stringutils.java

基于Jabber协议的即时消息服务器

     */
    public static String hash(String data, String algorithm) {
        try {
            return hash(data.getBytes("utf-8"), algorithm);
        }
        catch (UnsupportedEncodingException e) {
            Log.error(e);
        }
        return data;
    }

    /**
     * Hashes a byte array using the specified algorithm and returns the result as a
     * String of hexadecimal numbers. This method is synchronized to avoid
     * excessive MessageDigest object creation. If calling this method becomes
     * a bottleneck in your code, you may wish to maintain a pool of
     * MessageDigest objects instead of using this method.
     * <p/>
     * A hash is a one-way function -- that is, given an
     * input, an output is easily computed. However, given the output, the
     * input is almost impossible to compute. This is useful for passwords
     * since we can store the hash and a hacker will then have a very hard time
     * determining the original password.
     * <p/>
     * In Jive, every time a user logs in, we simply
     * take their plain text password, compute the hash, and compare the
     * generated hash to the stored hash. Since it is almost impossible that
     * two passwords will generate the same hash, we know if the user gave us
     * the correct password or not. The only negative to this system is that
     * password recovery is basically impossible. Therefore, a reset password
     * method is used instead.
     *
     * @param bytes the byte array to compute the hash of.
     * @param algorithm the name of the algorithm requested.
     * @return a hashed version of the passed-in String
     */
    public static String hash(byte[] bytes, String algorithm) {
        synchronized (algorithm.intern()) {
            MessageDigest digest = digests.get(algorithm);
            if (digest == null) {
                try {
                    digest = MessageDigest.getInstance(algorithm);
                    digests.put(algorithm, digest);
                }
                catch (NoSuchAlgorithmException nsae) {
                    Log.error("Failed to load the " + algorithm + " MessageDigest. " +
                            "Jive will be unable to function normally.", nsae);
                    return null;
                }
            }
            // Now, compute hash.
            digest.update(bytes);
            return encodeHex(digest.digest());
        }
    }

    /**
     * Turns an array of bytes into a String representing each byte as an
     * unsigned hex number.
     * <p/>
     * Method by Santeri Paavolainen, Helsinki Finland 1996<br>
     * (c) Santeri Paavolainen, Helsinki Finland 1996<br>
     * Distributed under LGPL.
     *
     * @param bytes an array of bytes to convert to a hex-string
     * @return generated hex string
     */
    public static String encodeHex(byte[] bytes) {
        StringBuilder buf = new StringBuilder(bytes.length * 2);
        int i;

        for (i = 0; i < bytes.length; i++) {
            if (((int)bytes[i] & 0xff) < 0x10) {
                buf.append("0");
            }
            buf.append(Long.toString((int)bytes[i] & 0xff, 16));
        }
        return buf.toString();
    }

    /**
     * Turns a hex encoded string into a byte array. It is specifically meant
     * to "reverse" the toHex(byte[]) method.
     *
     * @param hex a hex encoded String to transform into a byte array.
     * @return a byte array representing the hex String[
     */
    public static byte[] decodeHex(String hex) {
        char[] chars = hex.toCharArray();
        byte[] bytes = new byte[chars.length / 2];
        int byteCount = 0;
        for (int i = 0; i < chars.length; i += 2) {
            int newByte = 0x00;
            newByte |= hexCharToByte(chars[i]);
            newByte <<= 4;
            newByte |= hexCharToByte(chars[i + 1]);
            bytes[byteCount] = (byte)newByte;
            byteCount++;
        }
        return bytes;
    }

    /**
     * Returns the the byte value of a hexadecmical char (0-f). It's assumed
     * that the hexidecimal chars are lower case as appropriate.
     *
     * @param ch a hexedicmal character (0-f)
     * @return the byte value of the character (0x00-0x0F)
     */
    private static byte hexCharToByte(char ch) {
        switch (ch) {
            case '0':
                return 0x00;
            case '1':
                return 0x01;
            case '2':
                return 0x02;
            case '3':
                return 0x03;
            case '4':
                return 0x04;
            case '5':
                return 0x05;
            case '6':
                return 0x06;
            case '7':
                return 0x07;
            case '8':
                return 0x08;
            case '9':
                return 0x09;
            case 'a':
                return 0x0A;
            case 'b':
                return 0x0B;
            case 'c':
                return 0x0C;
            case 'd':
                return 0x0D;
            case 'e':
                return 0x0E;
            case 'f':
                return 0x0F;
        }
        return 0x00;
    }

    /**
     * Encodes a String as a base64 String.
     *
     * @param data a String to encode.
     * @return a base64 encoded String.
     */
    public static String encodeBase64(String data) {
        byte[] bytes = null;
        try {
            bytes = data.getBytes("UTF-8");
        }
        catch (UnsupportedEncodingException uee) {
            Log.error(uee);
        }
        return encodeBase64(bytes);
    }

    /**
     * Encodes a byte array into a base64 String.
     *
     * @param data a byte array to encode.
     * @return a base64 encode String.
     */
    public static String encodeBase64(byte[] data) {
        // Encode the String. We pass in a flag to specify that line
        // breaks not be added. This is consistent with our previous base64
        // implementation. Section 2.1 of 3548 (base64 spec) also specifies
        // no line breaks by default.
        return Base64.encodeBytes(data, Base64.DONT_BREAK_LINES);
    }

    /**
     * Decodes a base64 String.
     *
     * @param data a base64 encoded String to decode.
     * @return the decoded String.
     */
    public static byte[] decodeBase64(String data) {
        return Base64.decode(data);
    }

    /**
     * Converts a line of text into an array of lower case words using a
     * BreakIterator.wordInstance().<p>
     *
     * This method is under the Jive Open Source Software License and was
     * written by Mark Imbriaco.
     *
     * @param text a String of text to convert into an array of words
     * @return text broken up into an array of words.
     */
    public static String[] toLowerCaseWordArray(String text) {
        if (text == null || text.length() == 0) {
            return new String[0];
        }

        List<String> wordList = new ArrayList<String>();
        BreakIterator boundary = BreakIterator.getWordInstance();
        boundary.setText(text);
        int start = 0;

        for (int end = boundary.next(); end != BreakIterator.DONE;
             start = end, end = boundary.next()) {
            String tmp = text.substring(start, end).trim();
            // Remove characters that are not needed.
            tmp = replace(tmp, "+", "");
            tmp = replace(tmp, "/", "");
            tmp = replace(tmp, "\\", "");
            tmp = replace(tmp, "#", "");
            tmp = replace(tmp, "*", "");
            tmp = replace(tmp, ")", "");
            tmp = replace(tmp, "(", "");
            tmp = replace(tmp, "&", "");
            if (tmp.length() > 0) {
                wordList.add(tmp);
            }
        }
        return wordList.toArray(new String[wordList.size()]);
    }

    /**
     * Pseudo-random number generator object for use with randomString().
     * The Random class is not considered to be cryptographically secure, so
     * only use these random Strings for low to medium security applications.
     */
    private static Random randGen = new Random();

    /**
     * Array of numbers and letters of mixed case. Numbers appear in the list
     * twice so that there is a more equal chance that a number will be picked.
     * We can use the array to get a random number or letter by picking a random
     * array index.
     */
    private static char[] numbersAndLetters = ("0123456789abcdefghijklmnopqrstuvwxyz" +
            "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ").toCharArray();

    /**
     * Returns a random String of numbers and letters (lower and upper case)
     * of the specified length. The method uses the Random class that is
     * built-in to Java which is suitable for low to medium grade security uses.
     * This means that the output is only pseudo random, i.e., each number is
     * mathematically generated so is not truly random.<p>
     * <p/>
     * The specified length must be at least one. If not, the method will return
     * null.
     *
     * @param length the desired length of the random String to return.
     * @return a random String of numbers and letters of the specified length.
     */
    public static String randomString(int length) {
        if (length < 1) {
            return null;
        }
        // Create a char buffer to put random letters and numbers in.
        char[] randBuffer = new char[length];
        for (int i = 0; i < randBuffer.length; i++) {
            randBuffer[i] = numbersAndLetters[randGen.nextInt(71)];
        }
        return new String(randBuffer);
    }

    /**
     * Intelligently chops a String at a word boundary (whitespace) that occurs
     * at the specified index in the argument or before. However, if there is a
     * newline character before <code>length</code>, the String will be chopped
     * there. If no newline or whitespace is found in <code>string</code> up to
     * the index <code>length</code>, the String will chopped at <code>length</code>.
     * <p/>
     * For example, chopAtWord("This is a nice String", 10) will return
     * "This is a" which is the first word boundary less than or equal to 10
     * characters into the original String.
     *
     * @param string the String to chop.
     * @param length the index in <code>string</code> to start looking for a
     *               whitespace boundary at.
     * @return a substring of <code>string</code> whose length is less than or
     *         equal to <code>length</code>, and that is chopped at whitespace.
     */
    public static String chopAtWord(String string, int length) {
        if (string == null || string.length() == 0) {
            return string;
        }

        char[] charArray = string.toCharArray();
        int sLength = string.length();
        if (length < sLength) {
            sLength = length;
        }

        // First check if there is a newline character before length; if so,
        // chop word there.
        for (int i = 0; i < sLength - 1; i++) {
            // Windows
            if (charArray[i] == '\r' && charArray[i + 1] == '\n') {
                return string.substring(0, i + 1);
            }
            // Unix
            else if (charArray[i] == '\n') {
                return string.substring(0, i);
            }
        }
        // Also check boundary case of Unix newline
        if (charArray[sLength - 1] == '\n') {
            return string.substring(0, sLength - 1);
        }

        // Done checking for newline, now see if the total string is less than
        // the specified chop point.
        if (string.length() < length) {
            return string;
        }

        // No newline, so chop at the first whitespace.
        for (int i = length - 1; i > 0; i--) {
            if (charArray[i] == ' ') {
                return string.substring(0, i).trim();
            }
        }

        // Did not find word boundary so return original String chopped at
        // specified length.
        return string.substring(0, length);
    }

    /**
     * Reformats a string where lines that are longer than <tt>width</tt>
     * are split apart at the earliest wordbreak or at maxLength, whichever is
     * sooner. If the width specified is less than 5 or greater than the input