tools/java/org/unicode/cldr/util/Utf8StringByteConverter.java

/*
 **********************************************************************
 * Copyright (c) 2006-2007, Google and others.  All Rights Reserved.
 **********************************************************************
 * Author: Mark Davis
 **********************************************************************
 */
package org.unicode.cldr.util;

import java.io.IOException;

import com.ibm.icu.lang.UCharacter;
import com.ibm.icu.text.UTF16;
import com.ibm.icu.util.ICUUncheckedIOException;

/**
 * @author markdavis
 */
// TODO optimize this
public class Utf8StringByteConverter extends StringByteConverter {
    char lead = 0;

    @Override
    public int toBytes(char ch, byte[] output, int bytePosition) {
        // we may have state, if we were processing a supplemental char
        if (lead != 0) {
            if (UTF16.isTrailSurrogate(ch)) {
                int cp = UCharacter.getCodePoint(lead, ch);
                output[bytePosition++] = (byte) (0xF0 | (cp >>> 18));
                output[bytePosition++] = (byte) (0x80 | ((cp >>> 12) & 0x3F));
                output[bytePosition++] = (byte) (0x80 | ((cp >>> 6) & 0x3F));
                output[bytePosition++] = (byte) (0x80 | (cp & 0x3F));
                lead = 0;
                return bytePosition;
            }
            // write lead
            output[bytePosition++] = (byte) (0xE0 | (lead >>> 12));
            output[bytePosition++] = (byte) (0x80 | ((lead >>> 6) & 0x3F));
            output[bytePosition++] = (byte) (0x80 | (lead & 0x3F));
            lead = 0;
        }
        if (ch < 0x80) {
            output[bytePosition++] = (byte) ch;
        } else if (ch < 0x800) {
            output[bytePosition++] = (byte) (0xC0 | (ch >>> 6));
            output[bytePosition++] = (byte) (0x80 | (ch & 0x3F));
        } else if (ch >= 0xD800 && ch < 0xDC00) {
            lead = ch;
        } else {
            output[bytePosition++] = (byte) (0xE0 | (ch >>> 12));
            output[bytePosition++] = (byte) (0x80 | ((ch >>> 6) & 0x3F));
            output[bytePosition++] = (byte) (0x80 | (ch & 0x3F));
        }
        return bytePosition;
    }

    @Override
    public int toBytes(byte[] output, int bytePosition) {
        if (lead != 0) {
            output[bytePosition++] = (byte) (0xE0 | (lead >>> 12));
            output[bytePosition++] = (byte) (0x80 | ((lead >>> 6) & 0x3F));
            output[bytePosition++] = (byte) (0x80 | (lead & 0x3F));
            lead = 0;
        }
        return bytePosition;
    }

    @Override
    public int getMaxBytesPerChar() {
        return 4;
    }

    @Override
    public Appendable fromBytes(byte[] input, int byteStart, int byteLength,
        Appendable result) {
        try {
            while (byteStart < byteLength) {
                char b = (char) (input[byteStart++] & 0xFF);
                if (b < 0x80) {
                    // fall through
                } else if (b < 0xE0) {
                    b &= 0x1F;
                    b <<= 6;
                    b |= (char) (input[byteStart++] & 0x3F);
                } else if (b < 0xF0) {
                    b &= 0xF;
                    b <<= 6;
                    b |= (char) (input[byteStart++] & 0x3F);
                    b <<= 6;
                    b |= (char) (input[byteStart++] & 0x3F);
                } else {
                    // surrogate
                    int cp = (b & 0x7) << 6;
                    cp |= (char) (input[byteStart++] & 0x3F);
                    cp <<= 6;
                    cp |= (char) (input[byteStart++] & 0x3F);
                    cp <<= 6;
                    cp |= (char) (input[byteStart++] & 0x3F);
                    result.append(UTF16.getLeadSurrogate(cp));
                    b = UTF16.getTrailSurrogate(cp);
                }
                result.append(b);
            }
            return result;
        } catch (IOException e) {
            throw new ICUUncheckedIOException("Internal error", e);
        }
    }
}