Remove accents from a String

StringUtil.java

package net.andrewspencer.util;

import sun.text.Normalizer; // java.text.Normalizer in 1.6

/** 
 * Here's how to remove accents in Java 1.5. 
 * NB Doesn't separate ligatures.
 */
// WARNING Uses unofficial Sun classes, so JVM-dependent.
// Normalizer became standard in Java 1.6, though.
public final class StringUtil {
    ...
    public static String removeAccents(String notNullSource) {
        // Normalizer.normalise() converts each accented 
        // character into 1 non-accented character followed
        // by 1 or more characters representing the accent(s)
        // alone. These characters representing only
        // an accent belong to the Unicode category 
        // CombiningDiacriticalMarks. The call to replaceAll
        // strips out all characters in that category.
        String normalized = sun.text.Normalizer.normalize(
                notNullSource, 
                sun.text.Normalizer.DECOMP_COMPAT, // Normalizer.Form.NFKD in 1.6
                0);
        return normalized.replaceAll("\\p{InCombiningDiacriticalMarks}+", "");
    }
}

StringUtilTest.java

package net.andrewspencer.util;

import static org.fest.assertions.Assertions.assertThat;
import static net.andrewspencer.util.StringUtil.*;

import org.testng.annotations.Test;

public class StringUtilTest {
    @Test
    public void removeAccentsShouldRemoveAccents() {
        assertThat(removeAccents("ÀÁÂÃÄÅÇÈÉÊËÌÍÎÏÑÒÓÔÕÖÙÚÛÜÝ")).isEqualTo("AAAAAACEEEEIIIINOOOOOUUUUY");
        assertThat(removeAccents("àáâãäåçèéêëìíîïñòóôõöùúûüýÿ")).isEqualTo("aaaaaaceeeeiiiinooooouuuuyy");
    }
    
    @Test
    public void removeAccentsKeepsNonAsciiCharacters() {
        // Naïve versions of the accent stripping routine remove everything except ASCII characters.
        // This strips out characters that look like accented characters, but are actually distinct,
        // albeit non-ASCII, letter characters.  E.g. Ð != D, Ø != O
        assertThat(removeAccents("ÐØø")).isEqualTo("ÐØø");
    }
    
    @Test
    public void removeAccentsDoesNotSplitLigatures() {
        // If it did, then ligatures like "OE" would be split into their constitutive letters, which could be useful.
        // But I haven't found a way to do that.
        
        assertThat(removeAccents("\u0152\u0153\u00DF")).isEqualTo("\u0152\u0153\u00DF"); // ligatures OE, oe, ß (ss)
    }
    
    @Test
    public void capitaliseAndRemoveAccentsOnFirstLetterShouldDoWhatItSaysOnSingleWord() {
        assertThat(capitaliseAndRemoveAccentsOnFirstLetter("ëomer")).isEqualTo("Eomer");
    }

    @Test
    public void capitaliseAndRemoveAccentsOnFirstLetterShouldDoWhatItSaysOnMultipleWords() {
        assertThat(capitaliseAndRemoveAccentsOnFirstLetter("dol amroth")).isEqualTo("Dol Amroth");
    }
    
    @Test
    public void capitaliseAndRemoveAccentsOnFirstLetterShouldSwallowDoubleWhitespaceCharacters() {
        assertThat(capitaliseAndRemoveAccentsOnFirstLetter("minas  tirith")).isEqualTo("Minas Tirith");
    }
    
    @Test
    public void capitaliseAndRemoveAccentsOnFirstLetterShouldNotCapitaliseAfterHyphen() {
        assertThat(capitaliseAndRemoveAccentsOnFirstLetter("Kheled-zâram")).isEqualTo("Kheled-zâram");
    }

}