在Java中使用正则表达式使用不可打印的字符

Question

I'm using regex found here ( link ) to extract domain string that works fine.

the regex is

^((?!-)[A-Za-z0-9-]{1,63}(?<!-)\\.)+[A-Za-z]{2,6}$

I'm wondering, how could I change it in order to match domain which contains a non printable character instead of dot (.) ?

I know that regex code are like \\x01, \\x02, etc.. but if I replace dot with one of them, the regex doesn't match anymore

thanks in advance

Answer 1

Your dot is escaped here.

You need to remove the double-escape ( \\\\ ) and replace the dot with a literal to match it.

You could also just remove the double escape and keep the dot, which would match any character.

Answer 2

. will match any single character regardless of whether it is printable. Your current group [A-Za-z0-9-] restricts it. You could change this to "any character except literal dot"... ie [^.].

Pattern regex = Pattern.compile("^((?!-)[^.]{1,63}(?<!-)\\.)+[A-Za-z]{2,6}$");
System.out.println(regex.matcher("\u0001\u0002\u0003\u0004..com").find()); // => false
System.out.println(regex.matcher("\u0001\u0002\u0003\u0004.com").find()); // => true
System.out.println(regex.matcher("google.com").find()); // => true

If you're attempting to validate user entry of IDNs (international domain names), note note that there are new gTLDs that contain non alphanumeric characters Example .شبكة (.network).