stringutils.java

来自「mysql的jdbc驱动」· Java 代码 · 共 1,392 行 · 第 1/3 页

	 * @param s
	 *            the string to convert
	 * @param converter
	 *            the converter to reuse
	 * @param encoding
	 *            the character encoding to use
	 * @param serverEncoding
	 *            DOCUMENT ME!
	 * @param parserKnowsUnicode
	 *            DOCUMENT ME!
	 * 
	 * @return byte[] representation of the string
	 * 
	 * @throws SQLException
	 *             if an encoding unsupported by the JVM is supplied.
	 */
	public static final byte[] getBytes(String s,
			SingleByteCharsetConverter converter, String encoding,
			String serverEncoding, boolean parserKnowsUnicode)
			throws SQLException {
		try {
			byte[] b = null;

			if (converter != null) {
				b = converter.toBytes(s);
			} else if (encoding == null) {
				b = s.getBytes();
			} else {
				b = s.getBytes(encoding);

				if (!parserKnowsUnicode && (encoding.equalsIgnoreCase("SJIS") //$NON-NLS-1$
						|| encoding.equalsIgnoreCase("BIG5") //$NON-NLS-1$
				|| encoding.equalsIgnoreCase("GBK"))) { //$NON-NLS-1$

					if (!encoding.equalsIgnoreCase(serverEncoding)) {
						b = escapeEasternUnicodeByteStream(b, s, 0, s.length());
					}
				}
			}

			return b;
		} catch (UnsupportedEncodingException uee) {
			throw new SQLException(Messages.getString("StringUtils.5") //$NON-NLS-1$
					+ encoding + Messages.getString("StringUtils.6"),
					SQLError.SQL_STATE_ILLEGAL_ARGUMENT); //$NON-NLS-1$
		}
	}

	/**
	 * DOCUMENT ME!
	 * 
	 * @param s
	 *            DOCUMENT ME!
	 * @param converter
	 *            DOCUMENT ME!
	 * @param encoding
	 *            DOCUMENT ME!
	 * @param serverEncoding
	 *            DOCUMENT ME!
	 * @param offset
	 *            DOCUMENT ME!
	 * @param length
	 *            DOCUMENT ME!
	 * @param parserKnowsUnicode
	 *            DOCUMENT ME!
	 * 
	 * @return DOCUMENT ME!
	 * 
	 * @throws SQLException
	 *             DOCUMENT ME!
	 */
	public static final byte[] getBytes(String s,
			SingleByteCharsetConverter converter, String encoding,
			String serverEncoding, int offset, int length,
			boolean parserKnowsUnicode) throws SQLException {
		try {
			byte[] b = null;

			if (converter != null) {
				b = converter.toBytes(s, offset, length);
			} else if (encoding == null) {
				byte[] temp = s.substring(offset, offset + length).getBytes();

				length = temp.length;
				
				b = new byte[length];
				System.arraycopy(temp, 0, b, 0, length);
			} else {

				byte[] temp = s.substring(offset, offset + length)
					.getBytes(encoding);

				length = temp.length;
				
				b = new byte[length];
				System.arraycopy(temp, 0, b, 0, length);
				
				if (!parserKnowsUnicode && (encoding.equalsIgnoreCase("SJIS") //$NON-NLS-1$
						|| encoding.equalsIgnoreCase("BIG5") //$NON-NLS-1$
				|| encoding.equalsIgnoreCase("GBK"))) { //$NON-NLS-1$

					if (!encoding.equalsIgnoreCase(serverEncoding)) {
						b = escapeEasternUnicodeByteStream(b, s, offset, length);
					}
				}
			}

			return b;
		} catch (UnsupportedEncodingException uee) {
			throw new SQLException(Messages.getString("StringUtils.10") //$NON-NLS-1$
					+ encoding + Messages.getString("StringUtils.11"),
					SQLError.SQL_STATE_ILLEGAL_ARGUMENT); //$NON-NLS-1$
		}
	}

	/**
	 * Returns the byte[] representation of the given string using given
	 * encoding.
	 * 
	 * @param s
	 *            the string to convert
	 * @param encoding
	 *            the character encoding to use
	 * @param parserKnowsUnicode
	 *            DOCUMENT ME!
	 * 
	 * @return byte[] representation of the string
	 * 
	 * @throws SQLException
	 *             if an encoding unsupported by the JVM is supplied.
	 */
	public static final byte[] getBytes(String s, String encoding,
			String serverEncoding, boolean parserKnowsUnicode)
			throws SQLException {
		try {
			SingleByteCharsetConverter converter = SingleByteCharsetConverter
					.getInstance(encoding, null);

			return getBytes(s, converter, encoding, serverEncoding,
					parserKnowsUnicode);
		} catch (UnsupportedEncodingException uee) {
			throw new SQLException(Messages.getString("StringUtils.0") //$NON-NLS-1$
					+ encoding + Messages.getString("StringUtils.1"),
					SQLError.SQL_STATE_ILLEGAL_ARGUMENT); //$NON-NLS-1$
		}
	}

	public static int getInt(byte[] buf) throws NumberFormatException {
		int base = 10;

		int s = 0;

		/* Skip white space. */
		while (Character.isWhitespace((char) buf[s]) && (s < buf.length)) {
			++s;
		}

		if (s == buf.length) {
			throw new NumberFormatException(new String(buf));
		}

		/* Check for a sign. */
		boolean negative = false;

		if ((char) buf[s] == '-') {
			negative = true;
			++s;
		} else if ((char) buf[s] == '+') {
			++s;
		}

		/* Save the pointer so we can check later if anything happened. */
		int save = s;

		int cutoff = Integer.MAX_VALUE / base;
		int cutlim = (Integer.MAX_VALUE % base);

		if (negative) {
			cutlim++;
		}

		boolean overflow = false;

		int i = 0;

		for (; s < buf.length; s++) {
			char c = (char) buf[s];

			if (Character.isDigit(c)) {
				c -= '0';
			} else if (Character.isLetter(c)) {
				c = (char) (Character.toUpperCase(c) - 'A' + 10);
			} else {
				break;
			}

			if (c >= base) {
				break;
			}

			/* Check for overflow. */
			if ((i > cutoff) || ((i == cutoff) && (c > cutlim))) {
				overflow = true;
			} else {
				i *= base;
				i += c;
			}
		}

		if (s == save) {
			throw new NumberFormatException(new String(buf));
		}

		if (overflow) {
			throw new NumberFormatException(new String(buf));
		}

		/* Return the result of the appropriate sign. */
		return (negative ? (-i) : i);
	}

	public static long getLong(byte[] buf) throws NumberFormatException {
		int base = 10;

		int s = 0;

		/* Skip white space. */
		while (Character.isWhitespace((char) buf[s]) && (s < buf.length)) {
			++s;
		}

		if (s == buf.length) {
			throw new NumberFormatException(new String(buf));
		}

		/* Check for a sign. */
		boolean negative = false;

		if ((char) buf[s] == '-') {
			negative = true;
			++s;
		} else if ((char) buf[s] == '+') {
			++s;
		}

		/* Save the pointer so we can check later if anything happened. */
		int save = s;

		long cutoff = Long.MAX_VALUE / base;
		long cutlim = (int) (Long.MAX_VALUE % base);

		if (negative) {
			cutlim++;
		}

		boolean overflow = false;
		long i = 0;

		for (; s < buf.length; s++) {
			char c = (char) buf[s];

			if (Character.isDigit(c)) {
				c -= '0';
			} else if (Character.isLetter(c)) {
				c = (char) (Character.toUpperCase(c) - 'A' + 10);
			} else {
				break;
			}

			if (c >= base) {
				break;
			}

			/* Check for overflow. */
			if ((i > cutoff) || ((i == cutoff) && (c > cutlim))) {
				overflow = true;
			} else {
				i *= base;
				i += c;
			}
		}

		if (s == save) {
			throw new NumberFormatException(new String(buf));
		}

		if (overflow) {
			throw new NumberFormatException(new String(buf));
		}

		/* Return the result of the appropriate sign. */
		return (negative ? (-i) : i);
	}

	public static short getShort(byte[] buf) throws NumberFormatException {
		short base = 10;

		int s = 0;

		/* Skip white space. */
		while (Character.isWhitespace((char) buf[s]) && (s < buf.length)) {
			++s;
		}

		if (s == buf.length) {
			throw new NumberFormatException(new String(buf));
		}

		/* Check for a sign. */
		boolean negative = false;

		if ((char) buf[s] == '-') {
			negative = true;
			++s;
		} else if ((char) buf[s] == '+') {
			++s;
		}

		/* Save the pointer so we can check later if anything happened. */
		int save = s;

		short cutoff = (short) (Short.MAX_VALUE / base);
		short cutlim = (short) (Short.MAX_VALUE % base);

		if (negative) {
			cutlim++;
		}

		boolean overflow = false;
		short i = 0;

		for (; s < buf.length; s++) {
			char c = (char) buf[s];

			if (Character.isDigit(c)) {
				c -= '0';
			} else if (Character.isLetter(c)) {
				c = (char) (Character.toUpperCase(c) - 'A' + 10);
			} else {
				break;
			}

			if (c >= base) {
				break;
			}

			/* Check for overflow. */
			if ((i > cutoff) || ((i == cutoff) && (c > cutlim))) {
				overflow = true;
			} else {
				i *= base;
				i += c;
			}
		}

		if (s == save) {
			throw new NumberFormatException(new String(buf));
		}

		if (overflow) {
			throw new NumberFormatException(new String(buf));
		}

		/* Return the result of the appropriate sign. */
		return (negative ? (short) -i : (short) i);
	}

	public final static int indexOfIgnoreCase(int startingPosition,
			String searchIn, String searchFor) {
		if ((searchIn == null) || (searchFor == null)
				|| startingPosition > searchIn.length()) {
			return -1;
		}

		int patternLength = searchFor.length();
		int stringLength = searchIn.length();
		int stopSearchingAt = stringLength - patternLength;

		int i = startingPosition;

		if (patternLength == 0) {
			return -1;
		}

		// Brute force string pattern matching
		// Some locales don't follow upper-case rule, so need to check both
		char firstCharOfPatternUc = Character.toUpperCase(searchFor.charAt(0));
		char firstCharOfPatternLc = Character.toLowerCase(searchFor.charAt(0));

		lookForFirstChar: while (true) {
			while ((i < stopSearchingAt)
					&& (Character.toUpperCase(searchIn.charAt(i)) != firstCharOfPatternUc)
					&& Character.toLowerCase(searchIn.charAt(i)) != firstCharOfPatternLc) {
				i++;
			}

			if (i > stopSearchingAt) {
				return -1;
			}

			int j = i + 1;
			int end = (j + patternLength) - 1;

			int k = 1; // start at second char of pattern

			while (j < end) {
				int searchInPos = j++;
				int searchForPos = k++;

				if (Character.toUpperCase(searchIn.charAt(searchInPos)) != Character
						.toUpperCase(searchFor.charAt(searchForPos))) {
					i++;

					// start over
					continue lookForFirstChar;
				}

				// Georgian and Turkish locales don't have same convention, so
				// need to check lowercase
				// too!
				if (Character.toLowerCase(searchIn.charAt(searchInPos)) != Character
						.toLowerCase(searchFor.charAt(searchForPos))) {
					i++;

					// start over
					continue lookForFirstChar;
				}
			}

			return i; // found entire pattern
		}
	}

	/**
	 * DOCUMENT ME!
	 * 
	 * @param searchIn
	 *            DOCUMENT ME!
	 * @param searchFor
	 *            DOCUMENT ME!
	 * 
	 * @return DOCUMENT ME!
	 */
	public final static int indexOfIgnoreCase(String searchIn, String searchFor) {
		return indexOfIgnoreCase(0, searchIn, searchFor);
	}

	public static int indexOfIgnoreCaseRespectMarker(int startAt, String src,
			String target, String marker, String markerCloses,
			boolean allowBackslashEscapes) {
		char contextMarker = Character.MIN_VALUE;
		boolean escaped = false;
		int markerTypeFound = 0;
		int srcLength = src.length();
		int ind = 0;

		for (int i = startAt; i < srcLength; i++) {
			char c = src.charAt(i);

			if (allowBackslashEscapes && c == '\\') {
				escaped = !escaped;
			} else if (c == markerCloses.charAt(markerTypeFound) && !escaped) {
				contextMarker = Character.MIN_VALUE;
			} else if ((ind = marker.indexOf(c)) != -1 && !escaped