// @bug 4076676
//
// Bad canonicalization of same-class combining characters
//
void CollationRegressionTest::Test4076676(/* char* par */)
{
// These combining characters are all in the same class, so they should not
// be reordered, and they should compare as unequal.
static const UChar s1[] = {0x41, 0x0301, 0x0302, 0x0300, 0};
static const UChar s2[] = {0x41, 0x0302, 0x0300, 0x0301, 0};
RuleBasedCollator *c = (RuleBasedCollator *) en_us->clone();
c->setStrength(Collator::TERTIARY);
if (c->compare(s1,s2) == 0)
{
errln("Same-class combining chars were reordered");
}
delete c;
}
// @bug 4078588
//
// RuleBasedCollator breaks on "< a < bb" rule
//
void CollationRegressionTest::Test4078588(/* char *par */)
{
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *rbc = new RuleBasedCollator("&9 < a < bb", status);
if (rbc == NULL || U_FAILURE(status))
{
errln("Failed to create RuleBasedCollator.");
delete rbc;
return;
}
Collator::EComparisonResult result = rbc->compare("a","bb");
if (result != Collator::LESS)
{
errln((UnicodeString)"Compare(a,bb) returned " + (int)result
+ (UnicodeString)"; expected -1");
}
delete rbc;
}
// @bug 4081866
//
// Combining characters in different classes not reordered properly.
//
void CollationRegressionTest::Test4081866(/* char* par */)
{
// These combining characters are all in different classes,
// so they should be reordered and the strings should compare as equal.
static const UChar s1[] = {0x41, 0x0300, 0x0316, 0x0327, 0x0315, 0};
static const UChar s2[] = {0x41, 0x0327, 0x0316, 0x0315, 0x0300, 0};
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *c = (RuleBasedCollator *) en_us->clone();
c->setStrength(Collator::TERTIARY);
// Now that the default collators are set to NO_DECOMPOSITION
// (as a result of fixing bug 4114077), we must set it explicitly
// when we're testing reordering behavior. -- lwerner, 5/5/98
c->setAttribute(UCOL_NORMALIZATION_MODE, UCOL_ON, status);
if (c->compare(s1,s2) != 0)
{
errln("Combining chars were not reordered");
}
delete c;
}
void CollationCurrencyTest::currencyTest(/*char *par*/)
{
// All the currency symbols, in collation order
static const UChar currency[][2] =
{
{ 0x00A4, 0x0000}, /*00A4; L; [14 36, 03, 03] # [082B.0020.0002] # CURRENCY SIGN*/
{ 0x00A2, 0x0000}, /*00A2; L; [14 38, 03, 03] # [082C.0020.0002] # CENT SIGN*/
{ 0xFFE0, 0x0000}, /*FFE0; L; [14 38, 03, 05] # [082C.0020.0003] # FULLWIDTH CENT SIGN*/
{ 0x0024, 0x0000}, /*0024; L; [14 3A, 03, 03] # [082D.0020.0002] # DOLLAR SIGN*/
{ 0xFF04, 0x0000}, /*FF04; L; [14 3A, 03, 05] # [082D.0020.0003] # FULLWIDTH DOLLAR SIGN*/
{ 0xFE69, 0x0000}, /*FE69; L; [14 3A, 03, 1D] # [082D.0020.000F] # SMALL DOLLAR SIGN*/
{ 0x00A3, 0x0000}, /*00A3; L; [14 3C, 03, 03] # [082E.0020.0002] # POUND SIGN*/
{ 0xFFE1, 0x0000}, /*FFE1; L; [14 3C, 03, 05] # [082E.0020.0003] # FULLWIDTH POUND SIGN*/
{ 0x00A5, 0x0000}, /*00A5; L; [14 3E, 03, 03] # [082F.0020.0002] # YEN SIGN*/
{ 0xFFE5, 0x0000}, /*FFE5; L; [14 3E, 03, 05] # [082F.0020.0003] # FULLWIDTH YEN SIGN*/
{ 0x09F2, 0x0000}, /*09F2; L; [14 40, 03, 03] # [0830.0020.0002] # BENGALI RUPEE MARK*/
{ 0x09F3, 0x0000}, /*09F3; L; [14 42, 03, 03] # [0831.0020.0002] # BENGALI RUPEE SIGN*/
{ 0x0E3F, 0x0000}, /*0E3F; L; [14 44, 03, 03] # [0832.0020.0002] # THAI CURRENCY SYMBOL BAHT*/
{ 0x17DB, 0x0000}, /*17DB; L; [14 46, 03, 03] # [0833.0020.0002] # KHMER CURRENCY SYMBOL RIEL*/
{ 0x20A0, 0x0000}, /*20A0; L; [14 48, 03, 03] # [0834.0020.0002] # EURO-CURRENCY SIGN*/
{ 0x20A1, 0x0000}, /*20A1; L; [14 4A, 03, 03] # [0835.0020.0002] # COLON SIGN*/
{ 0x20A2, 0x0000}, /*20A2; L; [14 4C, 03, 03] # [0836.0020.0002] # CRUZEIRO SIGN*/
{ 0x20A3, 0x0000}, /*20A3; L; [14 4E, 03, 03] # [0837.0020.0002] # FRENCH FRANC SIGN*/
{ 0x20A4, 0x0000}, /*20A4; L; [14 50, 03, 03] # [0838.0020.0002] # LIRA SIGN*/
{ 0x20A5, 0x0000}, /*20A5; L; [14 52, 03, 03] # [0839.0020.0002] # MILL SIGN*/
{ 0x20A6, 0x0000}, /*20A6; L; [14 54, 03, 03] # [083A.0020.0002] # NAIRA SIGN*/
{ 0x20A7, 0x0000}, /*20A7; L; [14 56, 03, 03] # [083B.0020.0002] # PESETA SIGN*/
{ 0x20A9, 0x0000}, /*20A9; L; [14 58, 03, 03] # [083C.0020.0002] # WON SIGN*/
{ 0xFFE6, 0x0000}, /*FFE6; L; [14 58, 03, 05] # [083C.0020.0003] # FULLWIDTH WON SIGN*/
{ 0x20AA, 0x0000}, /*20AA; L; [14 5A, 03, 03] # [083D.0020.0002] # NEW SHEQEL SIGN*/
{ 0x20AB, 0x0000}, /*20AB; L; [14 5C, 03, 03] # [083E.0020.0002] # DONG SIGN*/
{ 0x20AC, 0x0000}, /*20AC; L; [14 5E, 03, 03] # [083F.0020.0002] # EURO SIGN*/
{ 0x20AD, 0x0000}, /*20AD; L; [14 60, 03, 03] # [0840.0020.0002] # KIP SIGN*/
{ 0x20AE, 0x0000}, /*20AE; L; [14 62, 03, 03] # [0841.0020.0002] # TUGRIK SIGN*/
{ 0x20AF, 0x0000}, /*20AF; L; [14 64, 03, 03] # [0842.0020.0002] # DRACHMA SIGN*/
};
uint32_t i, j;
UErrorCode status = U_ZERO_ERROR;
Collator::EComparisonResult expectedResult = Collator::EQUAL;
RuleBasedCollator *c = (RuleBasedCollator *)Collator::createInstance("en_US", status);
if (U_FAILURE(status))
{
errln ("Collator::createInstance() failed!");
return;
}
// Compare each currency symbol against all the
// currency symbols, including itself
for (i = 0; i < ARRAY_LENGTH(currency); i += 1)
{
for (j = 0; j < ARRAY_LENGTH(currency); j += 1)
{
UnicodeString source(currency[i], 1);
UnicodeString target(currency[j], 1);
if (i < j)
{
expectedResult = Collator::LESS;
}
else if ( i == j)
{
expectedResult = Collator::EQUAL;
}
else
{
expectedResult = Collator::GREATER;
}
Collator::EComparisonResult compareResult = c->compare(source, target);
CollationKey sourceKey, targetKey;
UErrorCode status = U_ZERO_ERROR;
c->getCollationKey(source, sourceKey, status);
if (U_FAILURE(status))
{
errln("Couldn't get collationKey for source");
continue;
}
c->getCollationKey(target, targetKey, status);
if (U_FAILURE(status))
{
errln("Couldn't get collationKey for target");
continue;
}
Collator::EComparisonResult keyResult = sourceKey.compareTo(targetKey);
reportCResult( source, target, sourceKey, targetKey, compareResult, keyResult, compareResult, expectedResult );
}
}
delete c;
}