1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
|
// Copyright (C) 2022 Google LLC
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#include <string_view>
#include <vector>
#include "gmock/gmock.h"
#include "gtest/gtest.h"
#include "icing/portable/platform.h"
#include "icing/proto/schema.pb.h"
#include "icing/testing/common-matchers.h"
#include "icing/testing/icu-data-file-helper.h"
#include "icing/testing/jni-test-helpers.h"
#include "icing/testing/test-data.h"
#include "icing/tokenization/language-segmenter-factory.h"
#include "icing/tokenization/language-segmenter.h"
#include "icing/tokenization/tokenizer-factory.h"
#include "icing/tokenization/tokenizer.h"
#include "unicode/uloc.h"
namespace icing {
namespace lib {
namespace {
using ::testing::ElementsAre;
// This test exists to ensure that the different tokenizers treat different
// segments of text in the same manner.
class CombinedTokenizerTest : public ::testing::Test {
protected:
void SetUp() override {
if (!IsCfStringTokenization() && !IsReverseJniTokenization()) {
ICING_ASSERT_OK(
// File generated via icu_data_file rule in //icing/BUILD.
icu_data_file_helper::SetUpICUDataFile(
GetTestFilePath("icing/icu.dat")));
}
jni_cache_ = GetTestJniCache();
language_segmenter_factory::SegmenterOptions options(ULOC_US,
jni_cache_.get());
ICING_ASSERT_OK_AND_ASSIGN(
lang_segmenter_,
language_segmenter_factory::Create(std::move(options)));
}
std::unique_ptr<const JniCache> jni_cache_;
std::unique_ptr<LanguageSegmenter> lang_segmenter_;
};
std::vector<std::string> GetTokenTerms(const std::vector<Token>& tokens) {
std::vector<std::string> terms;
terms.reserve(tokens.size());
for (const Token& token : tokens) {
if (token.type == Token::Type::REGULAR) {
terms.push_back(std::string(token.text));
}
}
return terms;
}
} // namespace
TEST_F(CombinedTokenizerTest, SpecialCharacters) {
const std::string_view kText = "😊 Hello! Goodbye?";
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> indexing_tokenizer,
tokenizer_factory::CreateIndexingTokenizer(
StringIndexingConfig::TokenizerType::PLAIN, lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> query_tokenizer,
CreateQueryTokenizer(tokenizer_factory::QueryTokenizerType::RAW_QUERY,
lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> indexing_tokens,
indexing_tokenizer->TokenizeAll(kText));
std::vector<std::string> indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("😊", "Hello", "Goodbye"));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> query_tokens,
query_tokenizer->TokenizeAll(kText));
std::vector<std::string> query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("😊", "Hello", "Goodbye"));
}
TEST_F(CombinedTokenizerTest, Parentheses) {
const std::string_view kText = "((paren1)(paren2) (last paren))";
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> indexing_tokenizer,
tokenizer_factory::CreateIndexingTokenizer(
StringIndexingConfig::TokenizerType::PLAIN, lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> query_tokenizer,
CreateQueryTokenizer(tokenizer_factory::QueryTokenizerType::RAW_QUERY,
lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> indexing_tokens,
indexing_tokenizer->TokenizeAll(kText));
std::vector<std::string> indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("paren1", "paren2", "last", "paren"));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> query_tokens,
query_tokenizer->TokenizeAll(kText));
std::vector<std::string> query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("paren1", "paren2", "last", "paren"));
}
TEST_F(CombinedTokenizerTest, Negation) {
const std::string_view kText = "-foo -bar -baz";
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> indexing_tokenizer,
tokenizer_factory::CreateIndexingTokenizer(
StringIndexingConfig::TokenizerType::PLAIN, lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> query_tokenizer,
CreateQueryTokenizer(tokenizer_factory::QueryTokenizerType::RAW_QUERY,
lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> indexing_tokens,
indexing_tokenizer->TokenizeAll(kText));
std::vector<std::string> indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("foo", "bar", "baz"));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> query_tokens,
query_tokenizer->TokenizeAll(kText));
std::vector<std::string> query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("foo", "bar", "baz"));
}
// TODO(b/254874614): Handle colon word breaks in ICU 72+
TEST_F(CombinedTokenizerTest, Colons) {
const std::string_view kText = ":foo: :bar baz:";
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> indexing_tokenizer,
tokenizer_factory::CreateIndexingTokenizer(
StringIndexingConfig::TokenizerType::PLAIN, lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> query_tokenizer,
CreateQueryTokenizer(tokenizer_factory::QueryTokenizerType::RAW_QUERY,
lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> indexing_tokens,
indexing_tokenizer->TokenizeAll(kText));
std::vector<std::string> indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("foo", "bar", "baz"));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> query_tokens,
query_tokenizer->TokenizeAll(kText));
std::vector<std::string> query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("foo", "bar", "baz"));
}
// TODO(b/254874614): Handle colon word breaks in ICU 72+
TEST_F(CombinedTokenizerTest, ColonsPropertyRestricts) {
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> indexing_tokenizer,
tokenizer_factory::CreateIndexingTokenizer(
StringIndexingConfig::TokenizerType::PLAIN, lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> query_tokenizer,
CreateQueryTokenizer(tokenizer_factory::QueryTokenizerType::RAW_QUERY,
lang_segmenter_.get()));
if (GetIcuTokenizationVersion() >= 72) {
// In ICU 72+ and above, ':' are no longer considered word connectors. The
// query tokenizer should still consider them to be property restricts.
constexpr std::string_view kText = "foo:bar";
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> indexing_tokens,
indexing_tokenizer->TokenizeAll(kText));
std::vector<std::string> indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("foo", "bar"));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> query_tokens,
query_tokenizer->TokenizeAll(kText));
std::vector<std::string> query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("bar"));
// This difference, however, should only apply to the first ':'. Both should
// consider a second ':' to be a word break.
constexpr std::string_view kText2 = "foo:bar:baz";
ICING_ASSERT_OK_AND_ASSIGN(indexing_tokens,
indexing_tokenizer->TokenizeAll(kText2));
indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("foo", "bar", "baz"));
ICING_ASSERT_OK_AND_ASSIGN(query_tokens,
query_tokenizer->TokenizeAll(kText2));
query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("bar", "baz"));
} else {
// This is a difference between the two tokenizers. "foo:bar" is a single
// token to the plain tokenizer because ':' is a word connector. But
// "foo:bar" is a property restrict to the query tokenizer - so "foo" is the
// property and "bar" is the only text term.
constexpr std::string_view kText = "foo:bar";
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> indexing_tokens,
indexing_tokenizer->TokenizeAll(kText));
std::vector<std::string> indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("foo:bar"));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> query_tokens,
query_tokenizer->TokenizeAll(kText));
std::vector<std::string> query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("bar"));
// This difference, however, should only apply to the first ':'. A
// second ':' should be treated by both tokenizers as a word connector.
constexpr std::string_view kText2 = "foo:bar:baz";
ICING_ASSERT_OK_AND_ASSIGN(indexing_tokens,
indexing_tokenizer->TokenizeAll(kText2));
indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("foo:bar:baz"));
ICING_ASSERT_OK_AND_ASSIGN(query_tokens,
query_tokenizer->TokenizeAll(kText2));
query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("bar:baz"));
}
}
TEST_F(CombinedTokenizerTest, Punctuation) {
const std::string_view kText = "Who? What!? Why & How.";
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> indexing_tokenizer,
tokenizer_factory::CreateIndexingTokenizer(
StringIndexingConfig::TokenizerType::PLAIN, lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(
std::unique_ptr<Tokenizer> query_tokenizer,
CreateQueryTokenizer(tokenizer_factory::QueryTokenizerType::RAW_QUERY,
lang_segmenter_.get()));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> indexing_tokens,
indexing_tokenizer->TokenizeAll(kText));
std::vector<std::string> indexing_terms = GetTokenTerms(indexing_tokens);
EXPECT_THAT(indexing_terms, ElementsAre("Who", "What", "Why", "How"));
ICING_ASSERT_OK_AND_ASSIGN(std::vector<Token> query_tokens,
query_tokenizer->TokenizeAll(kText));
std::vector<std::string> query_terms = GetTokenTerms(query_tokens);
EXPECT_THAT(query_terms, ElementsAre("Who", "What", "Why", "How"));
}
} // namespace lib
} // namespace icing
|
