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
|
// Copyright (C) 2019 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 "icing/tokenization/plain-tokenizer.h"
#include <algorithm>
#include <cstdint>
#include <vector>
#include "icing/text_classifier/lib3/utils/base/statusor.h"
#include "icing/tokenization/language-segmenter.h"
#include "icing/util/character-iterator.h"
#include "icing/util/i18n-utils.h"
#include "icing/util/status-macros.h"
namespace icing {
namespace lib {
namespace {
// Helper function to validate a term.
// A term is valid if:
// 1. it's not empty
// 2. it's not a whitespace
// 3. it's not a punctuation mark
//
// TODO(b/141007791): figure out how we'd like to support special characters
// like "+", "&", "@", "#" in indexing and query tokenizers.
bool IsValidTerm(std::string_view term) {
if (term.empty()) {
return false;
}
// Gets the first unicode character. We can know what the whole term is by
// checking only the first character.
return !i18n_utils::IsWhitespaceAt(term, /*position=*/0) &&
!i18n_utils::IsPunctuationAt(term, /*position=*/0);
}
} // namespace
// Plain tokenizer applies its rules to the results from language segmenter. It
// simply filters out invalid terms from language segmenter and returns
// everything else as tokens. Please refer to IsValidTerm() above for what terms
// are valid.
class PlainTokenIterator : public Tokenizer::Iterator {
public:
explicit PlainTokenIterator(
std::unique_ptr<LanguageSegmenter::Iterator> base_iterator)
: base_iterator_(std::move(base_iterator)) {}
bool Advance() override {
bool found_next_valid_term = false;
while (!found_next_valid_term && base_iterator_->Advance()) {
current_term_ = base_iterator_->GetTerm();
found_next_valid_term = IsValidTerm(current_term_);
}
return found_next_valid_term;
}
std::vector<Token> GetTokens() const override {
std::vector<Token> result;
if (!current_term_.empty()) {
result.push_back(Token(Token::Type::REGULAR, current_term_));
}
return result;
}
libtextclassifier3::StatusOr<CharacterIterator> CalculateTokenStart()
override {
return base_iterator_->CalculateTermStart();
}
libtextclassifier3::StatusOr<CharacterIterator> CalculateTokenEndExclusive()
override {
return base_iterator_->CalculateTermEndExclusive();
}
bool ResetToTokenStartingAfter(int32_t utf32_offset) override {
if (!base_iterator_->ResetToTermStartingAfterUtf32(utf32_offset).ok()) {
return false;
}
current_term_ = base_iterator_->GetTerm();
if (!IsValidTerm(current_term_)) {
// If the current value isn't valid, advance to the next valid value.
return Advance();
}
return true;
}
bool ResetToTokenEndingBefore(int32_t utf32_offset) override {
ICING_ASSIGN_OR_RETURN(
utf32_offset,
base_iterator_->ResetToTermEndingBeforeUtf32(utf32_offset), false);
current_term_ = base_iterator_->GetTerm();
while (!IsValidTerm(current_term_)) {
// Haven't found a valid term yet. Retrieve the term prior to this one
// from the segmenter.
ICING_ASSIGN_OR_RETURN(
utf32_offset,
base_iterator_->ResetToTermEndingBeforeUtf32(utf32_offset), false);
current_term_ = base_iterator_->GetTerm();
}
return true;
}
bool ResetToStart() override {
if (!base_iterator_->ResetToStartUtf32().ok()) {
return false;
}
current_term_ = base_iterator_->GetTerm();
if (!IsValidTerm(current_term_)) {
// If the current value isn't valid, advance to the next valid value.
return Advance();
}
return true;
}
private:
std::unique_ptr<LanguageSegmenter::Iterator> base_iterator_;
std::string_view current_term_;
};
libtextclassifier3::StatusOr<std::unique_ptr<Tokenizer::Iterator>>
PlainTokenizer::Tokenize(std::string_view text,
LanguageSegmenter::AccessType access_type) const {
ICING_ASSIGN_OR_RETURN(
std::unique_ptr<LanguageSegmenter::Iterator> base_iterator,
language_segmenter_.Segment(text, access_type));
return std::make_unique<PlainTokenIterator>(std::move(base_iterator));
}
libtextclassifier3::StatusOr<std::vector<Token>> PlainTokenizer::TokenizeAll(
std::string_view text) const {
ICING_ASSIGN_OR_RETURN(
std::unique_ptr<Tokenizer::Iterator> iterator,
Tokenize(text, LanguageSegmenter::AccessType::kForwardIterator));
std::vector<Token> tokens;
while (iterator->Advance()) {
std::vector<Token> batch_tokens = iterator->GetTokens();
tokens.insert(tokens.end(), batch_tokens.begin(), batch_tokens.end());
}
return tokens;
}
} // namespace lib
} // namespace icing
|
