stringutils.java

基于Jabber协议的即时消息服务器

/**
 * $Revision: 5718 $
 * $Date: 2006-10-13 17:18:18 -0700 (Fri, 13 Oct 2006) $
 *
 * Copyright (C) 2004-2006 Jive Software. All rights reserved.
 *
 * This software is published under the terms of the GNU Public License (GPL),
 * a copy of which is included in this distribution.
 */

package org.jivesoftware.util;

import java.io.UnsupportedEncodingException;
import java.security.MessageDigest;
import java.security.NoSuchAlgorithmException;
import java.text.BreakIterator;
import java.util.*;
import java.util.concurrent.ConcurrentHashMap;

/**
 * Utility class to peform common String manipulation algorithms.
 */
public class StringUtils {

    // Constants used by escapeHTMLTags
    private static final char[] QUOTE_ENCODE = """.toCharArray();
    private static final char[] AMP_ENCODE = "&".toCharArray();
    private static final char[] LT_ENCODE = "<".toCharArray();
    private static final char[] GT_ENCODE = ">".toCharArray();

    private StringUtils() {
        // Not instantiable.
    }

    /**
     * Replaces all instances of oldString with newString in string.
     *
     * @param string the String to search to perform replacements on.
     * @param oldString the String that should be replaced by newString.
     * @param newString the String that will replace all instances of oldString.
     * @return a String will all instances of oldString replaced by newString.
     */
    public static String replace(String string, String oldString, String newString) {
        if (string == null) {
            return null;
        }
        int i = 0;
        // Make sure that oldString appears at least once before doing any processing.
        if ((i = string.indexOf(oldString, i)) >= 0) {
            // Use char []'s, as they are more efficient to deal with.
            char[] string2 = string.toCharArray();
            char[] newString2 = newString.toCharArray();
            int oLength = oldString.length();
            StringBuilder buf = new StringBuilder(string2.length);
            buf.append(string2, 0, i).append(newString2);
            i += oLength;
            int j = i;
            // Replace all remaining instances of oldString with newString.
            while ((i = string.indexOf(oldString, i)) > 0) {
                buf.append(string2, j, i - j).append(newString2);
                i += oLength;
                j = i;
            }
            buf.append(string2, j, string2.length - j);
            return buf.toString();
        }
        return string;
    }

    /**
     * Replaces all instances of oldString with newString in line with the
     * added feature that matches of newString in oldString ignore case.
     *
     * @param line      the String to search to perform replacements on
     * @param oldString the String that should be replaced by newString
     * @param newString the String that will replace all instances of oldString
     * @return a String will all instances of oldString replaced by newString
     */
    public static String replaceIgnoreCase(String line, String oldString,
                                                 String newString) {
        if (line == null) {
            return null;
        }
        String lcLine = line.toLowerCase();
        String lcOldString = oldString.toLowerCase();
        int i = 0;
        if ((i = lcLine.indexOf(lcOldString, i)) >= 0) {
            char[] line2 = line.toCharArray();
            char[] newString2 = newString.toCharArray();
            int oLength = oldString.length();
            StringBuilder buf = new StringBuilder(line2.length);
            buf.append(line2, 0, i).append(newString2);
            i += oLength;
            int j = i;
            while ((i = lcLine.indexOf(lcOldString, i)) > 0) {
                buf.append(line2, j, i - j).append(newString2);
                i += oLength;
                j = i;
            }
            buf.append(line2, j, line2.length - j);
            return buf.toString();
        }
        return line;
    }

    /**
     * Replaces all instances of oldString with newString in line with the
     * added feature that matches of newString in oldString ignore case.
     * The count paramater is set to the number of replaces performed.
     *
     * @param line      the String to search to perform replacements on
     * @param oldString the String that should be replaced by newString
     * @param newString the String that will replace all instances of oldString
     * @param count     a value that will be updated with the number of replaces
     *                  performed.
     * @return a String will all instances of oldString replaced by newString
     */
    public static String replaceIgnoreCase(String line, String oldString,
            String newString, int[] count)
    {
        if (line == null) {
            return null;
        }
        String lcLine = line.toLowerCase();
        String lcOldString = oldString.toLowerCase();
        int i = 0;
        if ((i = lcLine.indexOf(lcOldString, i)) >= 0) {
            int counter = 1;
            char[] line2 = line.toCharArray();
            char[] newString2 = newString.toCharArray();
            int oLength = oldString.length();
            StringBuilder buf = new StringBuilder(line2.length);
            buf.append(line2, 0, i).append(newString2);
            i += oLength;
            int j = i;
            while ((i = lcLine.indexOf(lcOldString, i)) > 0) {
                counter++;
                buf.append(line2, j, i - j).append(newString2);
                i += oLength;
                j = i;
            }
            buf.append(line2, j, line2.length - j);
            count[0] = counter;
            return buf.toString();
        }
        return line;
    }

    /**
     * Replaces all instances of oldString with newString in line.
     * The count Integer is updated with number of replaces.
     *
     * @param line the String to search to perform replacements on.
     * @param oldString the String that should be replaced by newString.
     * @param newString the String that will replace all instances of oldString.
     * @return a String will all instances of oldString replaced by newString.
     */
    public static String replace(String line, String oldString,
            String newString, int[] count)
    {
        if (line == null) {
            return null;
        }
        int i = 0;
        if ((i = line.indexOf(oldString, i)) >= 0) {
            int counter = 1;
            char[] line2 = line.toCharArray();
            char[] newString2 = newString.toCharArray();
            int oLength = oldString.length();
            StringBuilder buf = new StringBuilder(line2.length);
            buf.append(line2, 0, i).append(newString2);
            i += oLength;
            int j = i;
            while ((i = line.indexOf(oldString, i)) > 0) {
                counter++;
                buf.append(line2, j, i - j).append(newString2);
                i += oLength;
                j = i;
            }
            buf.append(line2, j, line2.length - j);
            count[0] = counter;
            return buf.toString();
        }
        return line;
    }

    /**
     * This method takes a string and strips out all tags except <br> tags while still leaving
     * the tag body intact.
     *
     * @param in the text to be converted.
     * @return the input string with all tags removed.
     */
    public static String stripTags(String in) {
        if (in == null) {
            return null;
        }
        char ch;
        int i = 0;
        int last = 0;
        char[] input = in.toCharArray();
        int len = input.length;
        StringBuilder out = new StringBuilder((int)(len * 1.3));
        for (; i < len; i++) {
            ch = input[i];
            if (ch > '>') {
            }
            else if (ch == '<') {
                if (i + 3 < len && input[i + 1] == 'b' && input[i + 2] == 'r' && input[i + 3] == '>') {
                    i += 3;
                    continue;
                }
                if (i > last) {
                    out.append(input, last, i - last);
                }
                last = i + 1;
            }
            else if (ch == '>') {
                last = i + 1;
            }
        }
        if (last == 0) {
            return in;
        }
        if (i > last) {
            out.append(input, last, i - last);
        }
        return out.toString();
    }

    /**
     * This method takes a string which may contain HTML tags (ie, &lt;b&gt;,
     * &lt;table&gt;, etc) and converts the '&lt'' and '&gt;' characters to
     * their HTML escape sequences.
     *
     * @param in the text to be converted.
     * @return the input string with the characters '&lt;' and '&gt;' replaced
     *         with their HTML escape sequences.
     */
    public static String escapeHTMLTags(String in) {
        if (in == null) {
            return null;
        }
        char ch;
        int i = 0;
        int last = 0;
        char[] input = in.toCharArray();
        int len = input.length;
        StringBuilder out = new StringBuilder((int)(len * 1.3));
        for (; i < len; i++) {
            ch = input[i];
            if (ch > '>') {
            }
            else if (ch == '<') {
                if (i > last) {
                    out.append(input, last, i - last);
                }
                last = i + 1;
                out.append(LT_ENCODE);
            }
            else if (ch == '>') {
                if (i > last) {
                    out.append(input, last, i - last);
                }
                last = i + 1;
                out.append(GT_ENCODE);
            }
        }
        if (last == 0) {
            return in;
        }
        if (i > last) {
            out.append(input, last, i - last);
        }
        return out.toString();
    }

    /**
     * Used by the hash method.
     */
    private static Map<String, MessageDigest> digests =
            new ConcurrentHashMap<String, MessageDigest>();

    /**
     * Hashes a String using the Md5 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 data the String to compute the hash of.
     * @return a hashed version of the passed-in String
     */
    public static String hash(String data) {
        return hash(data, "MD5");
    }

    /**
     * Hashes a String 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 data the String to compute the hash of.
     * @param algorithm the name of the algorithm requested.
     * @return a hashed version of the passed-in String