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 "dictionary/structure/v4/ver4_patricia_trie_node_reader.h"
18
19 #include "dictionary/header/header_policy.h"
20 #include "dictionary/structure/pt_common/dynamic_pt_reading_utils.h"
21 #include "dictionary/structure/pt_common/patricia_trie_reading_utils.h"
22 #include "dictionary/structure/v4/content/language_model_dict_content.h"
23 #include "dictionary/structure/v4/content/probability_entry.h"
24 #include "dictionary/structure/v4/ver4_patricia_trie_reading_utils.h"
25 #include "dictionary/utils/buffer_with_extendable_buffer.h"
26 #include "dictionary/utils/forgetting_curve_utils.h"
27
28 namespace latinime {
29
fetchPtNodeInfoFromBufferAndProcessMovedPtNode(const int ptNodePos,const int siblingNodePos) const30 const PtNodeParams Ver4PatriciaTrieNodeReader::fetchPtNodeInfoFromBufferAndProcessMovedPtNode(
31 const int ptNodePos, const int siblingNodePos) const {
32 if (ptNodePos < 0 || ptNodePos >= mBuffer->getTailPosition()) {
33 // Reading invalid position because of bug or broken dictionary.
34 AKLOGE("Fetching PtNode info from invalid dictionary position: %d, dictionary size: %d",
35 ptNodePos, mBuffer->getTailPosition());
36 ASSERT(false);
37 return PtNodeParams();
38 }
39 const bool usesAdditionalBuffer = mBuffer->isInAdditionalBuffer(ptNodePos);
40 const uint8_t *const dictBuf = mBuffer->getBuffer(usesAdditionalBuffer);
41 int pos = ptNodePos;
42 const int headPos = ptNodePos;
43 if (usesAdditionalBuffer) {
44 pos -= mBuffer->getOriginalBufferSize();
45 }
46 const PatriciaTrieReadingUtils::NodeFlags flags =
47 PatriciaTrieReadingUtils::getFlagsAndAdvancePosition(dictBuf, &pos);
48 const int parentPosOffset =
49 DynamicPtReadingUtils::getParentPtNodePosOffsetAndAdvancePosition(
50 dictBuf, &pos);
51 const int parentPos =
52 DynamicPtReadingUtils::getParentPtNodePos(parentPosOffset, headPos);
53 int codePoints[MAX_WORD_LENGTH];
54 // Code point table is not used for ver4 dictionaries.
55 const int codePointCount = PatriciaTrieReadingUtils::getCharsAndAdvancePosition(
56 dictBuf, flags, MAX_WORD_LENGTH, nullptr /* codePointTable */, codePoints, &pos);
57 int terminalIdFieldPos = NOT_A_DICT_POS;
58 int terminalId = Ver4DictConstants::NOT_A_TERMINAL_ID;
59 if (PatriciaTrieReadingUtils::isTerminal(flags)) {
60 terminalIdFieldPos = pos;
61 if (usesAdditionalBuffer) {
62 terminalIdFieldPos += mBuffer->getOriginalBufferSize();
63 }
64 terminalId = Ver4PatriciaTrieReadingUtils::getTerminalIdAndAdvancePosition(dictBuf, &pos);
65 }
66 int childrenPosFieldPos = pos;
67 if (usesAdditionalBuffer) {
68 childrenPosFieldPos += mBuffer->getOriginalBufferSize();
69 }
70 int childrenPos = DynamicPtReadingUtils::readChildrenPositionAndAdvancePosition(
71 dictBuf, &pos);
72 if (usesAdditionalBuffer && childrenPos != NOT_A_DICT_POS) {
73 childrenPos += mBuffer->getOriginalBufferSize();
74 }
75 if (usesAdditionalBuffer) {
76 pos += mBuffer->getOriginalBufferSize();
77 }
78 // Sibling position is the tail position of original PtNode.
79 int newSiblingNodePos = (siblingNodePos == NOT_A_DICT_POS) ? pos : siblingNodePos;
80 // Read destination node if the read node is a moved node.
81 if (DynamicPtReadingUtils::isMoved(flags)) {
82 // The destination position is stored at the same place as the parent position.
83 return fetchPtNodeInfoFromBufferAndProcessMovedPtNode(parentPos, newSiblingNodePos);
84 } else {
85 return PtNodeParams(headPos, flags, parentPos, codePointCount, codePoints,
86 terminalIdFieldPos, terminalId, NOT_A_PROBABILITY, childrenPosFieldPos, childrenPos,
87 newSiblingNodePos);
88 }
89 }
90
91 } // namespace latinime
92