ver4_dict_constants.cpp revision a245d15da5d295af21ead9a01583c64796a31ad7
1/* 2 * Copyright (C) 2013, 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#include "suggest/policyimpl/dictionary/structure/v4/ver4_dict_constants.h" 18 19namespace latinime { 20 21// These values MUST match the definitions in FormatSpec.java. 22const char *const Ver4DictConstants::TRIE_FILE_EXTENSION = ".trie"; 23const char *const Ver4DictConstants::HEADER_FILE_EXTENSION = ".header"; 24const char *const Ver4DictConstants::FREQ_FILE_EXTENSION = ".freq"; 25// tat = Terminal Address Table 26const char *const Ver4DictConstants::TERMINAL_ADDRESS_TABLE_FILE_EXTENSION = ".tat"; 27const char *const Ver4DictConstants::BIGRAM_FILE_EXTENSION = ".bigram_freq"; 28const char *const Ver4DictConstants::BIGRAM_LOOKUP_TABLE_FILE_EXTENSION = ".bigram_lookup"; 29const char *const Ver4DictConstants::BIGRAM_CONTENT_TABLE_FILE_EXTENSION = ".bigram_index_freq"; 30const char *const Ver4DictConstants::SHORTCUT_FILE_EXTENSION = ".shortcut_shortcut"; 31const char *const Ver4DictConstants::SHORTCUT_LOOKUP_TABLE_FILE_EXTENSION = ".shortcut_lookup"; 32const char *const Ver4DictConstants::SHORTCUT_CONTENT_TABLE_FILE_EXTENSION = 33 ".shortcut_index_shortcut"; 34 35// Version 4 dictionary size is implicitly limited to 8MB due to 3-byte offsets. 36const int Ver4DictConstants::MAX_DICTIONARY_SIZE = 8 * 1024 * 1024; 37// Extended region size, which is not GCed region size in dict file + additional buffer size, is 38// limited to 1MB to prevent from inefficient traversing. 39const int Ver4DictConstants::MAX_DICT_EXTENDED_REGION_SIZE = 1 * 1024 * 1024; 40 41const int Ver4DictConstants::NOT_A_TERMINAL_ID = -1; 42const int Ver4DictConstants::PROBABILITY_SIZE = 1; 43const int Ver4DictConstants::FLAGS_IN_PROBABILITY_FILE_SIZE = 1; 44const int Ver4DictConstants::TERMINAL_ADDRESS_TABLE_ADDRESS_SIZE = 3; 45const int Ver4DictConstants::NOT_A_TERMINAL_ADDRESS = 0; 46const int Ver4DictConstants::TERMINAL_ID_FIELD_SIZE = 4; 47const int Ver4DictConstants::TIME_STAMP_FIELD_SIZE = 4; 48const int Ver4DictConstants::WORD_LEVEL_FIELD_SIZE = 1; 49const int Ver4DictConstants::WORD_COUNT_FIELD_SIZE = 1; 50 51const int Ver4DictConstants::BIGRAM_ADDRESS_TABLE_BLOCK_SIZE = 4; 52const int Ver4DictConstants::BIGRAM_ADDRESS_TABLE_DATA_SIZE = 4; 53const int Ver4DictConstants::SHORTCUT_ADDRESS_TABLE_BLOCK_SIZE = 16; 54const int Ver4DictConstants::SHORTCUT_ADDRESS_TABLE_DATA_SIZE = 4; 55 56const int Ver4DictConstants::BIGRAM_TARGET_TERMINAL_ID_FIELD_SIZE = 3; 57// Unsigned int max value of BIGRAM_TARGET_TERMINAL_ID_FIELD_SIZE-byte is used for representing 58// invalid terminal ID in bigram lists. 59const int Ver4DictConstants::INVALID_BIGRAM_TARGET_TERMINAL_ID = 60 (1 << (BIGRAM_TARGET_TERMINAL_ID_FIELD_SIZE * 8)) - 1; 61const int Ver4DictConstants::BIGRAM_FLAGS_FIELD_SIZE = 1; 62const int Ver4DictConstants::BIGRAM_PROBABILITY_MASK = 0x0F; 63const int Ver4DictConstants::BIGRAM_HAS_NEXT_MASK = 0x80; 64const int Ver4DictConstants::BIGRAM_LARGE_PROBABILITY_FIELD_SIZE = 1; 65 66const int Ver4DictConstants::SHORTCUT_FLAGS_FIELD_SIZE = 1; 67const int Ver4DictConstants::SHORTCUT_PROBABILITY_MASK = 0x0F; 68const int Ver4DictConstants::SHORTCUT_HAS_NEXT_MASK = 0x80; 69 70} // namespace latinime 71