|
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
|
|
|
"""abydos.tests.test_phonetic_hybrid. |
|
20
|
|
|
|
|
21
|
|
|
This module contains unit tests for abydos.phonetic.hybrid |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
|
|
from __future__ import unicode_literals |
|
25
|
|
|
|
|
26
|
|
|
import unittest |
|
27
|
|
|
|
|
28
|
|
|
from abydos.phonetic.hybrid import metasoundex, onca |
|
29
|
|
|
|
|
30
|
|
|
|
|
31
|
|
|
class ONCATestCases(unittest.TestCase): |
|
32
|
|
|
"""Test ONCA functions. |
|
33
|
|
|
|
|
34
|
|
|
test cases for abydos.phonetic.onca |
|
35
|
|
|
""" |
|
36
|
|
|
|
|
37
|
|
|
def test_onca(self): |
|
38
|
|
|
"""Test abydos.phonetic.onca.""" |
|
39
|
|
|
# https://nces.ed.gov/FCSM/pdf/RLT97.pdf |
|
40
|
|
|
self.assertEqual(onca('HALL'), 'H400') |
|
41
|
|
|
self.assertEqual(onca('SMITH'), 'S530') |
|
42
|
|
|
|
|
43
|
|
|
# http://nchod.uhce.ox.ac.uk/NCHOD%20Oxford%20E5%20Report%201st%20Feb_VerAM2.pdf |
|
44
|
|
|
self.assertEqual(onca('HAWTON'), 'H350') |
|
45
|
|
|
self.assertEqual(onca('HORTON'), 'H635') |
|
46
|
|
|
self.assertEqual(onca('HOUGHTON'), 'H235') |
|
47
|
|
|
|
|
48
|
|
|
|
|
49
|
|
|
class MetaSoundexTestCases(unittest.TestCase): |
|
50
|
|
|
"""Test MetaSoundex functions. |
|
51
|
|
|
|
|
52
|
|
|
test cases for abydos.phonetic.metasoundex |
|
53
|
|
|
""" |
|
54
|
|
|
|
|
55
|
|
|
def test_metasoundex(self): |
|
56
|
|
|
"""Test abydos.phonetic.metasoundex.""" |
|
57
|
|
|
# Base cases |
|
58
|
|
|
self.assertEqual(metasoundex(''), '0000') |
|
59
|
|
|
self.assertEqual(metasoundex('', lang='en'), '0000') |
|
60
|
|
|
self.assertEqual(metasoundex('', lang='es'), '') |
|
61
|
|
|
|
|
62
|
|
|
# Top 10 Anglo surnames in US |
|
63
|
|
|
self.assertEqual(metasoundex('Smith', lang='en'), '4500') |
|
64
|
|
|
self.assertEqual(metasoundex('Johnson', lang='en'), '1525') |
|
65
|
|
|
self.assertEqual(metasoundex('Williams', lang='en'), '7452') |
|
66
|
|
|
self.assertEqual(metasoundex('Brown', lang='en'), '7650') |
|
67
|
|
|
self.assertEqual(metasoundex('Jones', lang='en'), '1520') |
|
68
|
|
|
self.assertEqual(metasoundex('Miller', lang='en'), '6460') |
|
69
|
|
|
self.assertEqual(metasoundex('Davis', lang='en'), '3120') |
|
70
|
|
|
self.assertEqual(metasoundex('Wilson', lang='en'), '7425') |
|
71
|
|
|
self.assertEqual(metasoundex('Anderson', lang='en'), '0536') |
|
72
|
|
|
self.assertEqual(metasoundex('Thomas', lang='en'), '6200') |
|
73
|
|
|
|
|
74
|
|
|
self.assertEqual(metasoundex('Smith', lang='es'), '4632') |
|
75
|
|
|
self.assertEqual(metasoundex('Johnson', lang='es'), '82646') |
|
76
|
|
|
self.assertEqual(metasoundex('Williams', lang='es'), '564') |
|
77
|
|
|
self.assertEqual(metasoundex('Brown', lang='es'), '196') |
|
78
|
|
|
self.assertEqual(metasoundex('Jones', lang='es'), '864') |
|
79
|
|
|
self.assertEqual(metasoundex('Miller', lang='es'), '659') |
|
80
|
|
|
self.assertEqual(metasoundex('Davis', lang='es'), '314') |
|
81
|
|
|
self.assertEqual(metasoundex('Wilson', lang='es'), '546') |
|
82
|
|
|
self.assertEqual(metasoundex('Anderson', lang='es'), '63946') |
|
83
|
|
|
self.assertEqual(metasoundex('Thomas', lang='es'), '364') |
|
84
|
|
|
|
|
85
|
|
|
# Top 10 Mexican surnames |
|
86
|
|
|
self.assertEqual(metasoundex('Hernández', lang='en'), '5653') |
|
87
|
|
|
self.assertEqual(metasoundex('García', lang='en'), '5620') |
|
88
|
|
|
self.assertEqual(metasoundex('Lòpez', lang='en'), '8120') |
|
89
|
|
|
self.assertEqual(metasoundex('Martìnez', lang='en'), '6635') |
|
90
|
|
|
self.assertEqual(metasoundex('Rodrìguez', lang='en'), '9362') |
|
91
|
|
|
self.assertEqual(metasoundex('González', lang='en'), '5524') |
|
92
|
|
|
self.assertEqual(metasoundex('Pérez', lang='en'), '7620') |
|
93
|
|
|
self.assertEqual(metasoundex('Sánchez', lang='en'), '4520') |
|
94
|
|
|
self.assertEqual(metasoundex('Gómez', lang='en'), '5520') |
|
95
|
|
|
self.assertEqual(metasoundex('Flores', lang='en'), '7462') |
|
96
|
|
|
|
|
97
|
|
|
self.assertEqual(metasoundex('Hernández', lang='es'), '96634') |
|
98
|
|
|
self.assertEqual(metasoundex('García', lang='es'), '894') |
|
99
|
|
|
self.assertEqual(metasoundex('Lòpez', lang='es'), '504') |
|
100
|
|
|
self.assertEqual(metasoundex('Martìnez', lang='es'), '69364') |
|
101
|
|
|
self.assertEqual(metasoundex('Rodrìguez', lang='es'), '93984') |
|
102
|
|
|
self.assertEqual(metasoundex('González', lang='es'), '86454') |
|
103
|
|
|
self.assertEqual(metasoundex('Pérez', lang='es'), '094') |
|
104
|
|
|
self.assertEqual(metasoundex('Sánchez', lang='es'), '4644') |
|
105
|
|
|
self.assertEqual(metasoundex('Gómez', lang='es'), '864') |
|
106
|
|
|
self.assertEqual(metasoundex('Flores', lang='es'), '2594') |
|
107
|
|
|
|
|
108
|
|
|
|
|
109
|
|
|
if __name__ == '__main__': |
|
110
|
|
|
unittest.main() |
|
111
|
|
|
|