支持截取帶HTML代碼樣式的字符串

package me.jeff.util;
import java.util.ArrayList;
import java.util.List;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
/**
* @author http://hi.baidu.com/vnplalvyulin
* @version 1.0
*
* 按字節長度截取字符串(支持截取帶HTML代碼樣式的字符串)
* @param param 將要截取的字符串參數
* @param length 截取的字節長度
* @param end 字符串末尾補上的字符串
* @return 返回截取後的字符串
*/
public class SubStringHTML {

    public String subStringHTML(String param,int length,String end) {
         StringBuffer result = new StringBuffer();
         int n = 0;
         char temp;
         boolean isCode = false; //是不是HTML代碼
         boolean isHTML = false; //是不是HTML特殊字符,如 
         for (int i = 0; i < param.length(); i++) {
           temp = param.charAt(i);
           if (temp == '<') {
             isCode = true;
           }
           else if (temp == '&') {
             isHTML = true;
           }
           else if (temp == '>' && isCode) {
             n = n - 1;
             isCode = false;
           }
           else if (temp == ';' && isHTML) {
             isHTML = false;
           }
           if (!isCode && !isHTML) {
             n = n + 1;
             //UNICODE碼字符佔兩個字節
             if ( (temp + "").getBytes().length > 1) {
               n = n + 1;
             }
           }
           result.append(temp);
           if (n >= length) {
             break;
           }
         }
         result.append(end);
         //取出截取字符串中的HTML標記
         String temp_result = result.toString().replaceAll("(>)[^<>]*(<?)", "$1$2");
         //去掉不需要結素標記的HTML標記
         temp_result = temp_result.replaceAll("</?(AREA|BASE|BASEFONT|BODY|BR|COL|COLGROUP|DD|DT|FRAME|HEAD|HR|HTML|IMG|INPUT|ISINDEX|LI|LINK|META|OPTION|P|PARAM|TBODY|TD|TFOOT|TH|THEAD|TR|area|base|basefont|body|br|col|colgroup|dd|dt|frame|head|hr|html|img|input|isindex|li|link|meta|option|p|param|tbody|td|tfoot|th|thead|tr)[^<>]*/?>",
                                              "");
         //去掉成對的HTML標記
         temp_result=temp_result.replaceAll("<([a-zA-Z]+)[^<>]*>(.*?)</\\1>","$2");
         //用正則表達式取出標記
         Pattern p = Pattern.compile("<([a-zA-Z]+)[^<>]*>");
         Matcher m = p.matcher(temp_result);
         List endHTML = new ArrayList();
         while (m.find()) {
           endHTML.add(m.group(1));
         }
         //補全不成對的HTML標記
         for (int i = endHTML.size() - 1; i >= 0; i--) {
           result.append("</");
           result.append(endHTML.get(i));
           result.append(">");
         }
         return result.toString();
       }
}