Java按字节数截取字符串，一个中文长度为2

碰到可能会截取汉字的情况，当然是要不能截取出乱码来，就是不能对整个汉字截取一半。如"我ABC汉字d"这个字符串，截取５个字节的时候，应该是"我 ABC"，而截取８个字节的时候，应该是"我ABC汉"，而不应该是"我ABC汉？"，其中"？"为半个汉字，可理解为向前截取

public static String subStr(String str, int subSLength)  
               throws UnsupportedEncodingException{ 
           if (str == null)  
               return "";  
           else{ 
               int tempSubLength = subSLength;//截取字节数
               String subStr = str.substring(0, str.length()<subSLength ? str.length() : subSLength);//截取的子串  
               int subStrByetsL = subStr.getBytes("GBK").length;//截取子串的字节长度 
               //int subStrByetsL = subStr.getBytes().length;//截取子串的字节长度 
               // 说明截取的字符串中包含有汉字  
               while (subStrByetsL > tempSubLength){  
                   int subSLengthTemp = --subSLength;
                   subStr = str.substring(0, subSLengthTemp>str.length() ? str.length() : subSLengthTemp);  
                   subStrByetsL = subStr.getBytes("GBK").length;
                   //subStrByetsL = subStr.getBytes().length;
               }  
               return subStr; 
           }
       }

备注：将字符编码GBK改为UTF-8，则每个中文长度按3个字符计算

以下方法是向后截取字符串

public static String subStr_1(String str, int start, int end)
               throws UnsupportedEncodingException{
         
        if (str == null)  return null;
        String chinese = "\[\\u0391-\\uFFE5\]";
        byte\[\] b = str.getBytes("UTF-8");
         
        String temp = new String(b, start, end);
        String last = getLastStr(temp);
        while(!last.matches(chinese)){
            temp = new String(b, start, ++end);
            last = getLastStr(temp);
        }
 
        return new String(b, start, end);
       }
 
 
public static String getByteStr(String str, int start, int end) throws UnsupportedEncodingException{
        byte\[\] b = str.getBytes("UTF-8");
         
        return new String(b, start, end);
    }