Examples with Transliterator android.icu.text.Transliterator used on opensource projects

Example 31 with Transliterator

use of android.icu.text.Transliterator in project j2objc by google.

the class TransliteratorTest method TestT5160.

/*
     * Test case for threading problem in NormalizationTransliterator
     * reported by ticket#5160
     */
@Test
public void TestT5160() {
    final String[] testData = { "a", "b", "\u09BE", "A\u0301" };
    final String[] expected = { "a", "b", "\u09BE", "\u00C1" };
    Transliterator translit = Transliterator.getInstance("NFC");
    NormTranslitTask[] tasks = new NormTranslitTask[testData.length];
    for (int i = 0; i < tasks.length; i++) {
        tasks[i] = new NormTranslitTask(translit, testData[i], expected[i]);
    }
    TestUtil.runUntilDone(tasks);
    for (int i = 0; i < tasks.length; i++) {
        if (tasks[i].getErrorMessage() != null) {
            System.out.println("Fail: thread#" + i + " " + tasks[i].getErrorMessage());
            break;
        }
    }
}

Example 32 with Transliterator

use of android.icu.text.Transliterator in project j2objc by google.

the class TransliteratorTest method TestSpecialCases.

@Test
public void TestSpecialCases() {
    for (int i = 0; i < registerRules.length; ++i) {
        Transliterator t = Transliterator.createFromRules(registerRules[i][0], registerRules[i][1], Transliterator.FORWARD);
        DummyFactory.add(registerRules[i][0], t);
    }
    for (int i = 0; i < testCases.length; ++i) {
        String name = testCases[i][0];
        Transliterator t = Transliterator.getInstance(name);
        String id = t.getID();
        String source = testCases[i][1];
        String target = null;
        if (testCases[i].length > 2)
            target = testCases[i][2];
        else if (id.equalsIgnoreCase("NFD"))
            target = android.icu.text.Normalizer.normalize(source, android.icu.text.Normalizer.NFD);
        else if (id.equalsIgnoreCase("NFC"))
            target = android.icu.text.Normalizer.normalize(source, android.icu.text.Normalizer.NFC);
        else if (id.equalsIgnoreCase("NFKD"))
            target = android.icu.text.Normalizer.normalize(source, android.icu.text.Normalizer.NFKD);
        else if (id.equalsIgnoreCase("NFKC"))
            target = android.icu.text.Normalizer.normalize(source, android.icu.text.Normalizer.NFKC);
        else if (id.equalsIgnoreCase("Lower"))
            target = UCharacter.toLowerCase(Locale.US, source);
        else if (id.equalsIgnoreCase("Upper"))
            target = UCharacter.toUpperCase(Locale.US, source);
        expect(t, source, target);
    }
    for (int i = 0; i < registerRules.length; ++i) {
        Transliterator.unregister(registerRules[i][0]);
    }
}

Example 33 with Transliterator

use of android.icu.text.Transliterator in project j2objc by google.

the class TransliteratorTest method TestCompoundFilter.

/**
 * Compound filter semantics were orginially not implemented
 * correctly.  Originally, each component filter f(i) is replaced by
 * f'(i) = f(i) && g, where g is the filter for the compound
 * transliterator.
 *
 * From Mark:
 *
 * Suppose and I have a transliterator X. Internally X is
 * "Greek-Latin; Latin-Cyrillic; Any-Lower". I use a filter [^A].
 *
 * The compound should convert all greek characters (through latin) to
 * cyrillic, then lowercase the result. The filter should say "don't
 * touch 'A' in the original". But because an intermediate result
 * happens to go through "A", the Greek Alpha gets hung up.
 */
@Test
public void TestCompoundFilter() {
    Transliterator t = Transliterator.getInstance("Greek-Latin; Latin-Greek; Lower", Transliterator.FORWARD);
    t.setFilter(new UnicodeSet("[^A]"));
    // Only the 'A' at index 1 should remain unchanged
    expect(t, CharsToUnicodeString("BA\\u039A\\u0391"), CharsToUnicodeString("\\u03b2A\\u03ba\\u03b1"));
}

Example 34 with Transliterator

use of android.icu.text.Transliterator in project j2objc by google.

the class TransliteratorTest method TestJ277.

/**
 * Regression test for bugs found in Greek transliteration.
 */
@Test
public void TestJ277() {
    Transliterator gl = Transliterator.getInstance("Greek-Latin; NFD; [:M:]Remove; NFC");
    char sigma = (char) 0x3C3;
    char upsilon = (char) 0x3C5;
    char nu = (char) 0x3BD;
    // not used char PHI = (char)0x3A6;
    char alpha = (char) 0x3B1;
    // not used char omega = (char)0x3C9;
    // not used char omicron = (char)0x3BF;
    // not used char epsilon = (char)0x3B5;
    // sigma upsilon nu -> syn
    StringBuffer buf = new StringBuffer();
    buf.append(sigma).append(upsilon).append(nu);
    String syn = buf.toString();
    expect(gl, syn, "syn");
    // sigma alpha upsilon nu -> saun
    buf.setLength(0);
    buf.append(sigma).append(alpha).append(upsilon).append(nu);
    String sayn = buf.toString();
    expect(gl, sayn, "saun");
    // Again, using a smaller rule set
    String rules = "$alpha   = \u03B1;" + "$nu      = \u03BD;" + "$sigma   = \u03C3;" + "$ypsilon = \u03C5;" + "$vowel   = [aeiouAEIOU$alpha$ypsilon];" + "s <>           $sigma;" + "a <>           $alpha;" + "u <>  $vowel { $ypsilon;" + "y <>           $ypsilon;" + "n <>           $nu;";
    Transliterator mini = Transliterator.createFromRules("mini", rules, Transliterator.REVERSE);
    expect(mini, syn, "syn");
    expect(mini, sayn, "saun");
// |    // Transliterate the Greek locale data
// |    Locale el("el");
// |    DateFormatSymbols syms(el, status);
// |    if (U_FAILURE(status)) { errln("FAIL: Transliterator constructor failed"); return; }
// |    int32_t i, count;
// |    const UnicodeString* data = syms.getMonths(count);
// |    for (i=0; i<count; ++i) {
// |        if (data[i].length() == 0) {
// |            continue;
// |        }
// |        UnicodeString out(data[i]);
// |        gl->transliterate(out);
// |        bool_t ok = TRUE;
// |        if (data[i].length() >= 2 && out.length() >= 2 &&
// |            u_isupper(data[i].charAt(0)) && u_islower(data[i].charAt(1))) {
// |            if (!(u_isupper(out.charAt(0)) && u_islower(out.charAt(1)))) {
// |                ok = FALSE;
// |            }
// |        }
// |        if (ok) {
// |            logln(prettify(data[i] + " -> " + out));
// |        } else {
// |            errln(UnicodeString("FAIL: ") + prettify(data[i] + " -> " + out));
// |        }
// |    }
}

Example 35 with Transliterator

use of android.icu.text.Transliterator in project j2objc by google.

the class TransliteratorTest method TestHangul.

@Test
public void TestHangul() {
    Transliterator lh = Transliterator.getInstance("Latin-Hangul");
    Transliterator hl = lh.getInverse();
    assertTransform("Transform", "\uCE20", lh, "ch");
    assertTransform("Transform", "\uC544\uB530", lh, hl, "atta", "a-tta");
    assertTransform("Transform", "\uC544\uBE60", lh, hl, "appa", "a-ppa");
    assertTransform("Transform", "\uC544\uC9DC", lh, hl, "ajja", "a-jja");
    assertTransform("Transform", "\uC544\uAE4C", lh, hl, "akka", "a-kka");
    assertTransform("Transform", "\uC544\uC2F8", lh, hl, "assa", "a-ssa");
    assertTransform("Transform", "\uC544\uCC28", lh, hl, "acha", "a-cha");
    assertTransform("Transform", "\uC545\uC0AC", lh, hl, "agsa", "ag-sa");
    assertTransform("Transform", "\uC548\uC790", lh, hl, "anja", "an-ja");
    assertTransform("Transform", "\uC548\uD558", lh, hl, "anha", "an-ha");
    assertTransform("Transform", "\uC54C\uAC00", lh, hl, "alga", "al-ga");
    assertTransform("Transform", "\uC54C\uB9C8", lh, hl, "alma", "al-ma");
    assertTransform("Transform", "\uC54C\uBC14", lh, hl, "alba", "al-ba");
    assertTransform("Transform", "\uC54C\uC0AC", lh, hl, "alsa", "al-sa");
    assertTransform("Transform", "\uC54C\uD0C0", lh, hl, "alta", "al-ta");
    assertTransform("Transform", "\uC54C\uD30C", lh, hl, "alpa", "al-pa");
    assertTransform("Transform", "\uC54C\uD558", lh, hl, "alha", "al-ha");
    assertTransform("Transform", "\uC555\uC0AC", lh, hl, "absa", "ab-sa");
    assertTransform("Transform", "\uC548\uAC00", lh, hl, "anga", "an-ga");
    assertTransform("Transform", "\uC545\uC2F8", lh, hl, "agssa", "ag-ssa");
    assertTransform("Transform", "\uC548\uC9DC", lh, hl, "anjja", "an-jja");
    assertTransform("Transform", "\uC54C\uC2F8", lh, hl, "alssa", "al-ssa");
    assertTransform("Transform", "\uC54C\uB530", lh, hl, "altta", "al-tta");
    assertTransform("Transform", "\uC54C\uBE60", lh, hl, "alppa", "al-ppa");
    assertTransform("Transform", "\uC555\uC2F8", lh, hl, "abssa", "ab-ssa");
    assertTransform("Transform", "\uC546\uCE74", lh, hl, "akkka", "akk-ka");
    assertTransform("Transform", "\uC558\uC0AC", lh, hl, "asssa", "ass-sa");
}