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

Example 66 with RuleBasedCollator

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

the class CollationRegressionTest method Test4133509.

// @bug 4133509
// 
// The sorting using java.text.CollationKey is not in the exact order
// 
@Test
public void Test4133509() /* char* par */
{
    RuleBasedCollator en_us = (RuleBasedCollator) Collator.getInstance(Locale.US);
    String[] test1 = { "\u0045\u0078\u0063\u0065\u0070\u0074\u0069\u006f\u006e", "\u003c", "\u0045\u0078\u0063\u0065\u0070\u0074\u0069\u006f\u006e\u0049\u006e\u0049\u006e\u0069\u0074\u0069\u0061\u006c\u0069\u007a\u0065\u0072\u0045\u0072\u0072\u006f\u0072", "\u0047\u0072\u0061\u0070\u0068\u0069\u0063\u0073", "\u003c", "\u0047\u0072\u0061\u0070\u0068\u0069\u0063\u0073\u0045\u006e\u0076\u0069\u0072\u006f\u006e\u006d\u0065\u006e\u0074", "\u0053\u0074\u0072\u0069\u006e\u0067", "\u003c", "\u0053\u0074\u0072\u0069\u006e\u0067\u0042\u0075\u0066\u0066\u0065\u0072" };
    compareArray(en_us, test1);
}

Example 67 with RuleBasedCollator

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

the class CollationRegressionTest method Test4078588.

// @bug 4078588
// 
// RuleBasedCollator breaks on "< a < bb" rule
// 
@Test
public void Test4078588() /* char *par */
{
    RuleBasedCollator rbc = null;
    try {
        rbc = new RuleBasedCollator("&9 < a < bb");
    } catch (Exception e) {
        errln("Failed to create RuleBasedCollator.");
        return;
    }
    int result = rbc.compare("a", "bb");
    if (result >= 0) {
        errln("Compare(a,bb) returned " + result + "; expected -1");
    }
}

Example 68 with RuleBasedCollator

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

the class CollationRegressionTest method Test4114077.

// @bug 4114077
// 
// Collation with decomposition off doesn't work for Europe
// 
@Test
public void Test4114077() /* char* par */
{
    // Ensure that we get the same results with decomposition off
    // as we do with it on....
    RuleBasedCollator c = (RuleBasedCollator) Collator.getInstance(Locale.US);
    c.setStrength(Collator.TERTIARY);
    String[] test1 = { // Should be equivalent
    "\u00C0", // Should be equivalent
    "\u003d", // Should be equivalent
    "\u0041\u0300", "\u0070\u00ea\u0063\u0068\u0065", "\u003e", "\u0070\u00e9\u0063\u0068\u00e9", "\u0204", "\u003d", "\u0045\u030F", // a-ring-acute -> a-ring, acute
    "\u01fa", // a-ring-acute -> a-ring, acute
    "\u003d", // a-ring-acute -> a-ring, acute
    "\u0041\u030a\u0301", // No reordering --> unequal
    "\u0041\u0300\u0316", // No reordering --> unequal
    "\u003c", // No reordering --> unequal
    "\u0041\u0316\u0300" };
    c.setDecomposition(Collator.NO_DECOMPOSITION);
    compareArray(c, test1);
    String[] test2 = { // Reordering --> equal
    "\u0041\u0300\u0316", // Reordering --> equal
    "\u003d", // Reordering --> equal
    "\u0041\u0316\u0300" };
    c.setDecomposition(Collator.CANONICAL_DECOMPOSITION);
    compareArray(c, test2);
}

Example 69 with RuleBasedCollator

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

the class CollationIteratorTest method TestUnicodeChar.

/**
 * Test for CollationElementIterator previous and next for the whole set of
 * unicode characters.
 */
@Test
public void TestUnicodeChar() {
    RuleBasedCollator en_us = (RuleBasedCollator) Collator.getInstance(Locale.US);
    CollationElementIterator iter;
    char codepoint;
    StringBuffer source = new StringBuffer();
    source.append("\u0e4d\u0e4e\u0e4f");
    // source.append("\u04e8\u04e9");
    iter = en_us.getCollationElementIterator(source.toString());
    // A basic test to see if it's working at all
    CollationTest.backAndForth(this, iter);
    for (codepoint = 1; codepoint < 0xFFFE; ) {
        source.delete(0, source.length());
        while (codepoint % 0xFF != 0) {
            if (UCharacter.isDefined(codepoint)) {
                source.append(codepoint);
            }
            codepoint++;
        }
        if (UCharacter.isDefined(codepoint)) {
            source.append(codepoint);
        }
        if (codepoint != 0xFFFF) {
            codepoint++;
        }
        /*if (codepoint >= 0x04fc) {
                System.out.println("codepoint " + Integer.toHexString(codepoint));
                String str = source.substring(230, 232);
                System.out.println(android.icu.impl.Utility.escape(str));
                System.out.println("codepoint " + Integer.toHexString(codepoint) 
                                   + "length " + str.length());
                iter = en_us.getCollationElementIterator(str);
                CollationTest.backAndForth(this, iter);
            }
            */
        iter = en_us.getCollationElementIterator(source.toString());
        // A basic test to see if it's working at all
        CollationTest.backAndForth(this, iter);
    }
}

Example 70 with RuleBasedCollator

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

the class CollationIteratorTest method TestSearchCollatorElements.

/**
 * TestSearchCollatorElements tests iterator behavior (forwards and backwards) with
 * normalization on AND jamo tailoring, among other things.
 *
 * Note: This test is sensitive to changes of the root collator,
 * for example whether the ae-ligature maps to three CEs (as in the DUCET)
 * or to two CEs (as in the CLDR 24 FractionalUCA.txt).
 * It is also sensitive to how those CEs map to the iterator's 32-bit CE encoding.
 * For example, the DUCET's artificial secondary CE in the ae-ligature
 * may map to two 32-bit iterator CEs (as it did until ICU 52).
 */
@Test
public void TestSearchCollatorElements() {
    String tsceText = // simple LV Hangul
    " \uAC00" + // simple LVT Hangul
    " \uAC01" + // LVTT, last jamo expands for search
    " \uAC0F" + // LLVVVTT, every jamo expands for search
    " \uAFFF" + // 0xAC01 as conjoining jamo
    " \u1100\u1161\u11A8" + // 0xAC01 as compatibility jamo
    " \u3131\u314F\u3131" + // 0xAC0F as conjoining jamo; last expands for search
    " \u1100\u1161\u11B6" + // 0xAFFF as conjoining jamo; all expand for search
    " \u1101\u1170\u11B6" + // small letter ae, expands
    " \u00E6" + // small letter o with tilde and acute, decomposes
    " \u1E4D" + " ";
    int[] rootStandardOffsets = { 0, 1, 2, 2, 3, 4, 4, 4, 5, 6, 6, 6, 7, 8, 8, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, /* plus another 1-2 offset=26 if ae-ligature maps to three CEs */
    26, 27, 28, 28, 28, 29 };
    int[] rootSearchOffsets = { 0, 1, 2, 2, 3, 4, 4, 4, 5, 6, 6, 6, 6, 7, 8, 8, 8, 8, 8, 8, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 20, 21, 22, 22, 23, 23, 23, 24, 24, 25, 26, /* plus another 1-2 offset=26 if ae-ligature maps to three CEs */
    26, 27, 28, 28, 28, 29 };
    class TSCEItem {

        private String localeString;

        private int[] offsets;

        TSCEItem(String locStr, int[] offs) {
            localeString = locStr;
            offsets = offs;
        }

        public String getLocaleString() {
            return localeString;
        }

        public int[] getOffsets() {
            return offsets;
        }
    }
    final TSCEItem[] tsceItems = { new TSCEItem("root", rootStandardOffsets), new TSCEItem("root@collation=search", rootSearchOffsets) };
    for (TSCEItem tsceItem : tsceItems) {
        String localeString = tsceItem.getLocaleString();
        ULocale uloc = new ULocale(localeString);
        RuleBasedCollator col = null;
        try {
            col = (RuleBasedCollator) Collator.getInstance(uloc);
        } catch (Exception e) {
            errln("Error: in locale " + localeString + ", err in Collator.getInstance");
            continue;
        }
        CollationElementIterator uce = col.getCollationElementIterator(tsceText);
        int[] offsets = tsceItem.getOffsets();
        int ioff, noff = offsets.length;
        int offset, element;
        ioff = 0;
        do {
            offset = uce.getOffset();
            element = uce.next();
            logln(String.format("(%s) offset=%2d  ce=%08x\n", tsceItem.localeString, offset, element));
            if (element == 0) {
                errln("Error: in locale " + localeString + ", CEIterator next() returned element 0");
            }
            if (ioff < noff) {
                if (offset != offsets[ioff]) {
                    errln("Error: in locale " + localeString + ", expected CEIterator next()->getOffset " + offsets[ioff] + ", got " + offset);
                // ioff = noff;
                // break;
                }
                ioff++;
            } else {
                errln("Error: in locale " + localeString + ", CEIterator next() returned more elements than expected");
            }
        } while (element != CollationElementIterator.NULLORDER);
        if (ioff < noff) {
            errln("Error: in locale " + localeString + ", CEIterator next() returned fewer elements than expected");
        }
        // backwards test
        uce.setOffset(tsceText.length());
        ioff = noff;
        do {
            offset = uce.getOffset();
            element = uce.previous();
            if (element == 0) {
                errln("Error: in locale " + localeString + ", CEIterator previous() returned element 0");
            }
            if (ioff > 0) {
                ioff--;
                if (offset != offsets[ioff]) {
                    errln("Error: in locale " + localeString + ", expected CEIterator previous()->getOffset " + offsets[ioff] + ", got " + offset);
                // ioff = 0;
                // break;
                }
            } else {
                errln("Error: in locale " + localeString + ", CEIterator previous() returned more elements than expected");
            }
        } while (element != CollationElementIterator.NULLORDER);
        if (ioff > 0) {
            errln("Error: in locale " + localeString + ", CEIterator previous() returned fewer elements than expected");
        }
    }
}