java中文转unicode码

转载地址：http://www.iteye.com/topic/471901
http://blog.csdn.net/zhengjianqi/archive/2008/03/27/2222846.aspx

使用了两种方法进行判断

/**
	 * 把中文转成Unicode码
	 * @param str
	 * @return
	 */
	public String chinaToUnicode(String str){
		String result="";
		for (int i = 0; i < str.length(); i++){
            int chr1 = (char) str.charAt(i);
            if(chr1>=19968&&chr1<=171941){//汉字范围 \u4e00-\u9fa5 (中文)
                result+="\\u" + Integer.toHexString(chr1);
            }else{
            	result+=str.charAt(i);
            }
        }
		return result;
	}

	/**
	 * 判断是否为中文字符
	 * @param c
	 * @return
	 */
	public  boolean isChinese(char c) {
        Character.UnicodeBlock ub = Character.UnicodeBlock.of(c);
        if (ub == Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS
                || ub == Character.UnicodeBlock.CJK_COMPATIBILITY_IDEOGRAPHS
                || ub == Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A
                || ub == Character.UnicodeBlock.GENERAL_PUNCTUATION
                || ub == Character.UnicodeBlock.CJK_SYMBOLS_AND_PUNCTUATION
                || ub == Character.UnicodeBlock.HALFWIDTH_AND_FULLWIDTH_FORMS) {
            return true;
        }
        return false;
    }

static String string2Unicode(String s) {
    try {
      StringBuffer out = new StringBuffer("");
      byte[] bytes = s.getBytes("unicode");
      for (int i = 2; i < bytes.length - 1; i += 2) {
        out.append("u");
        String str = Integer.toHexString(bytes[i + 1] & 0xff);
        for (int j = str.length(); j < 2; j++) {
          out.append("0");
        }
        String str1 = Integer.toHexString(bytes[i] & 0xff);

        out.append(str);
        out.append(str1);
        out.append(" ");
      }
      return out.toString().toUpperCase();
    }
    catch (UnsupportedEncodingException e) {
      e.printStackTrace();
      return null;
    }
  } 

 

static String unicode2String(String unicodeStr){
    StringBuffer sb = new StringBuffer();
    String str[] = unicodeStr.toUpperCase().split("U");
    for(int i=0;i<str.length;i++){
      if(str[i].equals("")) continue;
      char c = (char)Integer.parseInt(str[i].trim(),16);
      sb.append(c);
    }
    return sb.toString();
  }

评论

1 楼 CoderDream 2013-12-04  

/**
* 输出十进制，格式为� 如“中国”输出：中国
*
* @param s
* @return
*/
public static String string2UnicodeNumber(String s) {
try {
StringBuffer out = new StringBuffer("");
StringBuffer temp = null;
StringBuffer number = null;
byte[] bytes = s.getBytes("unicode");
for (int i = 2; i < bytes.length - 1; i += 2) {
temp = new StringBuffer("&#");
number = new StringBuffer("");
String str = Integer.toHexString(bytes[i + 1] & 0xff);
for (int j = str.length(); j < 2; j++) {
temp.append("0");
}
String str1 = Integer.toHexString(bytes[i] & 0xff);
// 十进制转化为十六进制，结果为C8。
// Integer.toHexString(200);
// 十六进制转化为十进制，结果140。
// Integer.parseInt("8C",16);
number.append(str1);
number.append(str);

BigInteger bi = new BigInteger(number.toString(), 16);
String show = bi.toString(10);

temp.append(show + ";");
out.append(temp.toString());
}

return out.toString();
} catch (UnsupportedEncodingException e) {
e.printStackTrace();
return null;
}
}