1 /* 2 * Copyright (C) 2017 The Android Open Source Project 3 * 4 * Licensed under the Apache License, Version 2.0 (the "License"); 5 * you may not use this file except in compliance with the License. 6 * You may obtain a copy of the License at 7 * 8 * http://www.apache.org/licenses/LICENSE-2.0 9 * 10 * Unless required by applicable law or agreed to in writing, software 11 * distributed under the License is distributed on an "AS IS" BASIS, 12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 * See the License for the specific language governing permissions and 14 * limitations under the License. 15 */ 16 17 #ifndef LIBTEXTCLASSIFIER_LANG_ID_LIGHT_SENTENCE_H_ 18 #define LIBTEXTCLASSIFIER_LANG_ID_LIGHT_SENTENCE_H_ 19 20 #include <string> 21 #include <vector> 22 23 #include "util/base/logging.h" 24 #include "util/base/macros.h" 25 26 namespace libtextclassifier { 27 namespace nlp_core { 28 namespace lang_id { 29 30 // Simplified replacement for the Sentence proto, for internal use in the 31 // language identification code. 32 // 33 // In this simplified form, a sentence is a vector of words, each word being a 34 // string. 35 class LightSentence { 36 public: 37 LightSentence() {} 38 39 // Adds a new word after all existing ones, and returns a pointer to it. The 40 // new word is initialized to the empty string. 41 std::string *add_word() { 42 words_.emplace_back(); 43 return &(words_.back()); 44 } 45 46 // Returns number of words from this LightSentence. 47 int num_words() const { return words_.size(); } 48 49 // Returns the ith word from this LightSentence. Note: undefined behavior if 50 // i is out of bounds. 51 const std::string &word(int i) const { 52 TC_DCHECK((i >= 0) && (i < num_words())); 53 return words_[i]; 54 } 55 56 private: 57 std::vector<std::string> words_; 58 59 TC_DISALLOW_COPY_AND_ASSIGN(LightSentence); 60 }; 61 62 } // namespace lang_id 63 } // namespace nlp_core 64 } // namespace libtextclassifier 65 66 #endif // LIBTEXTCLASSIFIER_LANG_ID_LIGHT_SENTENCE_H_ 67
