Java HTML Unescape unescapeHTML(String s)

Description

Turn any HTML escape entities in the string into characters and return the resulting string.

License

Open Source License

Parameter

Parameter	Description
s	String to be unescaped.

Exception

Parameter	Description
NullPointerException	if s is null.

Return

unescaped String.

Declaration

public static String unescapeHTML(String s) 

Method Source Code

//package com.java2s;
/*/*ww w  .  ja va  2s  . c o  m*/
 * Static String formatting and query routines.
 * Copyright (C) 2001,2002 Stephen Ostermiller
 * http://ostermiller.org/contact.pl?regarding=Java+Utilities
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * See COPYING.TXT for details.
 */

import java.util.HashMap;

public class Main {
    private static HashMap htmlEntities = new HashMap();

    /**
     * Turn any HTML escape entities in the string into
     * characters and return the resulting string.
     *
     * @param s String to be unescaped.
     * @return unescaped String.
     * @throws NullPointerException if s is null.
     */
    public static String unescapeHTML(String s) {
        StringBuffer result = new StringBuffer(s.length());
        int ampInd = s.indexOf("&");
        int lastEnd = 0;
        while (ampInd >= 0) {
            int nextAmp = s.indexOf("&", ampInd + 1);
            int nextSemi = s.indexOf(";", ampInd + 1);
            if (nextSemi != -1 && (nextAmp == -1 || nextSemi < nextAmp)) {
                int value = -1;
                String escape = s.substring(ampInd + 1, nextSemi);
                try {
                    if (escape.startsWith("#")) {
                        value = Integer.parseInt(escape.substring(1), 10);
                    } else {
                        if (htmlEntities.containsKey(escape)) {
                            value = ((Integer) (htmlEntities.get(escape))).intValue();
                        }
                    }
                } catch (NumberFormatException x) {
                }
                result.append(s.substring(lastEnd, ampInd));
                lastEnd = nextSemi + 1;
                if (value >= 0 && value <= 0xffff) {
                    result.append((char) value);
                } else {
                    result.append("&").append(escape).append(";");
                }
            }
            ampInd = nextAmp;
        }
        result.append(s.substring(lastEnd));
        return result.toString();
    }
}

Related

1. unescapeHTML(String html)
2. unescapeHtml(String s)
3. unescapeHTML(String s)
4. unescapeHtml(String s)
5. unescapeHTML(String s)
6. unescapeHTML(String source)
7. unescapeHTML(String source)
8. unescapeHTML(String source, int start)
9. unescapeHTML(String str)