Home | History | Annotate | Download | only in makedict 
      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 package com.android.inputmethod.latin.makedict;
     18 
     19 import com.android.inputmethod.annotations.UsedForTesting;
     20 import com.android.inputmethod.latin.BinaryDictionary;
     21 import com.android.inputmethod.latin.common.FileUtils;
     22 
     23 import java.io.File;
     24 import java.io.FileNotFoundException;
     25 import java.io.IOException;
     26 import java.util.ArrayList;
     27 
     28 /**
     29  * An implementation of binary dictionary decoder for version 4 binary dictionary.
     30  */
     31 @UsedForTesting
     32 public class Ver4DictDecoder extends AbstractDictDecoder {
     33     final File mDictDirectory;
     34 
     35     @UsedForTesting
     36     /* package */ Ver4DictDecoder(final File dictDirectory) {
     37         mDictDirectory = dictDirectory;
     38 
     39     }
     40 
     41     @Override
     42     public DictionaryHeader readHeader() throws IOException, UnsupportedFormatException {
     43         // dictType is not being used in dicttool. Passing an empty string.
     44         final BinaryDictionary binaryDictionary= new BinaryDictionary(
     45               mDictDirectory.getAbsolutePath(), 0 /* offset */, 0 /* length */,
     46               true /* useFullEditDistance */, null /* locale */,
     47               "" /* dictType */, true /* isUpdatable */);
     48         final DictionaryHeader header = binaryDictionary.getHeader();
     49         binaryDictionary.close();
     50         if (header == null) {
     51             throw new IOException("Cannot read the dictionary header.");
     52         }
     53         return header;
     54     }
     55 
     56     @Override
     57     public FusionDictionary readDictionaryBinary(final boolean deleteDictIfBroken)
     58             throws FileNotFoundException, IOException, UnsupportedFormatException {
     59         // dictType is not being used in dicttool. Passing an empty string.
     60         final BinaryDictionary binaryDictionary = new BinaryDictionary(
     61               mDictDirectory.getAbsolutePath(), 0 /* offset */, 0 /* length */,
     62               true /* useFullEditDistance */, null /* locale */,
     63               "" /* dictType */, true /* isUpdatable */);
     64         final DictionaryHeader header = readHeader();
     65         final FusionDictionary fusionDict =
     66                 new FusionDictionary(new FusionDictionary.PtNodeArray(), header.mDictionaryOptions);
     67         int token = 0;
     68         final ArrayList<WordProperty> wordProperties = new ArrayList<>();
     69         do {
     70             final BinaryDictionary.GetNextWordPropertyResult result =
     71                     binaryDictionary.getNextWordProperty(token);
     72             final WordProperty wordProperty = result.mWordProperty;
     73             if (wordProperty == null) {
     74                 binaryDictionary.close();
     75                 if (deleteDictIfBroken) {
     76                     FileUtils.deleteRecursively(mDictDirectory);
     77                 }
     78                 return null;
     79             }
     80             wordProperties.add(wordProperty);
     81             token = result.mNextToken;
     82         } while (token != 0);
     83 
     84         // Insert unigrams into the fusion dictionary.
     85         for (final WordProperty wordProperty : wordProperties) {
     86             fusionDict.add(wordProperty.mWord, wordProperty.mProbabilityInfo,
     87                     wordProperty.mIsNotAWord,
     88                     wordProperty.mIsPossiblyOffensive);
     89         }
     90         // Insert bigrams into the fusion dictionary.
     91         // TODO: Support ngrams.
     92         for (final WordProperty wordProperty : wordProperties) {
     93             if (!wordProperty.mHasNgrams) {
     94                 continue;
     95             }
     96             final String word0 = wordProperty.mWord;
     97             for (final WeightedString bigram : wordProperty.getBigrams()) {
     98                 fusionDict.setBigram(word0, bigram.mWord, bigram.mProbabilityInfo);
     99             }
    100         }
    101         binaryDictionary.close();
    102         return fusionDict;
    103     }
    104 }
    105