Extract everything after second occurence of any kind of capital letter using regex

I need help with building a regex with Java flavor, which extracts everything after second occurence of any kind of capital letter.

For example, for the sample string "98A02D1" the result should be D1. Doesn't matter what comes before the second occurence of a capital letter.

What I came up with is this: (?:.*?[A-Z]){2}(.*)

Which doesn't work and only captures from the third occurence. Lowering the number in curly brackets doesn't help. Appreciate any help!

You can match from the second uppercase letter onwards. Or you can match anything that comes before the second uppercase letter and replace it with blank.

The following regex will match the part that should be removed:

^[^A-Z]*[A-Z][^A-Z]*

Regex Explanation:

• ^: start of string
• [^A-Z]*: any sequence of non-uppercase characters
• [A-Z]: an uppercase character
• [^A-Z]*: any sequence of non-uppercase characters

In your Java code:

String string = "98A02D1";
String pattern = "^[^A-Z]*[A-Z][^A-Z]*";

System.out.println(string.replaceFirst(pattern, ""));

Output:

D1

Check the regex demo here and the java demo here.

Using [A-Z]){2} matches 2 consecutive uppercase chars, which is different from "the second occurrence" as there can be other characters in between in that case.

If there has to be a second uppercase char present, you can use a capture group, and use that in the replacement with $1

If you don't want to cross newlines, you can exclude those in the negated character class.

^[^A-Z\r\n]*[A-Z][^A-Z\r\n]*([A-Z])

Regex demo | Java demo

String pattern = "(?m)^[^A-Z\\r\\n]*[A-Z][^A-Z\\r\\n]*([A-Z].*)";
String string = "98A02D1\n98A02";
System.out.println(string.replaceAll(pattern, "$1"));

Output

D1
98A02

<hr>

An other option could be for example matching any uppercase letter using \p{Lu}

^[^\p{Lu}]*[\p{Lu}][^\p{Lu}\r\n]*(\p{Lu})

Or asserting an uppercase char to the right, and in the replacement use an empty string instead of group 1:

^[^A-Z\r\n]*[A-Z][^A-Z\r\n]*(?=[A-Z])

I was able to match the second occurrence using the following.

(?<=[A-Z]).*?([A-Z].)

• The first group is a positive look-behind, attempting to match any uppercase letter.
• The second group is a capture, for any uppercase letter, followed by a single character.
• Between the two groups is a .*? pattern, which will attempt to match any character, 0 or more.

You can then use the following method, to return the value.

String secondOccurrence(String string) {
    Pattern pattern = Pattern.compile("(?<=[A-Z]).*?([A-Z].)");
    Matcher matcher = pattern.matcher(string);
    if (matcher.find()) 
        return matcher.group(1);
    return null;
}

Alternatively, you could create a for-loop, breaking when the second occurrence is found.

String secondOccurrence(String string) {
    int index = 0;
    boolean first = false;
    for (char character : string.toCharArray()) {
        if (String.valueOf(character).matches("[A-Z]")) {
            if (!first) first = true;
            else break;
        }
        index++;
    }
    return string.substring(index, index + 2);
}