Java判断
字符串中是否包含汉字
Java代码
import java.util.regex.Matcher;
import java.util.regex.Pattern;
public class IfHanZi {
public static void main(String[] args) {
//方法一:
String s1 = "我是中国人";
String s2 = "imchinese";
String s3 = "im中国人";
System.out.println(s1 + ":" + new String(s1).length());
System.out.println(s2 + ":" + new String(s2).length());
System.out.println(s3 + ":" + new String(s3).length());
System.out.println((s1.getBytes().length == s1.length()) ? "s1无汉字":"s1有汉字");
System.out.println((s2.getBytes().length == s2.length()) ? "s2无汉字":"s2有汉字");
System.out.println((s3.getBytes().length == s3.length()) ? "s3无汉字":"s3有汉字");
//方法二:
int count = 0;
String regEx = "[\\u4e00-\\u9fa5]";
String str = "中文fd我是中国人as ";
Pattern p = Pattern.compile(regEx);
Matcher m = p.matcher(str);
while (m.find()) {
for (int i = 0; i <= m.groupCount(); i++) {
count = count + 1;
}
}
System.out.println("共有 " + count + "个 ");
}
}