WordInputEventForPersonalization.java revision 16cc3992d7468ef781753df7b4227330e0834501 
1/*
2 * Copyright (C) 2014 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
17package com.android.inputmethod.latin.utils;
18
19import android.util.Log;
20
21import com.android.inputmethod.annotations.UsedForTesting;
22import com.android.inputmethod.latin.Constants;
23import com.android.inputmethod.latin.NgramContext;
24import com.android.inputmethod.latin.settings.SpacingAndPunctuations;
25import com.android.inputmethod.latin.utils.DistracterFilter.HandlingType;
26
27import java.util.ArrayList;
28import java.util.List;
29import java.util.Locale;
30
31// Note: this class is used as a parameter type of a native method. You should be careful when you
32// rename this class or field name. See BinaryDictionary#addMultipleDictionaryEntriesNative().
33public final class WordInputEventForPersonalization {
34    private static final String TAG = WordInputEventForPersonalization.class.getSimpleName();
35    private static final boolean DEBUG_TOKEN = false;
36
37    public final int[] mTargetWord;
38    public final int mPrevWordsCount;
39    public final int[][] mPrevWordArray = new int[Constants.MAX_PREV_WORD_COUNT_FOR_N_GRAM][];
40    public final boolean[] mIsPrevWordBeginningOfSentenceArray =
41            new boolean[Constants.MAX_PREV_WORD_COUNT_FOR_N_GRAM];
42    public final boolean mIsValid;
43    // Time stamp in seconds.
44    public final int mTimestamp;
45
46    @UsedForTesting
47    public WordInputEventForPersonalization(final CharSequence targetWord,
48            final NgramContext ngramContext, final boolean isValid, final int timestamp) {
49        mTargetWord = StringUtils.toCodePointArray(targetWord);
50        mPrevWordsCount = ngramContext.getPrevWordCount();
51        ngramContext.outputToArray(mPrevWordArray, mIsPrevWordBeginningOfSentenceArray);
52        mIsValid = isValid;
53        mTimestamp = timestamp;
54    }
55
56    // Process a list of words and return a list of {@link WordInputEventForPersonalization}
57    // objects.
58    public static ArrayList<WordInputEventForPersonalization> createInputEventFrom(
59            final List<String> tokens, final int timestamp,
60            final SpacingAndPunctuations spacingAndPunctuations, final Locale locale,
61            final DistracterFilter distracterFilter) {
62        final ArrayList<WordInputEventForPersonalization> inputEvents = new ArrayList<>();
63        final int N = tokens.size();
64        NgramContext ngramContext = NgramContext.EMPTY_PREV_WORDS_INFO;
65        for (int i = 0; i < N; ++i) {
66            final String tempWord = tokens.get(i);
67            if (StringUtils.isEmptyStringOrWhiteSpaces(tempWord)) {
68                // just skip this token
69                if (DEBUG_TOKEN) {
70                    Log.d(TAG, "--- isEmptyStringOrWhiteSpaces: \"" + tempWord + "\"");
71                }
72                continue;
73            }
74            if (!DictionaryInfoUtils.looksValidForDictionaryInsertion(
75                    tempWord, spacingAndPunctuations)) {
76                if (DEBUG_TOKEN) {
77                    Log.d(TAG, "--- not looksValidForDictionaryInsertion: \""
78                            + tempWord + "\"");
79                }
80                // Sentence terminator found. Split.
81                // TODO: Detect whether the context is beginning-of-sentence.
82                ngramContext = NgramContext.EMPTY_PREV_WORDS_INFO;
83                continue;
84            }
85            if (DEBUG_TOKEN) {
86                Log.d(TAG, "--- word: \"" + tempWord + "\"");
87            }
88            final WordInputEventForPersonalization inputEvent =
89                    detectWhetherVaildWordOrNotAndGetInputEvent(
90                            ngramContext, tempWord, timestamp, locale, distracterFilter);
91            if (inputEvent == null) {
92                continue;
93            }
94            inputEvents.add(inputEvent);
95            ngramContext = ngramContext.getNextNgramContext(new NgramContext.WordInfo(tempWord));
96        }
97        return inputEvents;
98    }
99
100    private static WordInputEventForPersonalization detectWhetherVaildWordOrNotAndGetInputEvent(
101            final NgramContext ngramContext, final String targetWord, final int timestamp,
102            final Locale locale, final DistracterFilter distracterFilter) {
103        if (locale == null) {
104            return null;
105        }
106        final int wordHandlingType = distracterFilter.getWordHandlingType(ngramContext,
107                targetWord, locale);
108        final String word = HandlingType.shouldBeLowerCased(wordHandlingType) ?
109                targetWord.toLowerCase(locale) : targetWord;
110        if (distracterFilter.isDistracterToWordsInDictionaries(ngramContext, targetWord, locale)) {
111            // The word is a distracter.
112            return null;
113        }
114        return new WordInputEventForPersonalization(word, ngramContext,
115                !HandlingType.shouldBeHandledAsOov(wordHandlingType), timestamp);
116    }
117}
118