void CollationIteratorTest::TestConstructors()
{
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *coll =
(RuleBasedCollator *)Collator::createInstance(status);
if (coll == NULL || U_FAILURE(status))
{
errln("Couldn't create a default collator.");
return;
}
// testing protected constructor with character iterator as argument
StringCharacterIterator chariter(test1);
CollationElementIterator *iter1 =
coll->createCollationElementIterator(chariter);
if (U_FAILURE(status)) {
errln("Couldn't create collation element iterator with character iterator.");
return;
}
CollationElementIterator *iter2 =
coll->createCollationElementIterator(test1);
// initially the 2 collation element iterators should be the same
if (*iter1 != *iter1 || *iter2 != *iter2 || *iter1 != *iter2
|| *iter2 != *iter1) {
errln("CollationElementIterators constructed with the same string data should be the same at the start");
}
assertEqual(*iter1, *iter2);
delete iter1;
delete iter2;
// tests empty strings
UnicodeString empty("");
iter1 = coll->createCollationElementIterator(empty);
chariter.setText(empty);
iter2 = coll->createCollationElementIterator(chariter);
if (*iter1 != *iter1 || *iter2 != *iter2 || *iter1 != *iter2
|| *iter2 != *iter1) {
errln("CollationElementIterators constructed with the same string data should be the same at the start");
}
if (iter1->next(status) != (int32_t)UCOL_NULLORDER) {
errln("Empty string should have no CEs.");
}
if (iter2->next(status) != (int32_t)UCOL_NULLORDER) {
errln("Empty string should have no CEs.");
}
delete iter1;
delete iter2;
delete coll;
}
// @bug 4101940
//
void CollationRegressionTest::Test4101940(/* char* par */)
{
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *c = NULL;
UnicodeString rules = "&9 < a < b";
UnicodeString nothing = "";
c = new RuleBasedCollator(rules, status);
if (c == NULL || U_FAILURE(status))
{
errln("Failed to create RuleBasedCollator");
delete c;
return;
}
CollationElementIterator *i = c->createCollationElementIterator(nothing);
i->reset();
if (i->next(status) != CollationElementIterator::NULLORDER)
{
errln("next did not return NULLORDER");
}
delete i;
delete c;
}
// @bug 4054238
//
// CollationElementIterator will not work correctly if the associated
// Collator object's mode is changed
//
void CollationRegressionTest::Test4054238(/* char* par */)
{
const UChar chars3[] = {0x61, 0x00FC, 0x62, 0x65, 0x63, 0x6b, 0x20, 0x47, 0x72, 0x00F6, 0x00DF, 0x65, 0x20, 0x4c, 0x00FC, 0x62, 0x63, 0x6b, 0};
const UnicodeString test3(chars3);
RuleBasedCollator *c = (RuleBasedCollator *) en_us->clone();
// NOTE: The Java code uses en_us to create the CollationElementIterators
// but I'm pretty sure that's wrong, so I've changed this to use c.
UErrorCode status = U_ZERO_ERROR;
c->setAttribute(UCOL_NORMALIZATION_MODE, UCOL_ON, status);
CollationElementIterator *i1 = c->createCollationElementIterator(test3);
delete i1;
delete c;
}
void CollationIteratorTest::TestStrengthOrder()
{
int order = 0x0123ABCD;
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *coll =
(RuleBasedCollator *)Collator::createInstance(status);
if (coll == NULL || U_FAILURE(status))
{
errln("Couldn't create a default collator.");
return;
}
coll->setStrength(Collator::PRIMARY);
CollationElementIterator *iter =
coll->createCollationElementIterator(test1);
if (iter == NULL) {
errln("Couldn't create a collation element iterator from default collator");
return;
}
if (iter->strengthOrder(order) != 0x01230000) {
errln("Strength order for a primary strength collator should be the first 2 bytes");
return;
}
coll->setStrength(Collator::SECONDARY);
if (iter->strengthOrder(order) != 0x0123AB00) {
errln("Strength order for a secondary strength collator should be the third byte");
return;
}
coll->setStrength(Collator::TERTIARY);
if (iter->strengthOrder(order) != order) {
errln("Strength order for a tertiary strength collator should be the third byte");
return;
}
delete iter;
delete coll;
}
void CollationIteratorTest::TestMaxExpansion(/* char* par */)
{
UErrorCode status = U_ZERO_ERROR;
UnicodeString rule("&a < ab < c/aba < d < z < ch");
RuleBasedCollator *coll = new RuleBasedCollator(rule, status);
UChar ch = 0;
UnicodeString str(ch);
CollationElementIterator *iter = coll->createCollationElementIterator(str);
while (ch < 0xFFFF && U_SUCCESS(status)) {
int count = 1;
uint32_t order;
ch ++;
UnicodeString str(ch);
iter->setText(str, status);
order = iter->previous(status);
/* thai management */
if (CollationElementIterator::isIgnorable(order))
order = iter->previous(status);
while (U_SUCCESS(status)
&& iter->previous(status) != (int32_t)UCOL_NULLORDER)
{
count ++;
}
if (U_FAILURE(status) && iter->getMaxExpansion(order) < count) {
errln("Failure at codepoint %d, maximum expansion count < %d\n",
ch, count);
}
}
delete iter;
delete coll;
}
void SSearchTest::offsetTest()
{
const char *test[] = {
// The sequence \u0FB3\u0F71\u0F71\u0F80 contains a discontiguous
// contraction (\u0FB3\u0F71\u0F80) logically followed by \u0F71.
"\\u1E33\\u0FB3\\u0F71\\u0F71\\u0F80\\uD835\\uDF6C\\u01B0",
"\\ua191\\u16ef\\u2036\\u017a",
#if 0
// This results in a complex interaction between contraction,
// expansion and normalization that confuses the backwards offset fixups.
"\\u0F7F\\u0F80\\u0F81\\u0F82\\u0F83\\u0F84\\u0F85",
#endif
"\\u0F80\\u0F81\\u0F82\\u0F83\\u0F84\\u0F85",
"\\u07E9\\u07EA\\u07F1\\u07F2\\u07F3",
"\\u02FE\\u02FF"
"\\u0300\\u0301\\u0302\\u0303\\u0304\\u0305\\u0306\\u0307\\u0308\\u0309\\u030A\\u030B\\u030C\\u030D\\u030E\\u030F"
"\\u0310\\u0311\\u0312\\u0313\\u0314\\u0315\\u0316\\u0317\\u0318\\u0319\\u031A\\u031B\\u031C\\u031D\\u031E\\u031F"
"\\u0320\\u0321\\u0322\\u0323\\u0324\\u0325\\u0326\\u0327\\u0328\\u0329\\u032A\\u032B\\u032C\\u032D\\u032E\\u032F"
"\\u0330\\u0331\\u0332\\u0333\\u0334\\u0335\\u0336\\u0337\\u0338\\u0339\\u033A\\u033B\\u033C\\u033D\\u033E\\u033F"
"\\u0340\\u0341\\u0342\\u0343\\u0344\\u0345\\u0346\\u0347\\u0348\\u0349\\u034A\\u034B\\u034C\\u034D\\u034E", // currently not working, see #8081
"\\u02FE\\u02FF\\u0300\\u0301\\u0302\\u0303\\u0316\\u0317\\u0318", // currently not working, see #8081
"a\\u02FF\\u0301\\u0316", // currently not working, see #8081
"a\\u02FF\\u0316\\u0301",
"a\\u0430\\u0301\\u0316",
"a\\u0430\\u0316\\u0301",
"abc\\u0E41\\u0301\\u0316",
"abc\\u0E41\\u0316\\u0301",
"\\u0E41\\u0301\\u0316",
"\\u0E41\\u0316\\u0301",
"a\\u0301\\u0316",
"a\\u0316\\u0301",
"\\uAC52\\uAC53",
"\\u34CA\\u34CB",
"\\u11ED\\u11EE",
"\\u30C3\\u30D0",
"p\\u00E9ch\\u00E9",
"a\\u0301\\u0325",
"a\\u0300\\u0325",
"a\\u0325\\u0300",
"A\\u0323\\u0300B",
"A\\u0300\\u0323B",
"A\\u0301\\u0323B",
"A\\u0302\\u0301\\u0323B",
"abc",
"ab\\u0300c",
"ab\\u0300\\u0323c",
" \\uD800\\uDC00\\uDC00",
"a\\uD800\\uDC00\\uDC00",
"A\\u0301\\u0301",
"A\\u0301\\u0323",
"A\\u0301\\u0323B",
"B\\u0301\\u0323C",
"A\\u0300\\u0323B",
"\\u0301A\\u0301\\u0301",
"abcd\\r\\u0301",
"p\\u00EAche",
"pe\\u0302che",
};
int32_t testCount = ARRAY_SIZE(test);
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *col = (RuleBasedCollator *) Collator::createInstance(Locale::getEnglish(), status);
if (U_FAILURE(status)) {
errcheckln(status, "Failed to create collator in offsetTest! - %s", u_errorName(status));
return;
}
char buffer[4096]; // A bit of a hack... just happens to be long enough for all the test cases...
// We could allocate one that's the right size by (CE_count * 10) + 2
// 10 chars is enough room for 8 hex digits plus ", ". 2 extra chars for "[" and "]"
col->setAttribute(UCOL_NORMALIZATION_MODE, UCOL_ON, status);
for(int32_t i = 0; i < testCount; i += 1) {
if (i>=4 && i<=6 && logKnownIssue("9156", "was 8081")) {
continue; // timebomb until ticket #9156 (was #8081) is resolved
}
UnicodeString ts = CharsToUnicodeString(test[i]);
CollationElementIterator *iter = col->createCollationElementIterator(ts);
OrderList forwardList;
OrderList backwardList;
int32_t order, low, high;
do {
low = iter->getOffset();
order = iter->next(status);
high = iter->getOffset();
forwardList.add(order, low, high);
} while (order != CollationElementIterator::NULLORDER);
iter->reset();
iter->setOffset(ts.length(), status);
backwardList.add(CollationElementIterator::NULLORDER, iter->getOffset(), iter->getOffset());
do {
high = iter->getOffset();
order = iter->previous(status);
low = iter->getOffset();
if (order == CollationElementIterator::NULLORDER) {
break;
}
backwardList.add(order, low, high);
} while (TRUE);
backwardList.reverse();
if (forwardList.compare(backwardList)) {
logln("Works with \"%s\"", test[i]);
logln("Forward offsets: [%s]", printOffsets(buffer, forwardList));
// logln("Backward offsets: [%s]", printOffsets(buffer, backwardList));
logln("Forward CEs: [%s]", printOrders(buffer, forwardList));
// logln("Backward CEs: [%s]", printOrders(buffer, backwardList));
logln();
} else {
errln("Fails with \"%s\"", test[i]);
infoln("Forward offsets: [%s]", printOffsets(buffer, forwardList));
infoln("Backward offsets: [%s]", printOffsets(buffer, backwardList));
infoln("Forward CEs: [%s]", printOrders(buffer, forwardList));
infoln("Backward CEs: [%s]", printOrders(buffer, backwardList));
infoln();
}
delete iter;
}
delete col;
}
void CollationIteratorTest::TestAssignment()
{
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *coll =
(RuleBasedCollator *)Collator::createInstance(status);
if (coll == NULL || U_FAILURE(status))
{
errln("Couldn't create a default collator.");
return;
}
UnicodeString source("abcd");
CollationElementIterator *iter1 =
coll->createCollationElementIterator(source);
CollationElementIterator iter2 = *iter1;
if (*iter1 != iter2) {
errln("Fail collation iterator assignment does not produce the same elements");
}
CollationElementIterator iter3(*iter1);
if (*iter1 != iter3) {
errln("Fail collation iterator copy constructor does not produce the same elements");
}
source = CharsToUnicodeString("a\\u0300\\u0325");
coll->setAttribute(UCOL_NORMALIZATION_MODE, UCOL_ON, status);
CollationElementIterator *iter4
= coll->createCollationElementIterator(source);
CollationElementIterator iter5(*iter4);
if (*iter4 != iter5) {
errln("collation iterator assignment does not produce the same elements");
}
iter4->next(status);
if (U_FAILURE(status) || *iter4 == iter5) {
errln("collation iterator not equal");
}
iter5.next(status);
if (U_FAILURE(status) || *iter4 != iter5) {
errln("collation iterator equal");
}
iter4->next(status);
if (U_FAILURE(status) || *iter4 == iter5) {
errln("collation iterator not equal");
}
iter5.next(status);
if (U_FAILURE(status) || *iter4 != iter5) {
errln("collation iterator equal");
}
CollationElementIterator iter6(*iter4);
if (*iter4 != iter6) {
errln("collation iterator equal");
}
iter4->next(status);
if (U_FAILURE(status) || *iter4 == iter5) {
errln("collation iterator not equal");
}
iter5.next(status);
if (U_FAILURE(status) || *iter4 != iter5) {
errln("collation iterator equal");
}
iter4->next(status);
if (U_FAILURE(status) || *iter4 == iter5) {
errln("collation iterator not equal");
}
iter5.next(status);
if (U_FAILURE(status) || *iter4 != iter5) {
errln("collation iterator equal");
}
delete iter1;
delete iter4;
delete coll;
}
void CollationIteratorTest::TestClearBuffers(/* char* par */)
{
UErrorCode status = U_ZERO_ERROR;
RuleBasedCollator *c = new RuleBasedCollator((UnicodeString)"&a < b < c & ab = d", status);
if (c == NULL || U_FAILURE(status))
{
errln("Couldn't create a RuleBasedCollator.");
delete c;
return;
}
UnicodeString source("abcd");
CollationElementIterator *i = c->createCollationElementIterator(source);
int32_t e0 = i->next(status); // save the first collation element
if (U_FAILURE(status))
{
errln("call to i->next() failed. err=%s", u_errorName(status));
}
else
{
i->setOffset(3, status); // go to the expanding character
if (U_FAILURE(status))
{
errln("call to i->setOffset(3) failed. err=%s", u_errorName(status));
}
else
{
i->next(status); // but only use up half of it
if (U_FAILURE(status))
{
errln("call to i->next() failed. err=%s", u_errorName(status));
}
else
{
i->setOffset(0, status); // go back to the beginning
if (U_FAILURE(status))
{
errln("call to i->setOffset(0) failed. err=%s", u_errorName(status));
}
else
{
int32_t e = i->next(status); // and get this one again
if (U_FAILURE(status))
{
errln("call to i->next() failed. err=%s", u_errorName(status));
}
else if (e != e0)
{
errln("got 0x%X, expected 0x%X", e, e0);
}
}
}
}
}
delete i;
delete c;
}