aboutsummaryrefslogtreecommitdiffstats
path: root/java/src/com/android/inputmethod/latin/makedict/DictDecoder.java
blob: e251f7df7c4fbb757a27dbd2b876e9203d73e295 (about) (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
/*
 * Copyright (C) 2013 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package com.android.inputmethod.latin.makedict;

import com.android.inputmethod.annotations.UsedForTesting;
import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.CharEncoding;
import com.android.inputmethod.latin.makedict.BinaryDictDecoderUtils.DictBuffer;
import com.android.inputmethod.latin.makedict.FormatSpec.FileHeader;
import com.android.inputmethod.latin.makedict.FormatSpec.FormatOptions;
import com.android.inputmethod.latin.makedict.FusionDictionary.WeightedString;
import com.android.inputmethod.latin.utils.ByteArrayDictBuffer;

import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.RandomAccessFile;
import java.nio.ByteBuffer;
import java.nio.channels.FileChannel;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.TreeMap;

/**
 * The base class of binary dictionary decoders.
 */
public abstract class DictDecoder {

    protected FileHeader readHeader(final DictBuffer dictBuffer)
            throws IOException, UnsupportedFormatException {
        if (dictBuffer == null) {
            openDictBuffer();
        }

        final int version = HeaderReader.readVersion(dictBuffer);
        if (version < FormatSpec.MINIMUM_SUPPORTED_VERSION
                || version > FormatSpec.MAXIMUM_SUPPORTED_VERSION) {
            throw new UnsupportedFormatException("Unsupported version : " + version);
        }
        // TODO: Remove this field.
        final int optionsFlags = HeaderReader.readOptionFlags(dictBuffer);

        final int headerSize = HeaderReader.readHeaderSize(dictBuffer);

        if (headerSize < 0) {
            throw new UnsupportedFormatException("header size can't be negative.");
        }

        final HashMap<String, String> attributes = HeaderReader.readAttributes(dictBuffer,
                headerSize);

        final FileHeader header = new FileHeader(headerSize,
                new FusionDictionary.DictionaryOptions(attributes,
                        0 != (optionsFlags & FormatSpec.GERMAN_UMLAUT_PROCESSING_FLAG),
                        0 != (optionsFlags & FormatSpec.FRENCH_LIGATURE_PROCESSING_FLAG)),
                        new FormatOptions(version,
                                0 != (optionsFlags & FormatSpec.SUPPORTS_DYNAMIC_UPDATE)));
        return header;
    }

    /**
     * Reads and returns the file header.
     */
    public abstract FileHeader readHeader() throws IOException, UnsupportedFormatException;

    /**
     * Reads PtNode from nodeAddress.
     * @param ptNodePos the position of PtNode.
     * @param formatOptions the format options.
     * @return PtNodeInfo.
     */
    public abstract PtNodeInfo readPtNode(final int ptNodePos, final FormatOptions formatOptions);

    /**
     * Reads a buffer and returns the memory representation of the dictionary.
     *
     * This high-level method takes a buffer and reads its contents, populating a
     * FusionDictionary structure. The optional dict argument is an existing dictionary to
     * which words from the buffer should be added. If it is null, a new dictionary is created.
     *
     * @param dict an optional dictionary to add words to, or null.
     * @param deleteDictIfBroken a flag indicating whether this method should remove the broken
     * dictionary or not.
     * @return the created (or merged) dictionary.
     */
    @UsedForTesting
    public abstract FusionDictionary readDictionaryBinary(final FusionDictionary dict,
            final boolean deleteDictIfBroken)
                    throws FileNotFoundException, IOException, UnsupportedFormatException;

    /**
     * Gets the address of the last PtNode of the exact matching word in the dictionary.
     * If no match is found, returns NOT_VALID_WORD.
     *
     * @param word the word we search for.
     * @return the address of the terminal node.
     * @throws IOException if the file can't be read.
     * @throws UnsupportedFormatException if the format of the file is not recognized.
     */
    @UsedForTesting
    public int getTerminalPosition(final String word)
            throws IOException, UnsupportedFormatException {
        if (!isDictBufferOpen()) {
            openDictBuffer();
        }
        return BinaryDictIOUtils.getTerminalPosition(this, word);
    }

    /**
     * Reads unigrams and bigrams from the binary file.
     * Doesn't store a full memory representation of the dictionary.
     *
     * @param words the map to store the address as a key and the word as a value.
     * @param frequencies the map to store the address as a key and the frequency as a value.
     * @param bigrams the map to store the address as a key and the list of address as a value.
     * @throws IOException if the file can't be read.
     * @throws UnsupportedFormatException if the format of the file is not recognized.
     */
    @UsedForTesting
    public void readUnigramsAndBigramsBinary(final TreeMap<Integer, String> words,
            final TreeMap<Integer, Integer> frequencies,
            final TreeMap<Integer, ArrayList<PendingAttribute>> bigrams)
            throws IOException, UnsupportedFormatException {
        if (!isDictBufferOpen()) {
            openDictBuffer();
        }
        BinaryDictIOUtils.readUnigramsAndBigramsBinary(this, words, frequencies, bigrams);
    }

    /**
     * Sets the position of the buffer to the given value.
     *
     * @param newPos the new position
     */
    public abstract void setPosition(final int newPos);

    /**
     * Gets the position of the buffer.
     *
     * @return the position
     */
    public abstract int getPosition();

    /**
     * Reads and returns the PtNode count out of a buffer and forwards the pointer.
     */
    public abstract int readPtNodeCount();

    /**
     * Reads the forward link and advances the position.
     *
     * @return true if this method moves the file pointer, false otherwise.
     */
    public abstract boolean readAndFollowForwardLink();
    public abstract boolean hasNextPtNodeArray();

    /**
     * Opens the dictionary file and makes DictBuffer.
     */
    @UsedForTesting
    public abstract void openDictBuffer() throws FileNotFoundException, IOException;
    @UsedForTesting
    public abstract boolean isDictBufferOpen();

    // Constants for DictionaryBufferFactory.
    public static final int USE_READONLY_BYTEBUFFER = 0x01000000;
    public static final int USE_BYTEARRAY = 0x02000000;
    public static final int USE_WRITABLE_BYTEBUFFER = 0x03000000;
    public static final int MASK_DICTBUFFER = 0x0F000000;

    public interface DictionaryBufferFactory {
        public DictBuffer getDictionaryBuffer(final File file)
                throws FileNotFoundException, IOException;
    }

    /**
     * Creates DictionaryBuffer using a ByteBuffer
     *
     * This class uses less memory than DictionaryBufferFromByteArrayFactory,
     * but doesn't perform as fast.
     * When operating on a big dictionary, this class is preferred.
     */
    public static final class DictionaryBufferFromReadOnlyByteBufferFactory
            implements DictionaryBufferFactory {
        @Override
        public DictBuffer getDictionaryBuffer(final File file)
                throws FileNotFoundException, IOException {
            FileInputStream inStream = null;
            ByteBuffer buffer = null;
            try {
                inStream = new FileInputStream(file);
                buffer = inStream.getChannel().map(FileChannel.MapMode.READ_ONLY,
                        0, file.length());
            } finally {
                if (inStream != null) {
                    inStream.close();
                }
            }
            if (buffer != null) {
                return new BinaryDictDecoderUtils.ByteBufferDictBuffer(buffer);
            }
            return null;
        }
    }

    /**
     * Creates DictionaryBuffer using a byte array
     *
     * This class performs faster than other classes, but consumes more memory.
     * When operating on a small dictionary, this class is preferred.
     */
    public static final class DictionaryBufferFromByteArrayFactory
            implements DictionaryBufferFactory {
        @Override
        public DictBuffer getDictionaryBuffer(final File file)
                throws FileNotFoundException, IOException {
            FileInputStream inStream = null;
            try {
                inStream = new FileInputStream(file);
                final byte[] array = new byte[(int) file.length()];
                inStream.read(array);
                return new ByteArrayDictBuffer(array);
            } finally {
                if (inStream != null) {
                    inStream.close();
                }
            }
        }
    }

    /**
     * Creates DictionaryBuffer using a writable ByteBuffer and a RandomAccessFile.
     *
     * This class doesn't perform as fast as other classes,
     * but this class is the only option available for destructive operations (insert or delete)
     * on a dictionary.
     */
    @UsedForTesting
    public static final class DictionaryBufferFromWritableByteBufferFactory
            implements DictionaryBufferFactory {
        @Override
        public DictBuffer getDictionaryBuffer(final File file)
                throws FileNotFoundException, IOException {
            RandomAccessFile raFile = null;
            ByteBuffer buffer = null;
            try {
                raFile = new RandomAccessFile(file, "rw");
                buffer = raFile.getChannel().map(FileChannel.MapMode.READ_WRITE, 0, file.length());
            } finally {
                if (raFile != null) {
                    raFile.close();
                }
            }
            if (buffer != null) {
                return new BinaryDictDecoderUtils.ByteBufferDictBuffer(buffer);
            }
            return null;
        }
    }

    /**
     * A utility class for reading a file header.
     */
    protected static class HeaderReader {
        protected static int readVersion(final DictBuffer dictBuffer)
                throws IOException, UnsupportedFormatException {
            return BinaryDictDecoderUtils.checkFormatVersion(dictBuffer);
        }

        protected static int readOptionFlags(final DictBuffer dictBuffer) {
            return dictBuffer.readUnsignedShort();
        }

        protected static int readHeaderSize(final DictBuffer dictBuffer) {
            return dictBuffer.readInt();
        }

        protected static HashMap<String, String> readAttributes(final DictBuffer dictBuffer,
                final int headerSize) {
            final HashMap<String, String> attributes = new HashMap<String, String>();
            while (dictBuffer.position() < headerSize) {
                // We can avoid an infinite loop here since dictBuffer.position() is always
                // increased by calling CharEncoding.readString.
                final String key = CharEncoding.readString(dictBuffer);
                final String value = CharEncoding.readString(dictBuffer);
                attributes.put(key, value);
            }
            dictBuffer.position(headerSize);
            return attributes;
        }
    }

    /**
     * A utility class for reading a PtNode.
     */
    protected static class PtNodeReader {
        protected static int readPtNodeOptionFlags(final DictBuffer dictBuffer) {
            return dictBuffer.readUnsignedByte();
        }

        protected static int readParentAddress(final DictBuffer dictBuffer,
                final FormatOptions formatOptions) {
            if (BinaryDictIOUtils.supportsDynamicUpdate(formatOptions)) {
                return BinaryDictDecoderUtils.readSInt24(dictBuffer);
            } else {
                return FormatSpec.NO_PARENT_ADDRESS;
            }
        }

        protected static int readChildrenAddress(final DictBuffer dictBuffer, final int optionFlags,
                final FormatOptions formatOptions) {
            if (BinaryDictIOUtils.supportsDynamicUpdate(formatOptions)) {
                final int address = BinaryDictDecoderUtils.readSInt24(dictBuffer);
                if (address == 0) return FormatSpec.NO_CHILDREN_ADDRESS;
                return address;
            } else {
                switch (optionFlags & FormatSpec.MASK_CHILDREN_ADDRESS_TYPE) {
                    case FormatSpec.FLAG_CHILDREN_ADDRESS_TYPE_ONEBYTE:
                        return dictBuffer.readUnsignedByte();
                    case FormatSpec.FLAG_CHILDREN_ADDRESS_TYPE_TWOBYTES:
                        return dictBuffer.readUnsignedShort();
                    case FormatSpec.FLAG_CHILDREN_ADDRESS_TYPE_THREEBYTES:
                        return dictBuffer.readUnsignedInt24();
                    case FormatSpec.FLAG_CHILDREN_ADDRESS_TYPE_NOADDRESS:
                    default:
                        return FormatSpec.NO_CHILDREN_ADDRESS;
                }
            }
        }

        // Reads shortcuts and returns the read length.
        protected static int readShortcut(final DictBuffer dictBuffer,
                final ArrayList<WeightedString> shortcutTargets) {
            final int pointerBefore = dictBuffer.position();
            dictBuffer.readUnsignedShort(); // skip the size
            while (true) {
                final int targetFlags = dictBuffer.readUnsignedByte();
                final String word = CharEncoding.readString(dictBuffer);
                shortcutTargets.add(new WeightedString(word,
                        targetFlags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_FREQUENCY));
                if (0 == (targetFlags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_HAS_NEXT)) break;
            }
            return dictBuffer.position() - pointerBefore;
        }

        protected static int readBigramAddresses(final DictBuffer dictBuffer,
                final ArrayList<PendingAttribute> bigrams, final int baseAddress) {
            int readLength = 0;
            int bigramCount = 0;
            while (bigramCount++ < FormatSpec.MAX_BIGRAMS_IN_A_PTNODE) {
                final int bigramFlags = dictBuffer.readUnsignedByte();
                ++readLength;
                final int sign = 0 == (bigramFlags & FormatSpec.FLAG_BIGRAM_ATTR_OFFSET_NEGATIVE)
                        ? 1 : -1;
                int bigramAddress = baseAddress + readLength;
                switch (bigramFlags & FormatSpec.MASK_BIGRAM_ATTR_ADDRESS_TYPE) {
                    case FormatSpec.FLAG_BIGRAM_ATTR_ADDRESS_TYPE_ONEBYTE:
                        bigramAddress += sign * dictBuffer.readUnsignedByte();
                        readLength += 1;
                        break;
                    case FormatSpec.FLAG_BIGRAM_ATTR_ADDRESS_TYPE_TWOBYTES:
                        bigramAddress += sign * dictBuffer.readUnsignedShort();
                        readLength += 2;
                        break;
                    case FormatSpec.FLAG_BIGRAM_ATTR_ADDRESS_TYPE_THREEBYTES:
                        bigramAddress += sign * dictBuffer.readUnsignedInt24();
                        readLength += 3;
                        break;
                    default:
                        throw new RuntimeException("Has bigrams with no address");
                }
                bigrams.add(new PendingAttribute(
                        bigramFlags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_FREQUENCY,
                        bigramAddress));
                if (0 == (bigramFlags & FormatSpec.FLAG_BIGRAM_SHORTCUT_ATTR_HAS_NEXT)) break;
            }
            return readLength;
        }
    }

    public abstract void skipPtNode(final FormatOptions formatOptions);
}