Java Unicode unicodeTrim(String s)

Description

unicode Trim

License

Open Source License

Declaration

public static String unicodeTrim(String s) 

Method Source Code

//package com.java2s;

public class Main {
    public static String unicodeTrim(String s) {
        // this is dumb because it looks like there aren't any whitespace
        // characters that need surrogate encoding. But, points for
        // pedantic correctness! It's future-proof or something.
        // String.trim() actually is broken, since there are plenty of
        // non-ASCII whitespace characters.
        final int length = s.length();
        if (length == 0)
            return s;

        int start = 0;
        while (start < length) {
            char c = s.charAt(start);
            if (c == ' ' || c == '\n') {
                start += 1;//  w ww . java2  s .  c  o m
            } else {
                int cp = s.codePointAt(start);
                if (isWhitespace(cp))
                    start += Character.charCount(cp);
                else
                    break;
            }
        }

        int end = length;
        while (end > start) {
            char c = s.charAt(end - 1);
            if (c == ' ' || c == '\n') {
                --end;
            } else {
                int cp;
                int delta;
                if (Character.isLowSurrogate(c)) {
                    cp = s.codePointAt(end - 2);
                    delta = 2;
                } else {
                    cp = s.codePointAt(end - 1);
                    delta = 1;
                }
                if (isWhitespace(cp))
                    end -= delta;
                else
                    break;
            }
        }
        return s.substring(start, end);
    }

    static boolean isWhitespace(int codepoint) {
        switch (codepoint) {
        // try to hit the most common ASCII ones first, then the nonbreaking
        // spaces that Java brokenly leaves out of isWhitespace.
        case ' ':
        case '\n':
        case '\u00A0':
        case '\u2007':
        case '\u202F':
            // this one is the BOM, see
            // http://www.unicode.org/faq/utf_bom.html#BOM
            // we just accept it as a zero-width nonbreaking space.
        case '\uFEFF':
            return true;
        default:
            return Character.isWhitespace(codepoint);
        }
    }
}

Related

1. unicodeHTMLEscape(final String s)
2. unicodePreservingIndex(String str, int index)
3. unicodePreservingSubstring(String str, int begin, int end)
4. unicodeToChar(char[] unicode)
5. unicodeToHTMLUnicodeEntity(final String text)