|
1
|
|
|
# Copyright 2014-2020 by Christopher C. Little. |
|
2
|
|
|
# This file is part of Abydos. |
|
3
|
|
|
# |
|
4
|
|
|
# Abydos is free software: you can redistribute it and/or modify |
|
5
|
|
|
# it under the terms of the GNU General Public License as published by |
|
6
|
|
|
# the Free Software Foundation, either version 3 of the License, or |
|
7
|
|
|
# (at your option) any later version. |
|
8
|
|
|
# |
|
9
|
|
|
# Abydos is distributed in the hope that it will be useful, |
|
10
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
11
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
12
|
|
|
# GNU General Public License for more details. |
|
13
|
|
|
# |
|
14
|
|
|
# You should have received a copy of the GNU General Public License |
|
15
|
|
|
# along with Abydos. If not, see <http://www.gnu.org/licenses/>. |
|
16
|
|
|
|
|
17
|
|
|
"""abydos.stemmer._snowball_dutch. |
|
18
|
|
|
|
|
19
|
1 |
|
Snowball Dutch stemmer |
|
20
|
|
|
""" |
|
21
|
|
|
|
|
22
|
|
|
from unicodedata import normalize |
|
23
|
|
|
|
|
24
|
1 |
|
from ._snowball import _Snowball |
|
25
|
|
|
|
|
26
|
|
|
__all__ = ['SnowballDutch'] |
|
27
|
|
|
|
|
28
|
|
|
|
|
29
|
|
|
class SnowballDutch(_Snowball): |
|
30
|
|
|
"""Snowball Dutch stemmer. |
|
31
|
1 |
|
|
|
32
|
|
|
The Snowball Dutch stemmer is defined at: |
|
33
|
1 |
|
http://snowball.tartarus.org/algorithms/dutch/stemmer.html |
|
34
|
|
|
|
|
35
|
1 |
|
.. versionadded:: 0.3.6 |
|
36
|
1 |
|
""" |
|
37
|
|
|
|
|
38
|
1 |
|
_vowels = {'a', 'e', 'i', 'o', 'u', 'y', 'è'} |
|
39
|
1 |
|
_not_s_endings = {'a', 'e', 'i', 'j', 'o', 'u', 'y', 'è'} |
|
40
|
|
|
_accented = dict(zip((ord(_) for _ in 'äëïöüáéíóú'), 'aeiouaeiou')) |
|
|
|
|
|
|
41
|
1 |
|
|
|
42
|
|
|
def _undouble(self, word: str) -> str: |
|
43
|
|
|
"""Undouble endings -kk, -dd, and -tt. |
|
44
|
1 |
|
|
|
45
|
|
|
Parameters |
|
46
|
|
|
---------- |
|
47
|
|
|
word : str |
|
48
|
|
|
The word to stem |
|
49
|
|
|
|
|
50
|
|
|
Returns |
|
51
|
|
|
------- |
|
52
|
|
|
str |
|
53
|
1 |
|
The word with doubled endings undoubled |
|
54
|
1 |
|
|
|
55
|
1 |
|
|
|
56
|
|
|
.. versionadded:: 0.1.0 |
|
57
|
1 |
|
.. versionchanged:: 0.3.6 |
|
58
|
|
|
Encapsulated in class |
|
59
|
|
|
|
|
60
|
|
|
""" |
|
61
|
|
|
if ( |
|
62
|
|
|
len(word) > 1 |
|
63
|
|
|
and word[-1] == word[-2] |
|
64
|
|
|
and word[-1] in {'d', 'k', 't'} |
|
65
|
|
|
): |
|
66
|
|
|
return word[:-1] |
|
67
|
|
|
return word |
|
68
|
|
|
|
|
69
|
|
|
def stem(self, word: str) -> str: |
|
70
|
|
|
"""Return Snowball Dutch stem. |
|
71
|
|
|
|
|
72
|
|
|
Parameters |
|
73
|
|
|
---------- |
|
74
|
|
|
word : str |
|
75
|
|
|
The word to stem |
|
76
|
1 |
|
|
|
77
|
|
|
Returns |
|
78
|
|
|
------- |
|
79
|
|
|
str |
|
80
|
|
|
Word stem |
|
81
|
1 |
|
|
|
82
|
1 |
|
Examples |
|
83
|
|
|
-------- |
|
84
|
1 |
|
>>> stmr = SnowballDutch() |
|
85
|
|
|
>>> stmr.stem('lezen') |
|
86
|
|
|
'lez' |
|
87
|
|
|
>>> stmr.stem('opschorting') |
|
88
|
|
|
'opschort' |
|
89
|
|
|
>>> stmr.stem('ongrijpbaarheid') |
|
90
|
|
|
'ongrijp' |
|
91
|
|
|
|
|
92
|
|
|
|
|
93
|
|
|
.. versionadded:: 0.1.0 |
|
94
|
|
|
.. versionchanged:: 0.3.6 |
|
95
|
|
|
Encapsulated in class |
|
96
|
|
|
|
|
97
|
|
|
""" |
|
98
|
|
|
# lowercase, normalize, decompose, filter umlauts & acutes out, and |
|
99
|
|
|
# compose |
|
100
|
|
|
word = normalize('NFC', word.lower()) |
|
101
|
|
|
word = word.translate(self._accented) |
|
102
|
|
|
|
|
103
|
|
|
for i in range(len(word)): |
|
104
|
|
|
if i == 0 and word[0] == 'y': |
|
105
|
|
|
word = 'Y' + word[1:] |
|
106
|
|
|
elif word[i] == 'y' and word[i - 1] in self._vowels: |
|
107
|
|
|
word = word[:i] + 'Y' + word[i + 1 :] |
|
108
|
|
|
elif ( |
|
109
|
|
|
word[i] == 'i' |
|
110
|
|
|
and word[i - 1] in self._vowels |
|
111
|
|
|
and i + 1 < len(word) |
|
112
|
|
|
and word[i + 1] in self._vowels |
|
113
|
|
|
): |
|
114
|
|
|
word = word[:i] + 'I' + word[i + 1 :] |
|
115
|
1 |
|
|
|
116
|
1 |
|
r1_start = max(3, self._sb_r1(word)) |
|
117
|
|
|
r2_start = self._sb_r2(word) |
|
118
|
1 |
|
|
|
119
|
1 |
|
# Step 1 |
|
120
|
1 |
|
if word[-5:] == 'heden': |
|
121
|
1 |
|
if len(word[r1_start:]) >= 5: |
|
122
|
1 |
|
word = word[:-3] + 'id' |
|
123
|
1 |
|
elif word[-3:] == 'ene': |
|
124
|
|
|
if len(word[r1_start:]) >= 3 and ( |
|
125
|
|
|
word[-4] not in self._vowels and word[-6:-3] != 'gem' |
|
126
|
|
|
): |
|
127
|
|
|
word = self._undouble(word[:-3]) |
|
128
|
|
|
elif word[-2:] == 'en': |
|
129
|
1 |
|
if len(word[r1_start:]) >= 2 and ( |
|
130
|
|
|
word[-3] not in self._vowels and word[-5:-2] != 'gem' |
|
131
|
1 |
|
): |
|
132
|
1 |
|
word = self._undouble(word[:-2]) |
|
133
|
|
|
elif word[-2:] == 'se': |
|
134
|
|
|
if ( |
|
135
|
1 |
|
len(word[r1_start:]) >= 2 |
|
136
|
1 |
|
and word[-3] not in self._not_s_endings |
|
137
|
1 |
|
): |
|
138
|
1 |
|
word = word[:-2] |
|
139
|
1 |
|
elif word[-1:] == 's': |
|
140
|
|
|
if ( |
|
141
|
|
|
len(word[r1_start:]) >= 1 |
|
142
|
1 |
|
and word[-2] not in self._not_s_endings |
|
143
|
1 |
|
): |
|
144
|
1 |
|
word = word[:-1] |
|
145
|
|
|
|
|
146
|
|
|
# Step 2 |
|
147
|
1 |
|
e_removed = False |
|
148
|
1 |
|
if word[-1:] == 'e': |
|
149
|
1 |
|
if len(word[r1_start:]) >= 1 and word[-2] not in self._vowels: |
|
150
|
|
|
word = self._undouble(word[:-1]) |
|
151
|
|
|
e_removed = True |
|
152
|
|
|
|
|
153
|
1 |
|
# Step 3a |
|
154
|
1 |
|
if word[-4:] == 'heid': |
|
155
|
1 |
|
if len(word[r2_start:]) >= 4 and word[-5] != 'c': |
|
156
|
|
|
word = word[:-4] |
|
157
|
|
|
if word[-2:] == 'en': |
|
158
|
|
|
if len(word[r1_start:]) >= 2 and ( |
|
159
|
1 |
|
word[-3] not in self._vowels and word[-5:-2] != 'gem' |
|
160
|
|
|
): |
|
161
|
|
|
word = self._undouble(word[:-2]) |
|
162
|
1 |
|
|
|
163
|
1 |
|
# Step 3b |
|
164
|
1 |
|
if word[-4:] == 'lijk': |
|
165
|
1 |
|
if len(word[r2_start:]) >= 4: |
|
166
|
1 |
|
word = word[:-4] |
|
167
|
|
|
# Repeat step 2 |
|
168
|
|
|
if word[-1:] == 'e': |
|
169
|
1 |
|
if ( |
|
170
|
1 |
|
len(word[r1_start:]) >= 1 |
|
171
|
1 |
|
and word[-2] not in self._vowels |
|
172
|
1 |
|
): |
|
173
|
1 |
|
word = self._undouble(word[:-1]) |
|
174
|
|
|
elif word[-4:] == 'baar': |
|
175
|
|
|
if len(word[r2_start:]) >= 4: |
|
176
|
1 |
|
word = word[:-4] |
|
177
|
|
|
elif word[-3:] in ('end', 'ing'): |
|
178
|
|
|
if len(word[r2_start:]) >= 3: |
|
179
|
1 |
|
word = word[:-3] |
|
180
|
1 |
|
if ( |
|
181
|
1 |
|
word[-2:] == 'ig' |
|
182
|
|
|
and len(word[r2_start:]) >= 2 |
|
183
|
1 |
|
and word[-3] != 'e' |
|
184
|
1 |
|
): |
|
185
|
|
|
word = word[:-2] |
|
186
|
|
|
else: |
|
187
|
|
|
word = self._undouble(word) |
|
188
|
1 |
|
elif word[-3:] == 'bar': |
|
189
|
1 |
|
if len(word[r2_start:]) >= 3 and e_removed: |
|
190
|
1 |
|
word = word[:-3] |
|
191
|
1 |
|
elif word[-2:] == 'ig': |
|
192
|
1 |
|
if len(word[r2_start:]) >= 2 and word[-3] != 'e': |
|
193
|
1 |
|
word = word[:-2] |
|
194
|
1 |
|
|
|
195
|
1 |
|
# Step 4 |
|
196
|
|
|
if ( |
|
197
|
|
|
len(word) >= 4 |
|
198
|
|
|
and word[-3] == word[-2] |
|
199
|
|
|
and word[-2] in {'a', 'e', 'o', 'u'} |
|
200
|
1 |
|
and word[-4] not in self._vowels |
|
201
|
|
|
and word[-1] not in self._vowels |
|
202
|
1 |
|
and word[-1] != 'I' |
|
203
|
1 |
|
): |
|
204
|
1 |
|
word = word[:-2] + word[-1] |
|
205
|
1 |
|
|
|
206
|
1 |
|
# Change 'Y' and 'U' back to lowercase if survived stemming |
|
207
|
1 |
|
for i in range(0, len(word)): |
|
208
|
1 |
|
if word[i] == 'Y': |
|
209
|
|
|
word = word[:i] + 'y' + word[i + 1 :] |
|
210
|
|
|
elif word[i] == 'I': |
|
211
|
1 |
|
word = word[:i] + 'i' + word[i + 1 :] |
|
212
|
|
|
|
|
213
|
|
|
return word |
|
214
|
|
|
|
|
215
|
|
|
|
|
216
|
|
|
if __name__ == '__main__': |
|
217
|
|
|
import doctest |
|
218
|
|
|
|
|
219
|
|
|
doctest.testmod() |
|
220
|
|
|
|
chrislit /
abydos
Loading history...