Package javax.xml.bind

Source Code of javax.xml.bind.DatatypeConverterImpl

package javax.xml.bind;

import javax.xml.namespace.QName;
import javax.xml.namespace.NamespaceContext;
import javax.xml.datatype.DatatypeFactory;
import javax.xml.datatype.DatatypeConfigurationException;
import java.math.BigInteger;
import java.math.BigDecimal;
import java.util.Calendar;
import java.util.GregorianCalendar;
import java.util.TimeZone;

/**
* This class is the JAXB RI's default implementation of the
* {@link DatatypeConverterInterface}.
*
* <p>
* When client apps specify the use of the static print/parse
* methods in {@link DatatypeConverter}, it will delegate
* to this class.
*
* <p>
* This class is responsible for whitespace normalization.
*
* @author <ul><li>Ryan Shoemaker, Sun Microsystems, Inc.</li></ul>
* @version $Revision: 1.1 $
* @since JAXB2.1
*/
final class DatatypeConverterImpl implements DatatypeConverterInterface {

    /**
     * To avoid re-creating instances, we cache one instance.
     */
    public static final DatatypeConverterInterface theInstance = new DatatypeConverterImpl();

    protected DatatypeConverterImpl() {
    }

    public String parseString(String lexicalXSDString) {
        return lexicalXSDString;
    }

    public BigInteger parseInteger(String lexicalXSDInteger) {
        return _parseInteger(lexicalXSDInteger);
    }

    public static BigInteger _parseInteger(CharSequence s) {
        return new BigInteger(removeOptionalPlus(WhiteSpaceProcessor.trim(s)).toString());
    }

    public String printInteger(BigInteger val) {
        return _printInteger(val);
    }

    public static String _printInteger(BigInteger val) {
        return val.toString();
    }

    public int parseInt(String s) {
        return _parseInt(s);
    }

    /**
     * Faster but less robust String->int conversion.
     *
     * Note that:
     * <ol>
     <li>XML Schema allows '+', but {@link Integer#valueOf(String)} is not.
     *  <li>XML Schema allows leading and trailing (but not in-between) whitespaces..
     *      {@link Integer#valueOf(String)} doesn't allow any.
     * </ol>
     */
    public static int _parseInt(CharSequence s) {
        int len = s.length();
        int sign = 1;

        int r = 0;

        for( int i=0; i<len; i++ ) {
            char ch = s.charAt(i);
            if(WhiteSpaceProcessor.isWhiteSpace(ch)) {
                // skip whitespace
            } else
            if('0'<=ch && ch<='9') {
                r = r*10 + (ch-'0');
            } else
            if(ch=='-') {
                sign = -1;
            } else
            if(ch=='+') {
                // noop
            } else
                throw new NumberFormatException("Not a number: "+s);
        }

        return r*sign;
    }

    public long parseLong(String lexicalXSLong) {
        return _parseLong(lexicalXSLong);
    }

    public static long _parseLong(CharSequence s) {
        return Long.valueOf(removeOptionalPlus(WhiteSpaceProcessor.trim(s)).toString());
    }

    public short parseShort(String lexicalXSDShort) {
        return _parseShort(lexicalXSDShort);
    }

    public static short _parseShort(CharSequence s) {
        return (short)_parseInt(s);
    }

    public String printShort(short val) {
        return _printShort(val);
    }

    public static String _printShort(short val) {
        return String.valueOf(val);
    }

    public BigDecimal parseDecimal(String content) {
        return _parseDecimal(content);
    }
    public static BigDecimal _parseDecimal(CharSequence content) {
        content = WhiteSpaceProcessor.trim(content);

        return new BigDecimal(content.toString());

        // from purely XML Schema perspective,
        // this implementation has a problem, since
        // in xs:decimal "1.0" and "1" is equal whereas the above
        // code will return different values for those two forms.
        //
        // the code was originally using com.sun.msv.datatype.xsd.NumberType.load,
        // but a profiling showed that the process of normalizing "1.0" into "1"
        // could take non-trivial time.
        //
        // also, from the user's point of view, one might be surprised if
        // 1 (not 1.0) is returned from "1.000"
    }

    public float parseFloat(String lexicalXSDFloat) {
        return _parseFloat(lexicalXSDFloat);
    }

    public static float _parseFloat( CharSequence _val ) {
        String s = WhiteSpaceProcessor.trim(_val).toString();
        /* Incompatibilities of XML Schema's float "xfloat" and Java's float "jfloat"

            * jfloat.valueOf ignores leading and trailing whitespaces,
              whereas this is not allowed in xfloat.
            * jfloat.valueOf allows "float type suffix" (f, F) to be
              appended after float literal (e.g., 1.52e-2f), whereare
              this is not the case of xfloat.

            gray zone
            ---------
            * jfloat allows ".523". And there is no clear statement that mentions
              this case in xfloat. Although probably this is allowed.
            *
        */

        if(s.equals("NaN"))         return Float.NaN;
        if(s.equals("INF"))         return Float.POSITIVE_INFINITY;
        if(s.equals("-INF"))        return Float.NEGATIVE_INFINITY;

        if(s.length()==0
        || !isDigitOrPeriodOrSign(s.charAt(0))
        || !isDigitOrPeriodOrSign(s.charAt(s.length()-1)) )
            throw new NumberFormatException();

        // these screening process is necessary due to the wobble of Float.valueOf method
        return Float.parseFloat(s);
    }

    public String printFloat(float v) {
        return _printFloat(v);
    }

    public static String _printFloat(float v) {
        if( v==Float.NaN )                  return "NaN";
        if( v==Float.POSITIVE_INFINITY )    return "INF";
        if( v==Float.NEGATIVE_INFINITY )    return "-INF";
        return String.valueOf(v);
    }



    public double parseDouble(String lexicalXSDDouble) {
        return _parseDouble(lexicalXSDDouble);
    }

    public static double _parseDouble( CharSequence _val ) {
        String val = WhiteSpaceProcessor.trim(_val).toString();

        if(val.equals("NaN"))    return Double.NaN;
        if(val.equals("INF"))    return Double.POSITIVE_INFINITY;
        if(val.equals("-INF"))    return Double.NEGATIVE_INFINITY;

        if(val.length()==0
        || !isDigitOrPeriodOrSign(val.charAt(0))
        || !isDigitOrPeriodOrSign(val.charAt(val.length()-1)) )
            throw new NumberFormatException(val);


        // these screening process is necessary due to the wobble of Float.valueOf method
        return Double.parseDouble(val);
    }

    public boolean parseBoolean(String lexicalXSDBoolean) {
        return _parseBoolean(lexicalXSDBoolean);
    }

    public static boolean _parseBoolean(CharSequence literal) {
        int i=0;
        int len = literal.length();
        char ch;
        do {
            ch = literal.charAt(i++);
        } while(WhiteSpaceProcessor.isWhiteSpace(ch) && i<len);

        // if we are strict about errors, check i==len. and report an error

        if( ch=='t' || ch=='1' )        return true;
        if( ch=='f' || ch=='0' )        return false;
        return false;
    }

    public String printBoolean(boolean val) {
        return val?"true":"false";
    }
    public static String _printBoolean(boolean val) {
        return val?"true":"false";
    }

    public byte parseByte(String lexicalXSDByte) {
        return _parseByte(lexicalXSDByte);
    }

    public static byte _parseByte(CharSequence literal) {
        return (byte)_parseInt(literal);
    }

    public String printByte(byte val) {
        return _printByte(val);
    }

    public static String _printByte(byte val) {
        return String.valueOf(val);
    }

    public QName parseQName(String lexicalXSDQName, NamespaceContext nsc) {
        return _parseQName(lexicalXSDQName,nsc);
    }

    /**
     * @return null if fails to convert.
     */
    public static QName _parseQName(CharSequence text, NamespaceContext nsc) {
        int length = text.length();

        // trim whitespace
        int start=0;
        while(start<length && WhiteSpaceProcessor.isWhiteSpace(text.charAt(start)))
            start++;

        int end = length;
        while(end>start && WhiteSpaceProcessor.isWhiteSpace(text.charAt(end-1)))
            end--;

        if(end==start)
            throw new IllegalArgumentException("input is empty");


        String uri;
        String localPart;
        String prefix;

        // search ':'
        int idx=start+1;    // no point in searching the first char. that's not valid.
        while(idx<end && text.charAt(idx)!=':' )
            idx++;

        if( idx==end ) {
            uri = nsc.getNamespaceURI("");
            localPart = text.subSequence(start,end).toString();
            prefix = "";
        } else {
            // Prefix exists, check everything
            prefix = text.subSequence(start,idx).toString();
            localPart = text.subSequence(idx+1,end).toString();
            uri = nsc.getNamespaceURI(prefix);
            // uri can never be null according to javadoc,
            // but some users reported that there are implementations that return null.
            if(uri==null || uri.length()==0) // crap. the NamespaceContext interface is broken.
                // error: unbound prefix
                throw new IllegalArgumentException("prefix "+prefix+" is not bound to a namespace");
        }

        return new QName(uri,localPart,prefix);
    }

    public Calendar parseDateTime(String lexicalXSDDateTime) {
        return _parseDateTime(lexicalXSDDateTime);
    }

    public static GregorianCalendar _parseDateTime(CharSequence s) {
        String val = WhiteSpaceProcessor.trim(s).toString();
        return datatypeFactory.newXMLGregorianCalendar(val).toGregorianCalendar();
    }

    public String printDateTime(Calendar val) {
        return _printDateTime(val);
    }

    public static String _printDateTime(Calendar val) {
        return CalendarFormatter.doFormat("%Y-%M-%DT%h:%m:%s%z",val);
    }

    public byte[] parseBase64Binary(String lexicalXSDBase64Binary) {
        return _parseBase64Binary(lexicalXSDBase64Binary);
    }


    public byte[] parseHexBinary(String s) {
        final int len = s.length();

        // "111" is not a valid hex encoding.
        if( len%2 != 0 )
            throw new IllegalArgumentException("hexBinary needs to be even-length: "+s);

        byte[] out = new byte[len/2];

        for( int i=0; i<len; i+=2 ) {
            int h = hexToBin(s.charAt());
            int l = hexToBin(s.charAt(i+1));
            if( h==-1 || l==-1 )
                throw new IllegalArgumentException("contains illegal character for hexBinary: "+s);

            out[i/2] = (byte)(h*16+l);
        }

        return out;
    }

    private static int hexToBin( char ch ) {
        if( '0'<=ch && ch<='9' )    return ch-'0';
        if( 'A'<=ch && ch<='F' )    return ch-'A'+10;
        if( 'a'<=ch && ch<='f' )    return ch-'a'+10;
        return -1;
    }

    private static final char[] hexCode = "0123456789ABCDEF".toCharArray();

    public String printHexBinary(byte[] data) {
        StringBuilder r = new StringBuilder(data.length*2);
        for ( byte b : data) {
            r.append(hexCode[(b >> 4) & 0xF]);
            r.append(hexCode[(b & 0xF)]);
        }
        return r.toString();
    }


    public long parseUnsignedInt(String lexicalXSDUnsignedInt) {
        return _parseLong(lexicalXSDUnsignedInt);
    }

    public String printUnsignedInt(long val) {
        return _printLong(val);
    }

    public int parseUnsignedShort(String lexicalXSDUnsignedShort) {
        return _parseInt(lexicalXSDUnsignedShort);
    }

    public Calendar parseTime(String lexicalXSDTime) {
        return datatypeFactory.newXMLGregorianCalendar(lexicalXSDTime).toGregorianCalendar();
    }

    public String printTime(Calendar val) {
        return CalendarFormatter.doFormat("%h:%m:%s%z",val);
    }

    public Calendar parseDate(String lexicalXSDDate) {
        return datatypeFactory.newXMLGregorianCalendar(lexicalXSDDate).toGregorianCalendar();
    }

    public String printDate(Calendar val) {

        return CalendarFormatter.doFormat((new StringBuilder("%Y-%M-%D").append("%z")).toString(),val);
    }

    public String parseAnySimpleType(String lexicalXSDAnySimpleType) {
        return lexicalXSDAnySimpleType;
//        return (String)SimpleURType.theInstance._createValue( lexicalXSDAnySimpleType, null );
    }

    public String printString(String val) {
//        return StringType.theInstance.convertToLexicalValue( val, null );
        return val;
    }


    public String printInt(int val) {
        return _printInt(val);
    }

    public static String _printInt(int val) {
        return String.valueOf(val);
    }

    public String printLong(long val) {
        return _printLong(val);
    }

    public static String _printLong(long val) {
        return String.valueOf(val);
    }

    public String printDecimal(BigDecimal val) {
        return _printDecimal(val);
    }

    public static String _printDecimal(BigDecimal val) {
        return val.toPlainString();
    }

    public String printDouble(double v) {
        return _printDouble(v);
    }

    public static String _printDouble(double v) {
        if( v==Double.NaN )                    return "NaN";
        if( v==Double.POSITIVE_INFINITY )    return "INF";
        if( v==Double.NEGATIVE_INFINITY )    return "-INF";
        return String.valueOf(v);
    }

    public String printQName(QName val, NamespaceContext nsc) {
        return _printQName(val,nsc);
    }

    public static String _printQName(QName val, NamespaceContext nsc) {
        // Double-check
        String qname;
        String prefix = nsc.getPrefix( val.getNamespaceURI() );
        String localPart = val.getLocalPart();

        if( prefix == null || prefix.length()==0 ) { // be defensive
            qname = localPart;
        } else {
            qname = prefix + ':' + localPart;
        }

        return qname;
    }

    public String printBase64Binary(byte[] val) {
        return _printBase64Binary(val);
    }

    public String printUnsignedShort(int val) {
        return String.valueOf(val);
    }

    public String printAnySimpleType(String val) {
        return val;
    }


    /**
     * Just return the string passed as a parameter but
     * installs an instance of this class as the DatatypeConverter
     * implementation. Used from static fixed value initializers.
     */
    public static String installHook( String s ) {
        DatatypeConverter.setDatatypeConverter(theInstance);
        return s;
    }




// base64 decoder
//====================================

    private static final byte[] decodeMap = initDecodeMap();
    private static final byte PADDING = 127;

    private static byte[] initDecodeMap() {
        byte[] map = new byte[128];
        int i;
        for( i=0; i<128; i++ )        map[i] = -1;

        for( i='A'; i<='Z'; i++ )    map[i] = (byte)(i-'A');
        for( i='a'; i<='z'; i++ )    map[i] = (byte)(i-'a'+26);
        for( i='0'; i<='9'; i++ )    map[i] = (byte)(i-'0'+52);
        map['+'] = 62;
        map['/'] = 63;
        map['='] = PADDING;

        return map;
    }

    /**
     * computes the length of binary data speculatively.
     *
     * <p>
     * Our requirement is to create byte[] of the exact length to store the binary data.
     * If we do this in a straight-forward way, it takes two passes over the data.
     * Experiments show that this is a non-trivial overhead (35% or so is spent on
     * the first pass in calculating the length.)
     *
     * <p>
     * So the approach here is that we compute the length speculatively, without looking
     * at the whole contents. The obtained speculative value is never less than the
     * actual length of the binary data, but it may be bigger. So if the speculation
     * goes wrong, we'll pay the cost of reallocation and buffer copying.
     *
     * <p>
     * If the base64 text is tightly packed with no indentation nor illegal char
     * (like what most web services produce), then the speculation of this method
     * will be correct, so we get the performance benefit.
     */
    private static int guessLength( String text ) {
        final int len = text.length();

        // compute the tail '=' chars
        int j=len-1;
        for(; j>=0; j-- ) {
            byte code = decodeMap[text.charAt(j)];
            if(code==PADDING)
                continue;
            if(code==-1)
                // most likely this base64 text is indented. go with the upper bound
                return text.length()/4*3;
            break;
        }

        j++;    // text.charAt(j) is now at some base64 char, so +1 to make it the size
        int padSize = len-j;
        if(padSize >2) // something is wrong with base64. be safe and go with the upper bound
            return text.length()/4*3;

        // so far this base64 looks like it's unindented tightly packed base64.
        // take a chance and create an array with the expected size
        return text.length()/4*3-padSize;
    }

    /**
     * @param text
     *      base64Binary data is likely to be long, and decoding requires
     *      each character to be accessed twice (once for counting length, another
     *      for decoding.)
     *
     *      A benchmark showed that taking {@link String} is faster, presumably
     *      because JIT can inline a lot of string access (with data of 1K chars, it was twice as fast)
     */
    public static byte[] _parseBase64Binary(String text) {
        final int buflen = guessLength(text);
        final byte[] out = new byte[buflen];
        int o=0;

        final int len = text.length();
        int i;

        final byte[] quadruplet = new byte[4];
        int q=0;

        // convert each quadruplet to three bytes.
        for( i=0; i<len; i++ ) {
            char ch = text.charAt(i);
            byte v = decodeMap[ch];

            if( v!=-1 )
                quadruplet[q++] = v;

            if(q==4) {
                // quadruplet is now filled.
                out[o++] = (byte)((quadruplet[0]<<2)|(quadruplet[1]>>4));
                if( quadruplet[2]!=PADDING )
                    out[o++] = (byte)((quadruplet[1]<<4)|(quadruplet[2]>>2));
                if( quadruplet[3]!=PADDING )
                    out[o++] = (byte)((quadruplet[2]<<6)|(quadruplet[3]));
                q=0;
            }
        }

        if(buflen==o) // speculation worked out to be OK
            return out;

        // we overestimated, so need to create a new buffer
        byte[] nb = new byte[o];
        System.arraycopy(out,0,nb,0,o);
        return nb;
    }

    private static final char[] encodeMap = initEncodeMap();

    private static char[] initEncodeMap() {
        char[] map = new char[64];
        int i;
        for( i= 0; i<26; i++ )        map[i] = (char)('A'+i);
        for( i=26; i<52; i++ )        map[i] = (char)('a'+(i-26));
        for( i=52; i<62; i++ )        map[i] = (char)('0'+(i-52));
        map[62] = '+';
        map[63] = '/';

        return map;
    }

    public static char encode( int i ) {
        return encodeMap[i&0x3F];
    }

    public static byte encodeByte( int i ) {
        return (byte)encodeMap[i&0x3F];
    }

    public static String _printBase64Binary(byte[] input) {
        return _printBase64Binary(input, 0, input.length);
    }
    public static String _printBase64Binary(byte[] input, int offset, int len) {
        char[] buf = new char[((len+2)/3)*4];
        int ptr = _printBase64Binary(input,offset,len,buf,0);
        assert ptr==buf.length;
        return new String(buf);
    }

    /**
     * Encodes a byte array into a char array by doing base64 encoding.
     *
     * The caller must supply a big enough buffer.
     *
     * @return
     *      the value of {@code ptr+((len+2)/3)*4}, which is the new offset
     *      in the output buffer where the further bytes should be placed.
     */
    public static int _printBase64Binary(byte[] input, int offset, int len, char[] buf, int ptr) {
        for( int i=offset; i<len; i+=3 ) {
            switch( len-i ) {
            case 1:
                buf[ptr++] = encode(input[i]>>2);
                buf[ptr++] = encode(((input[i])&0x3)<<4);
                buf[ptr++] = '=';
                buf[ptr++] = '=';
                break;
            case 2:
                buf[ptr++] = encode(input[i]>>2);
                buf[ptr++] = encode(
                            ((input[i]&0x3)<<4) |
                            ((input[i+1]>>4)&0xF));
                buf[ptr++] = encode((input[i+1]&0xF)<<2);
                buf[ptr++] = '=';
                break;
            default:
                buf[ptr++] = encode(input[i]>>2);
                buf[ptr++] = encode(
                            ((input[i]&0x3)<<4) |
                            ((input[i+1]>>4)&0xF));
                buf[ptr++] = encode(
                            ((input[i+1]&0xF)<<2)|
                            ((input[i+2]>>6)&0x3));
                buf[ptr++] = encode(input[i+2]&0x3F);
                break;
            }
        }
        return ptr;
    }

    /**
     * Encodes a byte array into another byte array by first doing base64 encoding
     * then encoding the result in ASCII.
     *
     * The caller must supply a big enough buffer.
     *
     * @return
     *      the value of {@code ptr+((len+2)/3)*4}, which is the new offset
     *      in the output buffer where the further bytes should be placed.
     */
    public static int _printBase64Binary(byte[] input, int offset, int len, byte[] out, int ptr) {
        byte[] buf = out;
        int max = len+offset;
        for( int i=offset; i<max; i+=3 ) {
            switch( max-i ) {
            case 1:
                buf[ptr++] = encodeByte(input[i]>>2);
                buf[ptr++] = encodeByte(((input[i])&0x3)<<4);
                buf[ptr++] = '=';
                buf[ptr++] = '=';
                break;
            case 2:
                buf[ptr++] = encodeByte(input[i]>>2);
                buf[ptr++] = encodeByte(
                            ((input[i]&0x3)<<4) |
                            ((input[i+1]>>4)&0xF));
                buf[ptr++] = encodeByte((input[i+1]&0xF)<<2);
                buf[ptr++] = '=';
                break;
            default:
                buf[ptr++] = encodeByte(input[i]>>2);
                buf[ptr++] = encodeByte(
                            ((input[i]&0x3)<<4) |
                            ((input[i+1]>>4)&0xF));
                buf[ptr++] = encodeByte(
                            ((input[i+1]&0xF)<<2)|
                            ((input[i+2]>>6)&0x3));
                buf[ptr++] = encodeByte(input[i+2]&0x3F);
                break;
            }
        }

        return ptr;
    }

    private static CharSequence removeOptionalPlus(CharSequence s) {
        int len = s.length();

        if(len<=1 || s.charAt(0)!='+')    return s;

        s = s.subSequence(1,len);
        char ch = s.charAt(0);
        if('0'<=ch && ch<='9')    return s;
        if('.'==ch )    return s;

        throw new NumberFormatException();
    }

    private static boolean isDigitOrPeriodOrSign( char ch ) {
        if( '0'<=ch && ch<='9' )    return true;
        if( ch=='+' || ch=='-' || ch=='.' )    return true;
        return false;
    }

    private static final DatatypeFactory datatypeFactory;

    static {
        try {
            datatypeFactory = DatatypeFactory.newInstance();
        } catch (DatatypeConfigurationException e) {
            throw new Error(e);
        }
    }


    private static final class CalendarFormatter {
        public static String doFormat( String format, Calendar cal ) throws IllegalArgumentException {
            int fidx = 0;
            int flen = format.length();
            StringBuilder buf = new StringBuilder();

            while(fidx<flen) {
                char fch = format.charAt(fidx++);

                if(fch!='%') {  // not a meta character
                    buf.append(fch);
                    continue;
                }

                // seen meta character. we don't do error check against the format
                switch (format.charAt(fidx++)) {
                case 'Y' : // year
                    formatYear(cal, buf);
                    break;

                case 'M' : // month
                    formatMonth(cal, buf);
                    break;

                case 'D' : // days
                    formatDays(cal, buf);
                    break;

                case 'h' : // hours
                    formatHours(cal, buf);
                    break;

                case 'm' : // minutes
                    formatMinutes(cal, buf);
                    break;

                case 's' : // parse seconds.
                    formatSeconds(cal, buf);
                    break;

                case 'z' : // time zone
                    formatTimeZone(cal,buf);
                    break;

                default :
                    // illegal meta character. impossible.
                    throw new InternalError();
                }
            }

            return buf.toString();
        }


        private static void formatYear(Calendar cal, StringBuilder buf) {
            int year = cal.get(Calendar.YEAR);

            String s;
            if (year <= 0) // negative value
                s = Integer.toString(1 - year);
            else // positive value
                s = Integer.toString(year);

            while (s.length() < 4)
                s = '0' + s;
            if (year <= 0)
                s = '-' + s;

            buf.append(s);
        }

        private static void formatMonth(Calendar cal, StringBuilder buf) {
            formatTwoDigits(cal.get(Calendar.MONTH)+1,buf);
        }

        private static void formatDays(Calendar cal, StringBuilder buf) {
            formatTwoDigits(cal.get(Calendar.DAY_OF_MONTH),buf);
        }

        private static void formatHours(Calendar cal, StringBuilder buf) {
            formatTwoDigits(cal.get(Calendar.HOUR_OF_DAY),buf);
        }

        private static void formatMinutes(Calendar cal, StringBuilder buf) {
            formatTwoDigits(cal.get(Calendar.MINUTE),buf);
        }

        private static void formatSeconds(Calendar cal, StringBuilder buf) {
            formatTwoDigits(cal.get(Calendar.SECOND),buf);
            if (cal.isSet(Calendar.MILLISECOND)) { // milliseconds
                int n = cal.get(Calendar.MILLISECOND);
                if(n!=0) {
                    String ms = Integer.toString(n);
                    while (ms.length() < 3)
                        ms = '0' + ms; // left 0 paddings.

                    buf.append('.');
                    buf.append(ms);
                }
            }
        }

        /** formats time zone specifier. */
        private static void formatTimeZone(Calendar cal,StringBuilder buf) {
            TimeZone tz = cal.getTimeZone();

            if (tz == null)      return;

            // otherwise print out normally.
            int offset;
            if (tz.inDaylightTime(cal.getTime())) {
                offset = tz.getRawOffset() + (tz.useDaylightTime()?3600000:0);
            } else {
                offset = tz.getRawOffset();
            }

            if(offset==0) {
                buf.append('Z');
                return;
            }

            if (offset >= 0)
                buf.append('+');
            else {
                buf.append('-');
                offset *= -1;
            }

            offset /= 60 * 1000; // offset is in milli-seconds

            formatTwoDigits(offset / 60, buf);
            buf.append(':');
            formatTwoDigits(offset % 60, buf);
        }

        /** formats Integer into two-character-wide string. */
        private static void formatTwoDigits(int n,StringBuilder buf) {
            // n is always non-negative.
            if (n < 10) buf.append('0');
            buf.append(n);
        }
    }
}
TOP

Related Classes of javax.xml.bind.DatatypeConverterImpl

TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.