WordInputEventForPersonalization.java revision 0f7d881dc72132dfd75c8b4fe61a69fc5cdcd460 
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.NgramContext;
23import com.android.inputmethod.latin.common.StringUtils;
24import com.android.inputmethod.latin.define.DecoderSpecificConstants;
25import com.android.inputmethod.latin.settings.SpacingAndPunctuations;
26import com.android.inputmethod.latin.utils.DistracterFilter.HandlingType;
27
28import java.util.ArrayList;
29import java.util.List;
30import java.util.Locale;
31
32// Note: this class is used as a parameter type of a native method. You should be careful when you
33// rename this class or field name. See BinaryDictionary#addMultipleDictionaryEntriesNative().
34public final class WordInputEventForPersonalization {
35    private static final String TAG = WordInputEventForPersonalization.class.getSimpleName();
36    private static final boolean DEBUG_TOKEN = false;
37
38    public final int[] mTargetWord;
39    public final int mPrevWordsCount;
40    public final int[][] mPrevWordArray =
41            new int[DecoderSpecificConstants.MAX_PREV_WORD_COUNT_FOR_N_GRAM][];
42    public final boolean[] mIsPrevWordBeginningOfSentenceArray =
43            new boolean[DecoderSpecificConstants.MAX_PREV_WORD_COUNT_FOR_N_GRAM];
44    public final boolean mIsValid;
45    // Time stamp in seconds.
46    public final int mTimestamp;
47
48    @UsedForTesting
49    public WordInputEventForPersonalization(final CharSequence targetWord,
50            final NgramContext ngramContext, final boolean isValid, final int timestamp) {
51        mTargetWord = StringUtils.toCodePointArray(targetWord);
52        mPrevWordsCount = ngramContext.getPrevWordCount();
53        ngramContext.outputToArray(mPrevWordArray, mIsPrevWordBeginningOfSentenceArray);
54        mIsValid = isValid;
55        mTimestamp = timestamp;
56    }
57
58    // Process a list of words and return a list of {@link WordInputEventForPersonalization}
59    // objects.
60    public static ArrayList<WordInputEventForPersonalization> createInputEventFrom(
61            final List<String> tokens, final int timestamp,
62            final SpacingAndPunctuations spacingAndPunctuations, final Locale locale,
63            final DistracterFilter distracterFilter) {
64        final ArrayList<WordInputEventForPersonalization> inputEvents = new ArrayList<>();
65        final int N = tokens.size();
66        NgramContext ngramContext = NgramContext.EMPTY_PREV_WORDS_INFO;
67        for (int i = 0; i < N; ++i) {
68            final String tempWord = tokens.get(i);
69            if (StringUtils.isEmptyStringOrWhiteSpaces(tempWord)) {
70                // just skip this token
71                if (DEBUG_TOKEN) {
72                    Log.d(TAG, "--- isEmptyStringOrWhiteSpaces: \"" + tempWord + "\"");
73                }
74                continue;
75            }
76            if (!DictionaryInfoUtils.looksValidForDictionaryInsertion(
77                    tempWord, spacingAndPunctuations)) {
78                if (DEBUG_TOKEN) {
79                    Log.d(TAG, "--- not looksValidForDictionaryInsertion: \""
80                            + tempWord + "\"");
81                }
82                // Sentence terminator found. Split.
83                // TODO: Detect whether the context is beginning-of-sentence.
84                ngramContext = NgramContext.EMPTY_PREV_WORDS_INFO;
85                continue;
86            }
87            if (DEBUG_TOKEN) {
88                Log.d(TAG, "--- word: \"" + tempWord + "\"");
89            }
90            final WordInputEventForPersonalization inputEvent =
91                    detectWhetherVaildWordOrNotAndGetInputEvent(
92                            ngramContext, tempWord, timestamp, locale, distracterFilter);
93            if (inputEvent == null) {
94                continue;
95            }
96            inputEvents.add(inputEvent);
97            ngramContext = ngramContext.getNextNgramContext(new NgramContext.WordInfo(tempWord));
98        }
99        return inputEvents;
100    }
101
102    private static WordInputEventForPersonalization detectWhetherVaildWordOrNotAndGetInputEvent(
103            final NgramContext ngramContext, final String targetWord, final int timestamp,
104            final Locale locale, final DistracterFilter distracterFilter) {
105        if (locale == null) {
106            return null;
107        }
108        final int wordHandlingType = distracterFilter.getWordHandlingType(ngramContext,
109                targetWord, locale);
110        final String word = HandlingType.shouldBeLowerCased(wordHandlingType) ?
111                targetWord.toLowerCase(locale) : targetWord;
112        if (distracterFilter.isDistracterToWordsInDictionaries(ngramContext, targetWord, locale)) {
113            // The word is a distracter.
114            return null;
115        }
116        return new WordInputEventForPersonalization(word, ngramContext,
117                !HandlingType.shouldBeHandledAsOov(wordHandlingType), timestamp);
118    }
119}
120