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