|
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_fingerprint_synoname. |
|
20
|
|
|
|
|
21
|
|
|
This module contains unit tests for abydos.fingerprint.synoname |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
|
|
from __future__ import unicode_literals |
|
25
|
|
|
|
|
26
|
|
|
import unittest |
|
27
|
|
|
|
|
28
|
|
|
from abydos.fingerprint.synoname import synoname_toolcode |
|
29
|
|
|
|
|
30
|
|
|
|
|
31
|
|
|
class SynonameToolcodeTestCases(unittest.TestCase): |
|
32
|
|
|
"""Test Synoname Toolcode function. |
|
33
|
|
|
|
|
34
|
|
|
abydos.fingerprint.synoname.synoname_toolcode |
|
35
|
|
|
""" |
|
36
|
|
|
|
|
37
|
|
|
def test_synoname_toolcode(self): |
|
38
|
|
|
"""Test abydos.fingerprint.synoname.synoname_toolcode.""" |
|
39
|
|
|
# Base case |
|
40
|
|
|
self.assertEqual(synoname_toolcode(''), ('', '', '0000000000$$')) |
|
41
|
|
|
|
|
42
|
|
|
# from Synoname demo |
|
43
|
|
|
self.assertEqual(synoname_toolcode('angelico', 'fra'), |
|
44
|
|
|
('angelico', 'fra', '0000000308$044a$af')) |
|
45
|
|
|
self.assertEqual(synoname_toolcode('Aelst', 'Willem van', ''), |
|
46
|
|
|
('aelst', 'willem van', '0000001005$143a$awv')) |
|
47
|
|
|
self.assertEqual(synoname_toolcode('Afro'), |
|
48
|
|
|
('afro', '', '0000000004$$a')) |
|
49
|
|
|
self.assertEqual(synoname_toolcode('Afro', 'Basaldella'), |
|
50
|
|
|
('afro', 'basaldella', '0000001004$$ab')) |
|
51
|
|
|
self.assertEqual(synoname_toolcode('Albright', 'Ivan'), |
|
52
|
|
|
('albright', 'ivan', '0000000408$$ai')) |
|
53
|
|
|
self.assertEqual(synoname_toolcode('Antonello da Messina'), |
|
54
|
|
|
('antonello da messina', '', '0000000020$022b$adm')) |
|
55
|
|
|
self.assertEqual(synoname_toolcode('Albright', 'Ivan Le Lorraine'), |
|
56
|
|
|
('albright', 'ivan le lorraine', |
|
57
|
|
|
'0000001608$067b$ail')) |
|
58
|
|
|
self.assertEqual(synoname_toolcode('Bazille', 'Frederic', |
|
59
|
|
|
'Attributed to'), |
|
60
|
|
|
('bazille', 'frederic', '1000000807$$bf')) |
|
61
|
|
|
self.assertEqual(synoname_toolcode('Bazille', 'Frederick', |
|
62
|
|
|
'Attributed to'), |
|
63
|
|
|
('bazille', 'frederick', '1000000907$$bf')) |
|
64
|
|
|
self.assertEqual(synoname_toolcode('Beerstraaten', 'Jan Abrahamsz.'), |
|
65
|
|
|
('beerstraaten', 'jan abrahamsz.', '0200001412$$bja')) |
|
66
|
|
|
self.assertEqual(synoname_toolcode('Bonifacio di Pitati'), |
|
67
|
|
|
('bonifacio di pitati', '', '0000000019$035b$bdp')) |
|
68
|
|
|
self.assertEqual(synoname_toolcode('Breughel the Younger', 'Jan'), |
|
69
|
|
|
('breughel the younger', 'jan', |
|
70
|
|
|
'0020000320$134b$btyj')) |
|
71
|
|
|
self.assertEqual(synoname_toolcode('Brown', 'W. W.'), |
|
72
|
|
|
('brown', 'w. w.', '0200000505$$bw')) |
|
73
|
|
|
self.assertEqual(synoname_toolcode('Brueghel II (the Younger)', 'Jan'), |
|
74
|
|
|
('brueghel ii (the younger)', 'jan', |
|
75
|
|
|
'0120490325$049b134b$bityj')) |
|
76
|
|
|
self.assertEqual(synoname_toolcode('Brueghel II (the Younger)', |
|
77
|
|
|
'Pieter', 'Workshop of'), |
|
78
|
|
|
('brueghel ii (the younger)', 'pieter', |
|
79
|
|
|
'3120490625$049b134b$bityp')) |
|
80
|
|
|
self.assertEqual(synoname_toolcode('Bugiardini', |
|
81
|
|
|
'Guiliano di Piero di Simone'), |
|
82
|
|
|
('bugiardini', 'guiliano di piero di simone', |
|
83
|
|
|
'0000002710$035b035b$bgdps')) |
|
84
|
|
|
self.assertEqual(synoname_toolcode('Caravaggio', '', 'Follower of'), |
|
85
|
|
|
('caravaggio', '', '3000000010$$c')) |
|
86
|
|
|
self.assertEqual(synoname_toolcode('Caravaggio', |
|
87
|
|
|
'Michelangelo Merisi da', |
|
88
|
|
|
'Follower of'), |
|
89
|
|
|
('caravaggio', 'michelangelo merisi da', |
|
90
|
|
|
'3000002210$022a$cmd')) |
|
91
|
|
|
self.assertEqual(synoname_toolcode('Oost the Younger', 'Jacob van'), |
|
92
|
|
|
('oost the younger', 'jacob van', |
|
93
|
|
|
'0020000916$134b143a$otyjv')) |
|
94
|
|
|
|
|
95
|
|
|
# additional tests for coverage |
|
96
|
|
|
self.assertEqual(synoname_toolcode('Cato the Elder', '', 'Copy of'), |
|
97
|
|
|
('cato the elder', '', '2010000014$133b$cte')) |
|
98
|
|
|
self.assertEqual(synoname_toolcode('Cato, the Elder', normalize=2), |
|
99
|
|
|
('cato the elder', '', '0110000014$133b$cte')) |
|
100
|
|
|
self.assertEqual(synoname_toolcode('Cato the Elder', normalize=2), |
|
101
|
|
|
('cato the elder', '', '0010000014$133b$cte')) |
|
102
|
|
|
self.assertEqual(synoname_toolcode('Lorem ipsum dolor sit amet, ' + |
|
103
|
|
|
'consectetur adipiscing elit, ' + |
|
104
|
|
|
'sed do eiusmod tempor ' + |
|
105
|
|
|
'incididunt ut labore et dolore ' + |
|
106
|
|
|
'magna aliqua. Nulla aliquet ' + |
|
107
|
|
|
'porttitor lacus luctus accumsan ' + |
|
108
|
|
|
'tortor posuere. Egestas purus ' + |
|
109
|
|
|
'viverra accumsan in. Ultrices ' + |
|
110
|
|
|
'mi tempus imperdiet nulla ' + |
|
111
|
|
|
'malesuada pellentesque elit ' + |
|
112
|
|
|
'eget gravida. Proin libero nunc ' + |
|
113
|
|
|
'consequat interdum varius sit ' + |
|
114
|
|
|
'amet mattis vulputate. Mauris ' + |
|
115
|
|
|
'ultrices eros in cursus turpis ' + |
|
116
|
|
|
'massa tincidunt dui. Faucibus ' + |
|
117
|
|
|
'in ornare quam viverra orci ' + |
|
118
|
|
|
'sagittis eu volutpat odio. Enim ' + |
|
119
|
|
|
'blandit volutpat maecenas ' + |
|
120
|
|
|
'volutpat blandit aliquam etiam. ' + |
|
121
|
|
|
'Vel quam elementum pulvinar ' + |
|
122
|
|
|
'etiam. Duis ut diam quam nulla ' + |
|
123
|
|
|
'porttitor massa id.', |
|
124
|
|
|
normalize=1)[2], |
|
125
|
|
|
'02000060626$068d$lidsacetumnpvgflo') |
|
126
|
|
|
self.assertEqual(synoname_toolcode('Sainte-Vincent'), |
|
127
|
|
|
('sainte-vincent', '', '0100000014$110c$sv')) |
|
128
|
|
|
self.assertEqual(synoname_toolcode('Lorem', 'Sainte-Vincent'), |
|
129
|
|
|
('lorem', 'sainte-vincent', |
|
130
|
|
|
'0100001405$068d110b$lsvlo')) |
|
131
|
|
|
self.assertEqual(synoname_toolcode('Louis II', 'Jean'), |
|
132
|
|
|
('louis ii', 'jean', '0000490408$049b068d$lijlo')) |
|
133
|
|
|
self.assertEqual(synoname_toolcode('Louis', 'Jean II', normalize=2), |
|
134
|
|
|
('louis ii', 'jean', '0000490705$049a068d$ljilo')) |
|
135
|
|
|
self.assertEqual(synoname_toolcode('Louis', 'Jean II ', normalize=2), |
|
136
|
|
|
('louis ii', 'jean', '0000490805$049b068d$ljilo')) |
|
137
|
|
|
self.assertEqual(synoname_toolcode('Louis', 'Jean II-', normalize=2), |
|
138
|
|
|
('louis', 'jean ii-', '0100490805$049b068d$ljilo')) |
|
139
|
|
|
self.assertEqual(synoname_toolcode('Louis V.', 'Jean', normalize=2), |
|
140
|
|
|
('louis v.', 'jean', '0200000408$068d$lvjlo')) |
|
141
|
|
|
self.assertEqual(synoname_toolcode('Louis V.', 'Ste.-Jean Ste.', |
|
142
|
|
|
normalize=2), |
|
143
|
|
|
('louis v.', 'ste.-jean ste.', |
|
144
|
|
|
'0200001408$068d127b127X$lvsjlo ste')) |
|
145
|
|
|
self.assertEqual(synoname_toolcode('Louis IX', 'Jean III II', |
|
146
|
|
|
normalize=2), |
|
147
|
|
|
('louis ix iii ii', 'jean', |
|
148
|
|
|
'0000481108$048b049a056b068d$lijlo')) |
|
149
|
|
|
self.assertEqual(synoname_toolcode('Louis IX', 'Jean II III', |
|
150
|
|
|
normalize=2), |
|
151
|
|
|
('louis ix iii ii', 'jean', |
|
152
|
|
|
'0000481108$048a049a056b068d$lijlo')) |
|
153
|
|
|
self.assertEqual(synoname_toolcode('Louis IX', 'Jean II III', |
|
154
|
|
|
normalize=1), |
|
155
|
|
|
('louis ix', 'jean ii iii', |
|
156
|
|
|
'0000481108$048a049a056b068d$lijlo')) |
|
157
|
|
|
self.assertEqual(synoname_toolcode('Lorem', 'Sainte-Sainte-Vincent'), |
|
158
|
|
|
('lorem', 'sainte-sainte-vincent', |
|
159
|
|
|
'0100002105$068d110b$lsvlo')) |
|
160
|
|
|
self.assertEqual(synoname_toolcode('Brueghel II', 'I. Jan', |
|
161
|
|
|
normalize=2), |
|
162
|
|
|
('brueghel ii', 'i. jan', '0200000611$$bij')) |
|
163
|
|
|
self.assertEqual(synoname_toolcode('Brueghel', 'I. Jan II', |
|
164
|
|
|
normalize=2), |
|
165
|
|
|
('brueghel', 'i. jan ii', '0200000908$$bij')) |
|
166
|
|
|
self.assertEqual(synoname_toolcode('Lorem', 'Laurent Ormond'), |
|
167
|
|
|
('lorem', 'laurent ormond', '0000001405$068d$lo')) |
|
168
|
|
|
|
|
169
|
|
|
|
|
170
|
|
|
if __name__ == '__main__': |
|
171
|
|
|
unittest.main() |
|
172
|
|
|
|
chrislit /
abydos
