = _metaph_add('T') |
|
1027
|
1 |
|
continue |
|
1028
|
|
|
|
|
1029
|
1 |
|
elif _get_at(current) == 'V': |
|
1030
|
1 |
|
if _get_at(current + 1) == 'V': |
|
1031
|
1 |
|
current += 2 |
|
1032
|
|
|
else: |
|
1033
|
1 |
|
current += 1 |
|
1034
|
1 |
|
primary, secondary = _metaph_add('F') |
|
1035
|
1 |
|
continue |
|
1036
|
|
|
|
|
1037
|
1 |
|
elif _get_at(current) == 'W': |
|
1038
|
|
|
# can also be in middle of word |
|
1039
|
1 |
|
if _string_at(current, 2, {'WR'}): |
|
1040
|
1 |
|
primary, secondary = _metaph_add('R') |
|
1041
|
1 |
|
current += 2 |
|
1042
|
1 |
|
continue |
|
1043
|
1 |
|
elif (current == 0) and ( |
|
1044
|
|
|
_is_vowel(current + 1) or _string_at(current, 2, {'WH'}) |
|
|
|
|
|
|
1045
|
|
|
): |
|
1046
|
|
|
# Wasserman should match Vasserman |
|
1047
|
1 |
|
if _is_vowel(current + 1): |
|
1048
|
1 |
|
primary, secondary = _metaph_add('A', 'F') |
|
1049
|
|
|
else: |
|
1050
|
|
|
# need Uomo to match Womo |
|
1051
|
1 |
|
primary, secondary = _metaph_add('A') |
|
1052
|
|
|
|
|
1053
|
|
|
# Arnow should match Arnoff |
|
1054
|
1 |
|
if ( |
|
1055
|
|
|
((current == last) and _is_vowel(current - 1)) |
|
|
|
|
|
|
1056
|
|
|
or _string_at( |
|
|
|
|
|
|
1057
|
|
|
(current - 1), 5, {'EWSKI', 'EWSKY', 'OWSKI', 'OWSKY'} |
|
1058
|
|
|
) |
|
1059
|
|
|
or _string_at(0, 3, ['SCH']) |
|
|
|
|
|
|
1060
|
|
|
): |
|
1061
|
1 |
|
primary, secondary = _metaph_add('', 'F') |
|
1062
|
1 |
|
current += 1 |
|
1063
|
1 |
|
continue |
|
1064
|
|
|
# Polish e.g. 'filipowicz' |
|
1065
|
1 |
|
elif _string_at(current, 4, {'WICZ', 'WITZ'}): |
|
1066
|
1 |
|
primary, secondary = _metaph_add('TS', 'FX') |
|
1067
|
1 |
|
current += 4 |
|
1068
|
1 |
|
continue |
|
1069
|
|
|
# else skip it |
|
1070
|
|
|
else: |
|
1071
|
1 |
|
current += 1 |
|
1072
|
1 |
|
continue |
|
1073
|
|
|
|
|
1074
|
1 |
|
elif _get_at(current) == 'X': |
|
1075
|
|
|
# French e.g. breaux |
|
1076
|
1 |
|
if not ( |
|
1077
|
|
|
(current == last) |
|
|
|
|
|
|
1078
|
|
|
and ( |
|
|
|
|
|
|
1079
|
|
|
_string_at((current - 3), 3, {'IAU', 'EAU'}) |
|
1080
|
|
|
or _string_at((current - 2), 2, {'AU', 'OU'}) |
|
1081
|
|
|
) |
|
1082
|
|
|
): |
|
1083
|
1 |
|
primary, secondary = _metaph_add('KS') |
|
1084
|
|
|
|
|
1085
|
1 |
|
if _string_at((current + 1), 1, {'C', 'X'}): |
|
1086
|
1 |
|
current += 2 |
|
1087
|
|
|
else: |
|
1088
|
1 |
|
current += 1 |
|
1089
|
1 |
|
continue |
|
1090
|
|
|
|
|
1091
|
1 |
|
elif _get_at(current) == 'Z': |
|
1092
|
|
|
# Chinese Pinyin e.g. 'zhao' |
|
1093
|
1 |
|
if _get_at(current + 1) == 'H': |
|
1094
|
1 |
|
primary, secondary = _metaph_add('J') |
|
1095
|
1 |
|
current += 2 |
|
1096
|
1 |
|
continue |
|
1097
|
1 |
|
elif _string_at((current + 1), 2, {'ZO', 'ZI', 'ZA'}) or ( |
|
1098
|
|
|
_slavo_germanic() |
|
|
|
|
|
|
1099
|
|
|
and ((current > 0) and _get_at(current - 1) != 'T') |
|
|
|
|
|
|
1100
|
|
|
): |
|
1101
|
1 |
|
primary, secondary = _metaph_add('S', 'TS') |
|
1102
|
|
|
else: |
|
1103
|
1 |
|
primary, secondary = _metaph_add('S') |
|
1104
|
|
|
|
|
1105
|
1 |
|
if _get_at(current + 1) == 'Z': |
|
1106
|
1 |
|
current += 2 |
|
1107
|
|
|
else: |
|
1108
|
1 |
|
current += 1 |
|
1109
|
1 |
|
continue |
|
1110
|
|
|
|
|
1111
|
|
|
else: |
|
1112
|
1 |
|
current += 1 |
|
1113
|
|
|
|
|
1114
|
1 |
|
if max_length > 0: |
|
1115
|
1 |
|
primary = primary[:max_length] |
|
1116
|
1 |
|
secondary = secondary[:max_length] |
|
1117
|
1 |
|
if primary == secondary: |
|
1118
|
1 |
|
secondary = '' |
|
1119
|
|
|
|
|
1120
|
1 |
|
return primary, secondary |
|
1121
|
|
|
|
|
1122
|
|
|
|
|
1123
|
1 |
|
def double_metaphone(word, max_length=-1): |
|
1124
|
|
|
"""Return the Double Metaphone code for a word. |
|
1125
|
|
|
|
|
1126
|
|
|
Based on Lawrence Philips' (Visual) C++ code from 1999 |
|
1127
|
|
|
:cite:`Philips:2000`. |
|
1128
|
|
|
|
|
1129
|
|
|
:param word: the word to transform |
|
1130
|
|
|
:param max_length: the maximum length of the returned Double Metaphone |
|
1131
|
|
|
codes (defaults to 64, but in Philips' original implementation this |
|
1132
|
|
|
was 4) |
|
1133
|
|
|
:returns: the Double Metaphone value(s) |
|
1134
|
|
|
:rtype: tuple |
|
1135
|
|
|
|
|
1136
|
|
|
>>> double_metaphone('Christopher') |
|
1137
|
|
|
('KRSTFR', '') |
|
1138
|
|
|
>>> double_metaphone('Niall') |
|
1139
|
|
|
('NL', '') |
|
1140
|
|
|
>>> double_metaphone('Smith') |
|
1141
|
|
|
('SM0', 'XMT') |
|
1142
|
|
|
>>> double_metaphone('Schmidt') |
|
1143
|
|
|
('XMT', 'SMT') |
|
1144
|
|
|
""" |
|
1145
|
1 |
|
return DoubleMetaphone().encode(word, max_length) |
|
1146
|
|
|
|
|
1147
|
|
|
|
|
1148
|
|
|
if __name__ == '__main__': |
|
1149
|
|
|
import doctest |
|
1150
|
|
|
|
|
1151
|
|
|
doctest.testmod() |
|
1152
|
|
|
|
chrislit /
abydos
Loading history...