src/com/android/providers/contacts/HanziToPinyin.java

/*
 * Copyright (C) 2011 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.providers.contacts;

import android.text.TextUtils;
import android.util.Log;

import java.util.ArrayList;
import java.util.Locale;

import libcore.icu.Transliterator;

/**
 * An object to convert Chinese character to its corresponding pinyin string.
 * For characters with multiple possible pinyin string, only one is selected
 * according to ICU Transliterator class. Polyphone is not supported in this
 * implementation.
 */
public class HanziToPinyin {
    private static final String TAG = "HanziToPinyin";

    private static HanziToPinyin sInstance;
    private final Transliterator mPinyinTransliterator;

    public static class Token {
        /**
         * Separator between target string for each source char
         */
        public static final String SEPARATOR = " ";

        public static final int LATIN = 1;
        public static final int PINYIN = 2;
        public static final int UNKNOWN = 3;

        public Token() {
        }

        public Token(int type, String source, String target) {
            this.type = type;
            this.source = source;
            this.target = target;
        }

        /**
         * Type of this token, ASCII, PINYIN or UNKNOWN.
         */
        public int type;
        /**
         * Original string before translation.
         */
        public String source;
        /**
         * Translated string of source. For Han, target is corresponding Pinyin. Otherwise target is
         * original string in source.
         */
        public String target;
    }

    private HanziToPinyin() {
        Transliterator t = null;
        try {
            t = new Transliterator("Han-Latin/Names; Latin-Ascii; Any-Upper");
        } catch (RuntimeException e) {
            Log.w(TAG, "Han-Latin/Names transliterator data is missing,"
                  + " HanziToPinyin is disabled");
        }
        mPinyinTransliterator = t;
    }

    public boolean hasChineseTransliterator() {
        return mPinyinTransliterator != null;
    }

    public static HanziToPinyin getInstance() {
        synchronized (HanziToPinyin.class) {
            if (sInstance == null) {
                sInstance = new HanziToPinyin();
            }
            return sInstance;
        }
    }

    private Token getToken(char character) {
        Token token = new Token();
        token.source = Character.toString(character);

        if (character < 256) {
            token.type = Token.LATIN;
            token.target = token.source;
            return token;
        }

        token.type = Token.PINYIN;
        token.target = mPinyinTransliterator.transliterate(token.source);
        if (TextUtils.isEmpty(token.target) ||
            TextUtils.equals(token.source, token.target)) {
            token.type = Token.UNKNOWN;
            token.target = token.source;
        }
        return token;
    }

    /**
     * Convert the input to a array of tokens. The sequence of ASCII or Unknown characters without
     * space will be put into a Token, One Hanzi character which has pinyin will be treated as a
     * Token. If there is no Chinese transliterator, the empty token array is returned.
     */
    public ArrayList<Token> get(final String input) {
        ArrayList<Token> tokens = new ArrayList<Token>();
        if (!hasChineseTransliterator() || TextUtils.isEmpty(input)) {
            // return empty tokens.
            return tokens;
        }
        final int inputLength = input.length();
        final StringBuilder sb = new StringBuilder();
        int tokenType = Token.LATIN;
        // Go through the input, create a new token when
        // a. Token type changed
        // b. Get the Pinyin of current charater.
        // c. current character is space.
        for (int i = 0; i < inputLength; i++) {
            final char character = input.charAt(i);
            if (character == ' ') {
                if (sb.length() > 0) {
                    addToken(sb, tokens, tokenType);
                }
            } else if (character < 256) {
                if (tokenType != Token.LATIN && sb.length() > 0) {
                    addToken(sb, tokens, tokenType);
                }
                tokenType = Token.LATIN;
                sb.append(character);
            } else {
                Token t = getToken(character);
                if (t.type == Token.PINYIN) {
                    if (sb.length() > 0) {
                        addToken(sb, tokens, tokenType);
                    }
                    tokens.add(t);
                    tokenType = Token.PINYIN;
                } else {
                    if (tokenType != t.type && sb.length() > 0) {
                        addToken(sb, tokens, tokenType);
                    }
                    tokenType = t.type;
                    sb.append(character);
                }
            }
        }
        if (sb.length() > 0) {
            addToken(sb, tokens, tokenType);
        }
        return tokens;
    }

    private void addToken(
            final StringBuilder sb, final ArrayList<Token> tokens, final int tokenType) {
        String str = sb.toString();
        tokens.add(new Token(tokenType, str, str));
        sb.setLength(0);
    }
}