Coverage for documents/_snowball_stopwords.py: 100%
2 statements
« prev ^ index » next coverage.py v7.15.2, created at 2026-10-10 09:07 +0000
« prev ^ index » next coverage.py v7.15.2, created at 2026-10-10 09:07 +0000
1"""
2English stop words from the Snowball project, kept as published at
3https://snowballstem.org/algorithms/english/stop.txt, under this license:
5Copyright (c) 2001, Dr Martin Porter
6Copyright (c) 2004,2005, Richard Boulton
7Copyright (c) 2013, Yoshiki Shibukawa
8Copyright (c) 2006-2025, Olly Betts
9All rights reserved.
11Redistribution and use in source and binary forms, with or without
12modification, are permitted provided that the following conditions
13are met:
15 1. Redistributions of source code must retain the above copyright notice,
16 this list of conditions and the following disclaimer.
17 2. Redistributions in binary form must reproduce the above copyright notice,
18 this list of conditions and the following disclaimer in the documentation
19 and/or other materials provided with the distribution.
20 3. Neither the name of the Snowball project nor the names of its contributors
21 may be used to endorse or promote products derived from this software
22 without specific prior written permission.
24THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
25ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
26WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
27DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
28ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
29(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
30LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
31ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
32(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
33SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
34"""
36from typing import Final
38ENGLISH: Final[tuple[str, ...]] = (
39 "i",
40 "me",
41 "my",
42 "myself",
43 "we",
44 "our",
45 "ours",
46 "ourselves",
47 "you",
48 "your",
49 "yours",
50 "yourself",
51 "yourselves",
52 "he",
53 "him",
54 "his",
55 "himself",
56 "she",
57 "her",
58 "hers",
59 "herself",
60 "it",
61 "its",
62 "itself",
63 "they",
64 "them",
65 "their",
66 "theirs",
67 "themselves",
68 "what",
69 "which",
70 "who",
71 "whom",
72 "this",
73 "that",
74 "these",
75 "those",
76 "am",
77 "is",
78 "are",
79 "was",
80 "were",
81 "be",
82 "been",
83 "being",
84 "have",
85 "has",
86 "had",
87 "having",
88 "do",
89 "does",
90 "did",
91 "doing",
92 "would",
93 "should",
94 "could",
95 "ought",
96 "i'm",
97 "you're",
98 "he's",
99 "she's",
100 "it's",
101 "we're",
102 "they're",
103 "i've",
104 "you've",
105 "we've",
106 "they've",
107 "i'd",
108 "you'd",
109 "he'd",
110 "she'd",
111 "we'd",
112 "they'd",
113 "i'll",
114 "you'll",
115 "he'll",
116 "she'll",
117 "we'll",
118 "they'll",
119 "isn't",
120 "aren't",
121 "wasn't",
122 "weren't",
123 "hasn't",
124 "haven't",
125 "hadn't",
126 "doesn't",
127 "don't",
128 "didn't",
129 "won't",
130 "wouldn't",
131 "shan't",
132 "shouldn't",
133 "can't",
134 "cannot",
135 "couldn't",
136 "mustn't",
137 "let's",
138 "that's",
139 "who's",
140 "what's",
141 "here's",
142 "there's",
143 "when's",
144 "where's",
145 "why's",
146 "how's",
147 "a",
148 "an",
149 "the",
150 "and",
151 "but",
152 "if",
153 "or",
154 "because",
155 "as",
156 "until",
157 "while",
158 "of",
159 "at",
160 "by",
161 "for",
162 "with",
163 "about",
164 "against",
165 "between",
166 "into",
167 "through",
168 "during",
169 "before",
170 "after",
171 "above",
172 "below",
173 "to",
174 "from",
175 "up",
176 "down",
177 "in",
178 "out",
179 "on",
180 "off",
181 "over",
182 "under",
183 "again",
184 "further",
185 "then",
186 "once",
187 "here",
188 "there",
189 "when",
190 "where",
191 "why",
192 "how",
193 "all",
194 "any",
195 "both",
196 "each",
197 "few",
198 "more",
199 "most",
200 "other",
201 "some",
202 "such",
203 "no",
204 "nor",
205 "not",
206 "only",
207 "own",
208 "same",
209 "so",
210 "than",
211 "too",
212 "very",
213)