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

Example 16 with UnicodeSet

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

the class TransliteratorTest method TestToRules.

@Test
public void TestToRules() {
    String RBT = "rbt";
    String SET = "set";
    String[] DATA = { RBT, "$a=\\u4E61; [$a] > A;", "[\\u4E61] > A;", RBT, "$white=[[:Zs:][:Zl:]]; $white{a} > A;", "[[:Zs:][:Zl:]]{a} > A;", SET, "[[:Zs:][:Zl:]]", "[[:Zs:][:Zl:]]", SET, "[:Ps:]", "[:Ps:]", SET, "[:L:]", "[:L:]", SET, "[[:L:]-[A]]", "[[:L:]-[A]]", SET, "[~[:Lu:][:Ll:]]", "[~[:Lu:][:Ll:]]", SET, "[~[a-z]]", "[~[a-z]]", RBT, "$white=[:Zs:]; $black=[^$white]; $black{a} > A;", "[^[:Zs:]]{a} > A;", RBT, "$a=[:Zs:]; $b=[[a-z]-$a]; $b{a} > A;", "[[a-z]-[:Zs:]]{a} > A;", RBT, "$a=[:Zs:]; $b=[$a&[a-z]]; $b{a} > A;", "[[:Zs:]&[a-z]]{a} > A;", RBT, "$a=[:Zs:]; $b=[x$a]; $b{a} > A;", "[x[:Zs:]]{a} > A;", RBT, "$accentMinus = [ [\\u0300-\\u0345] & [:M:] - [\\u0338]] ;" + "$macron = \\u0304 ;" + "$evowel = [aeiouyAEIOUY] ;" + "$iotasub = \\u0345 ;" + "($evowel $macron $accentMinus *) i > | $1 $iotasub ;", "([AEIOUYaeiouy]\\u0304[[\\u0300-\\u0345]&[:M:]-[\\u0338]]*)i > | $1 \\u0345;", RBT, "([AEIOUYaeiouy]\\u0304[[:M:]-[\\u0304\\u0345]]*)i > | $1 \\u0345;", "([AEIOUYaeiouy]\\u0304[[:M:]-[\\u0304\\u0345]]*)i > | $1 \\u0345;" };
    for (int d = 0; d < DATA.length; d += 3) {
        if (DATA[d] == RBT) {
            // Transliterator test
            Transliterator t = Transliterator.createFromRules("ID", DATA[d + 1], Transliterator.FORWARD);
            if (t == null) {
                errln("FAIL: createFromRules failed");
                return;
            }
            String rules, escapedRules;
            rules = t.toRules(false);
            escapedRules = t.toRules(true);
            String expRules = Utility.unescape(DATA[d + 2]);
            String expEscapedRules = DATA[d + 2];
            if (rules.equals(expRules)) {
                logln("Ok: " + DATA[d + 1] + " => " + Utility.escape(rules));
            } else {
                errln("FAIL: " + DATA[d + 1] + " => " + Utility.escape(rules + ", exp " + expRules));
            }
            if (escapedRules.equals(expEscapedRules)) {
                logln("Ok: " + DATA[d + 1] + " => " + escapedRules);
            } else {
                errln("FAIL: " + DATA[d + 1] + " => " + escapedRules + ", exp " + expEscapedRules);
            }
        } else {
            // UnicodeSet test
            String pat = DATA[d + 1];
            String expToPat = DATA[d + 2];
            UnicodeSet set = new UnicodeSet(pat);
            // Adjust spacing etc. as necessary.
            String toPat;
            toPat = set.toPattern(true);
            if (expToPat.equals(toPat)) {
                logln("Ok: " + pat + " => " + toPat);
            } else {
                errln("FAIL: " + pat + " => " + Utility.escape(toPat) + ", exp " + Utility.escape(pat));
            }
        }
    }
}

Example 17 with UnicodeSet

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

the class RoundTripTest method getRepresentativeBoundaryHangul.

private static UnicodeSet getRepresentativeBoundaryHangul() {
    UnicodeSet resultToAddTo = new UnicodeSet();
    // U+1100 HANGUL CHOSEONG KIYEOK
    // U+1161 HANGUL JUNGSEONG A
    UnicodeSet L = new UnicodeSet("[:hst=L:]");
    UnicodeSet V = new UnicodeSet("[:hst=V:]");
    UnicodeSet T = new UnicodeSet("[:hst=T:]");
    String prefixLV = "\u1100\u1161";
    String prefixL = "\u1100";
    String suffixV = "\u1161";
    // HANGUL CHOSEONG IEUNG
    String nullL = "\u110B";
    UnicodeSet L0 = new UnicodeSet("[\u1100\u110B]");
    for (UnicodeSetIterator iL0 = new UnicodeSetIterator(L0); iL0.next(); ) {
        for (UnicodeSetIterator iV = new UnicodeSetIterator(V); iV.next(); ) {
            for (UnicodeSetIterator iV2 = new UnicodeSetIterator(V); iV2.next(); ) {
                String sample = iL0.getString() + iV.getString() + nullL + iV2.getString();
                String trial = Normalizer.compose(sample, false);
                if (trial.length() == 2) {
                    resultToAddTo.add(trial);
                }
            }
        }
    }
    for (UnicodeSetIterator iL = new UnicodeSetIterator(L); iL.next(); ) {
        // do all combinations of "g" + V + L + "a"
        final String suffix = iL.getString() + suffixV;
        for (UnicodeSetIterator iV = new UnicodeSetIterator(V); iV.next(); ) {
            String sample = prefixL + iV.getString() + suffix;
            String trial = Normalizer.compose(sample, false);
            if (trial.length() == 2) {
                resultToAddTo.add(trial);
            }
        }
        // do all combinations of "ga" + T + L + "a"
        for (UnicodeSetIterator iT = new UnicodeSetIterator(T); iT.next(); ) {
            String sample = prefixLV + iT.getString() + suffix;
            String trial = Normalizer.compose(sample, false);
            if (trial.length() == 2) {
                resultToAddTo.add(trial);
            }
        }
    }
    return resultToAddTo;
}

Example 18 with UnicodeSet

use of android.icu.text.UnicodeSet 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 19 with UnicodeSet

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

the class TransliteratorTest method checkRegistry.

private void checkRegistry(String id) {
    Transliterator fie = Transliterator.getInstance(id);
    final UnicodeSet fae = new UnicodeSet("[a-z5]");
    fie.setFilter(fae);
    Transliterator foe = Transliterator.getInstance(id);
    UnicodeFilter fee = foe.getFilter();
    if (fae.equals(fee)) {
        errln("Changed what is in registry for " + id);
    }
}

Example 20 with UnicodeSet

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

the class TransliteratorTest method TestGurmukhiDevanagari.

/**
 * Test Gurmukhi-Devanagari Tippi and Bindi
 */
@Test
public void TestGurmukhiDevanagari() {
    // the rule says:
    // (\u0902) (when preceded by vowel)      --->  (\u0A02)
    // (\u0902) (when preceded by consonant)  --->  (\u0A70)
    UnicodeSet vowel = new UnicodeSet("[\u0905-\u090A \u090F\u0910\u0913\u0914 \u093e-\u0942\u0947\u0948\u094B\u094C\u094D]");
    UnicodeSet non_vowel = new UnicodeSet("[\u0915-\u0928\u092A-\u0930]");
    UnicodeSetIterator vIter = new UnicodeSetIterator(vowel);
    UnicodeSetIterator nvIter = new UnicodeSetIterator(non_vowel);
    Transliterator trans = Transliterator.getInstance("Devanagari-Gurmukhi");
    StringBuffer src = new StringBuffer(" \u0902");
    StringBuffer expect = new StringBuffer(" \u0A02");
    while (vIter.next()) {
        src.setCharAt(0, (char) vIter.codepoint);
        expect.setCharAt(0, (char) (vIter.codepoint + 0x0100));
        expect(trans, src.toString(), expect.toString());
    }
    expect.setCharAt(1, '\u0A70');
    while (nvIter.next()) {
        // src.setCharAt(0,(char) nvIter.codepoint);
        src.setCharAt(0, (char) nvIter.codepoint);
        expect.setCharAt(0, (char) (nvIter.codepoint + 0x0100));
        expect(trans, src.toString(), expect.toString());
    }
}