JAVA判断字符串是否为乱码

   转载，源文件http://blog.csdn.net/jonay/article/details/6644167

 

import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class MessyCodeCheck {

  public static 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;
  }

  public static boolean isMessyCode(String strName) {
    Pattern p = Pattern.compile("\\s*|\t*|\r*|\n*");
    Matcher m = p.matcher(strName);
    String after = m.replaceAll("");
    String temp = after.replaceAll("\\p{P}", "");
    char[] ch = temp.trim().toCharArray();
    float chLength = ch.length;
    float count = 0;
    for (int i = 0; i < ch.length; i++) {
      char c = ch[i];
      if (!Character.isLetterOrDigit(c)) {

        if (!isChinese(c)) {
          count = count + 1;
          System.out.print(c);
        }
      }
    }
    float result = count / chLength;
    if (result > 0.4) {
      return true;
    } else {
      return false;
    }

  }

  public static void main(String[] args) {
    System.out.println(isMessyCode("*��JTP.jar�ļ����JTP�ļ���ȡ��ͼƬ��Դ"));
    System.out.println(isMessyCode("你好"));
  }
}

评论

1 楼 NeverGiveUpToChange 2015-02-26  

%D6%D0%B9%FA%B8%DF%B9%E6%B8%F1
%D6%D0%B9%FA%B8%DF%B9%E6%B8%F1===GB2312-----中国高规格====---false
%E4%B8%AD%E5%9B%BD%E9%AB%98%E8%A7%84%E6%A0%BC
%E4%B8%AD%E5%9B%BD%E9%AB%98%E8%A7%84%E6%A0%BC===GB2312-----涓��楂����====---true
%D6%D0%B9%FA%B8%DF%B9%E6%B8%F1
%D6%D0%B9%FA%B8%DF%B9%E6%B8%F1===GB2312-----中国高规格====---false
%3F%3F%3F%3F%3F
%3F%3F%3F%3F%3F===GB2312-----?????====---false

最后一个乱码他提示不乱码