1*993b0882SAndroid Build Coastguard Worker /* 2*993b0882SAndroid Build Coastguard Worker * Copyright (C) 2018 The Android Open Source Project 3*993b0882SAndroid Build Coastguard Worker * 4*993b0882SAndroid Build Coastguard Worker * Licensed under the Apache License, Version 2.0 (the "License"); 5*993b0882SAndroid Build Coastguard Worker * you may not use this file except in compliance with the License. 6*993b0882SAndroid Build Coastguard Worker * You may obtain a copy of the License at 7*993b0882SAndroid Build Coastguard Worker * 8*993b0882SAndroid Build Coastguard Worker * http://www.apache.org/licenses/LICENSE-2.0 9*993b0882SAndroid Build Coastguard Worker * 10*993b0882SAndroid Build Coastguard Worker * Unless required by applicable law or agreed to in writing, software 11*993b0882SAndroid Build Coastguard Worker * distributed under the License is distributed on an "AS IS" BASIS, 12*993b0882SAndroid Build Coastguard Worker * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13*993b0882SAndroid Build Coastguard Worker * See the License for the specific language governing permissions and 14*993b0882SAndroid Build Coastguard Worker * limitations under the License. 15*993b0882SAndroid Build Coastguard Worker */ 16*993b0882SAndroid Build Coastguard Worker 17*993b0882SAndroid Build Coastguard Worker #ifndef LIBTEXTCLASSIFIER_UTILS_TOKENFREE_BYTE_ENCODER_H_ 18*993b0882SAndroid Build Coastguard Worker #define LIBTEXTCLASSIFIER_UTILS_TOKENFREE_BYTE_ENCODER_H_ 19*993b0882SAndroid Build Coastguard Worker 20*993b0882SAndroid Build Coastguard Worker #include <vector> 21*993b0882SAndroid Build Coastguard Worker 22*993b0882SAndroid Build Coastguard Worker #include "utils/base/logging.h" 23*993b0882SAndroid Build Coastguard Worker #include "utils/container/string-set.h" 24*993b0882SAndroid Build Coastguard Worker #include "utils/strings/stringpiece.h" 25*993b0882SAndroid Build Coastguard Worker 26*993b0882SAndroid Build Coastguard Worker namespace libtextclassifier3 { 27*993b0882SAndroid Build Coastguard Worker 28*993b0882SAndroid Build Coastguard Worker // Encoder to segment/tokenize strings into bytes 29*993b0882SAndroid Build Coastguard Worker class ByteEncoder { 30*993b0882SAndroid Build Coastguard Worker public: 31*993b0882SAndroid Build Coastguard Worker bool Encode(StringPiece input_text, std::vector<int64_t>* encoded_text) const; ByteEncoder()32*993b0882SAndroid Build Coastguard Worker ByteEncoder() {} 33*993b0882SAndroid Build Coastguard Worker }; 34*993b0882SAndroid Build Coastguard Worker 35*993b0882SAndroid Build Coastguard Worker } // namespace libtextclassifier3 36*993b0882SAndroid Build Coastguard Worker 37*993b0882SAndroid Build Coastguard Worker #endif // LIBTEXTCLASSIFIER_UTILS_TOKENFREE_BYTE_ENCODER_H_ 38