How to solve SonarQube complaints about my code?

Question

I'd like to know how to get rid of all non-alphabet in a string please.

I have a String s, where many non-alphabetical characters can be inside. For instance, space, dots, slashes, and any other crazy stuff that are not from a, b, c ... z and not from A, B, C ... Z.

I just want to retain those a, b, c ... z and A, B, C ... Z.

Hence, I wrote:

private static String getGoodString(String s) {
    return s.replaceAll("[^a-zA-Z]", "");
}

This is actually working, quite happy.

However, SonarQube is complaining with:

Refactor this code to use a "static final" Pattern.

Replace these character ranges with Unicode-aware character classes.

How may I achieve the same (get ride of any non-alphabet), while making SonarQube very happy at the same time please?

Answer 1

Try using this version:

// inside your class
private static final Pattern p = Pattern.compile("[^\\p{Alpha}]");

private static String getGoodString(String s) {
    return p.matcher(s).replaceAll("");
}

Here I am using a static final Pattern. Also, the regex \p{Alpha} is the Unicode version for matching any letter character.