void genericOrderingTestWithResult(UCollator *coll, const char * const s[], uint32_t size, UCollationResult result) {
UChar t1[2048] = {0};
UChar t2[2048] = {0};
UCollationElements *iter;
UErrorCode status = U_ZERO_ERROR;
uint32_t i = 0, j = 0;
log_verbose("testing sequence:\n");
for(i = 0; i < size; i++) {
log_verbose("%s\n", s[i]);
}
iter = ucol_openElements(coll, t1, u_strlen(t1), &status);
if (U_FAILURE(status)) {
log_err("Creation of iterator failed\n");
}
for(i = 0; i < size-1; i++) {
for(j = i+1; j < size; j++) {
u_unescape(s[i], t1, 2048);
u_unescape(s[j], t2, 2048);
doTest(coll, t1, t2, result);
/* synwee : added collation element iterator test */
ucol_setText(iter, t1, u_strlen(t1), &status);
backAndForth(iter);
ucol_setText(iter, t2, u_strlen(t2), &status);
backAndForth(iter);
}
}
ucol_closeElements(iter);
}
/**
* Test for CollationElementIterator previous and next for the whole set of
* unicode characters with normalization on.
*/
static void TestNormalizedUnicodeChar()
{
UChar source[0x100];
UCollator *th_th;
UCollationElements *iter;
UErrorCode status = U_ZERO_ERROR;
UChar codepoint;
UChar *test;
/* thai should have normalization on */
th_th = ucol_open("th_TH", &status);
if (U_FAILURE(status)){
log_err_status(status, "ERROR: in creation of thai collation using ucol_open()\n %s\n",
myErrorName(status));
return;
}
for (codepoint = 1; codepoint < 0xFFFE;)
{
test = source;
while (codepoint % 0xFF != 0)
{
if (u_isdefined(codepoint))
*(test ++) = codepoint;
codepoint ++;
}
if (u_isdefined(codepoint))
*(test ++) = codepoint;
if (codepoint != 0xFFFF)
codepoint ++;
*test = 0;
iter=ucol_openElements(th_th, source, u_strlen(source), &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
ucol_close(th_th);
return;
}
backAndForth(iter);
ucol_closeElements(iter);
iter=ucol_openElements(th_th, source, -1, &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
ucol_close(th_th);
return;
}
backAndForth(iter);
ucol_closeElements(iter);
}
ucol_close(th_th);
}
void CollationIteratorTest::TestUnicodeChar()
{
CollationElementIterator *iter;
UChar codepoint;
UnicodeString source;
for (codepoint = 1; codepoint < 0xFFFE;)
{
source.remove();
while (codepoint % 0xFF != 0)
{
if (u_isdefined(codepoint))
source += codepoint;
codepoint ++;
}
if (u_isdefined(codepoint))
source += codepoint;
if (codepoint != 0xFFFF)
codepoint ++;
iter = en_us->createCollationElementIterator(source);
/* A basic test to see if it's working at all */
backAndForth(*iter);
delete iter;
}
}
U_CDECL_END
#define LINES 6
void CollationThaiTest::TestInvalidThai(void) {
const char *tests[LINES] = {
"\\u0E44\\u0E01\\u0E44\\u0E01",
"\\u0E44\\u0E01\\u0E01\\u0E44",
"\\u0E01\\u0E44\\u0E01\\u0E44",
"\\u0E01\\u0E01\\u0E44\\u0E44",
"\\u0E44\\u0E44\\u0E01\\u0E01",
"\\u0E01\\u0E44\\u0E44\\u0E01",
};
UChar strings[LINES][20];
UChar *toSort[LINES];
int32_t i = 0, j = 0, len = 0;
UErrorCode coll_status = U_ZERO_ERROR;
UnicodeString iteratorText;
thaiColl = ucol_open ("th_TH", &coll_status);
if (U_FAILURE(coll_status)) {
errln("Error opening Thai collator: %s", u_errorName(coll_status));
return;
}
CollationElementIterator* c = ((RuleBasedCollator *)coll)->createCollationElementIterator( iteratorText );
for(i = 0; i < (int32_t)(sizeof(tests)/sizeof(tests[0])); i++) {
len = u_unescape(tests[i], strings[i], 20);
strings[i][len] = 0;
toSort[i] = strings[i];
}
qsort (toSort, LINES, sizeof (UChar *), StrCmp);
for (i=0; i < LINES; i++)
{
logln("%i", i);
for (j=i+1; j < LINES; j++) {
if (ucol_strcoll (thaiColl, toSort[i], -1, toSort[j], -1) == UCOL_GREATER)
{
// inconsistency ordering found!
errln("Inconsistent ordering between strings %i and %i", i, j);
}
}
iteratorText.setTo(toSort[i]);
c->setText(iteratorText, coll_status);
backAndForth(*c);
}
ucol_close(thaiColl);
delete c;
}
void
IntlTestCollator::doTest(Collator* col, const UnicodeString &source, const UnicodeString &target, Collator::EComparisonResult result)
{
if(col) {
doTestVariant(col, source, target, result);
if(result == Collator::LESS) {
doTestVariant(col, target, source, Collator::GREATER);
} else if (result == Collator::GREATER) {
doTestVariant(col, target, source, Collator::LESS);
}
UErrorCode status = U_ZERO_ERROR;
LocalPointer<CollationElementIterator> c(((RuleBasedCollator *)col)->createCollationElementIterator(source));
logln("Testing iterating source: "+source);
backAndForth(*c);
c->setText(target, status);
logln("Testing iterating target: "+target);
backAndForth(*c);
}
}
/**
* Test for CollationElementIterator.previous()
*
* @bug 4108758 - Make sure it works with contracting characters
*
*/
static void TestPrevious()
{
UCollator *coll=NULL;
UChar rule[50];
UChar *source;
UCollator *c1, *c2, *c3;
UCollationElements *iter;
UErrorCode status = U_ZERO_ERROR;
UChar test1[50];
UChar test2[50];
u_uastrcpy(test1, "What subset of all possible test cases?");
u_uastrcpy(test2, "has the highest probability of detecting");
coll = ucol_open("en_US", &status);
iter=ucol_openElements(coll, test1, u_strlen(test1), &status);
log_verbose("English locale testing back and forth\n");
if(U_FAILURE(status)){
log_err_status(status, "ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
ucol_close(coll);
return;
}
/* A basic test to see if it's working at all */
backAndForth(iter);
ucol_closeElements(iter);
ucol_close(coll);
/* Test with a contracting character sequence */
u_uastrcpy(rule, "&a,A < b,B < c,C, d,D < z,Z < ch,cH,Ch,CH");
c1 = ucol_openRules(rule, u_strlen(rule), UCOL_OFF, UCOL_DEFAULT_STRENGTH, NULL, &status);
log_verbose("Contraction rule testing back and forth with no normalization\n");
if (c1 == NULL || U_FAILURE(status))
{
log_err("Couldn't create a RuleBasedCollator with a contracting sequence\n %s\n",
myErrorName(status));
return;
}
source=(UChar*)malloc(sizeof(UChar) * 20);
u_uastrcpy(source, "abchdcba");
iter=ucol_openElements(c1, source, u_strlen(source), &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
return;
}
backAndForth(iter);
ucol_closeElements(iter);
ucol_close(c1);
/* Test with an expanding character sequence */
u_uastrcpy(rule, "&a < b < c/abd < d");
c2 = ucol_openRules(rule, u_strlen(rule), UCOL_OFF, UCOL_DEFAULT_STRENGTH, NULL, &status);
log_verbose("Expansion rule testing back and forth with no normalization\n");
if (c2 == NULL || U_FAILURE(status))
{
log_err("Couldn't create a RuleBasedCollator with a contracting sequence.\n %s\n",
myErrorName(status));
return;
}
u_uastrcpy(source, "abcd");
iter=ucol_openElements(c2, source, u_strlen(source), &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
return;
}
backAndForth(iter);
ucol_closeElements(iter);
ucol_close(c2);
/* Now try both */
u_uastrcpy(rule, "&a < b < c/aba < d < z < ch");
c3 = ucol_openRules(rule, u_strlen(rule), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH,NULL, &status);
log_verbose("Expansion/contraction rule testing back and forth with no normalization\n");
if (c3 == NULL || U_FAILURE(status))
{
log_err("Couldn't create a RuleBasedCollator with a contracting sequence.\n %s\n",
myErrorName(status));
return;
}
u_uastrcpy(source, "abcdbchdc");
iter=ucol_openElements(c3, source, u_strlen(source), &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
return;
}
backAndForth(iter);
ucol_closeElements(iter);
ucol_close(c3);
source[0] = 0x0e41;
source[1] = 0x0e02;
source[2] = 0x0e41;
source[3] = 0x0e02;
source[4] = 0x0e27;
source[5] = 0x61;
source[6] = 0x62;
source[7] = 0x63;
source[8] = 0;
coll = ucol_open("th_TH", &status);
log_verbose("Thai locale testing back and forth with normalization\n");
iter=ucol_openElements(coll, source, u_strlen(source), &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
return;
}
backAndForth(iter);
ucol_closeElements(iter);
ucol_close(coll);
/* prev test */
source[0] = 0x0061;
source[1] = 0x30CF;
source[2] = 0x3099;
source[3] = 0x30FC;
source[4] = 0;
coll = ucol_open("ja_JP", &status);
log_verbose("Japanese locale testing back and forth with normalization\n");
iter=ucol_openElements(coll, source, u_strlen(source), &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
return;
}
backAndForth(iter);
ucol_closeElements(iter);
ucol_close(coll);
free(source);
}
/**
* Test the incremental normalization
*/
static void TestNormalization()
{
UErrorCode status = U_ZERO_ERROR;
const char *str =
"&a < \\u0300\\u0315 < A\\u0300\\u0315 < \\u0316\\u0315B < \\u0316\\u0300\\u0315";
UCollator *coll;
UChar rule[50];
int rulelen = u_unescape(str, rule, 50);
int count = 0;
const char *testdata[] =
{"\\u1ED9", "o\\u0323\\u0302",
"\\u0300\\u0315", "\\u0315\\u0300",
"A\\u0300\\u0315B", "A\\u0315\\u0300B",
"A\\u0316\\u0315B", "A\\u0315\\u0316B",
"\\u0316\\u0300\\u0315", "\\u0315\\u0300\\u0316",
"A\\u0316\\u0300\\u0315B", "A\\u0315\\u0300\\u0316B",
"\\u0316\\u0315\\u0300", "A\\u0316\\u0315\\u0300B"};
int32_t srclen;
UChar source[10];
UCollationElements *iter;
coll = ucol_openRules(rule, rulelen, UCOL_ON, UCOL_TERTIARY, NULL, &status);
ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &status);
if (U_FAILURE(status)){
log_err_status(status, "ERROR: in creation of collator using ucol_openRules()\n %s\n",
myErrorName(status));
return;
}
srclen = u_unescape(testdata[0], source, 10);
iter = ucol_openElements(coll, source, srclen, &status);
backAndForth(iter);
ucol_closeElements(iter);
srclen = u_unescape(testdata[1], source, 10);
iter = ucol_openElements(coll, source, srclen, &status);
backAndForth(iter);
ucol_closeElements(iter);
while (count < 12) {
srclen = u_unescape(testdata[count], source, 10);
iter = ucol_openElements(coll, source, srclen, &status);
if (U_FAILURE(status)){
log_err("ERROR: in creation of collator element iterator\n %s\n",
myErrorName(status));
return;
}
backAndForth(iter);
ucol_closeElements(iter);
iter = ucol_openElements(coll, source, -1, &status);
if (U_FAILURE(status)){
log_err("ERROR: in creation of collator element iterator\n %s\n",
myErrorName(status));
return;
}
backAndForth(iter);
ucol_closeElements(iter);
count ++;
}
ucol_close(coll);
}
/**
* Test for CollationElementIterator previous and next for the whole set of
* unicode characters.
*/
static void TestUnicodeChar()
{
UChar source[0x100];
UCollator *en_us;
UCollationElements *iter;
UErrorCode status = U_ZERO_ERROR;
UChar codepoint;
UChar *test;
en_us = ucol_open("en_US", &status);
if (U_FAILURE(status)){
log_err_status(status, "ERROR: in creation of collation data using ucol_open()\n %s\n",
myErrorName(status));
return;
}
for (codepoint = 1; codepoint < 0xFFFE;)
{
test = source;
while (codepoint % 0xFF != 0)
{
if (u_isdefined(codepoint))
*(test ++) = codepoint;
codepoint ++;
}
if (u_isdefined(codepoint))
*(test ++) = codepoint;
if (codepoint != 0xFFFF)
codepoint ++;
*test = 0;
iter=ucol_openElements(en_us, source, u_strlen(source), &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
ucol_close(en_us);
return;
}
/* A basic test to see if it's working at all */
log_verbose("codepoint testing %x\n", codepoint);
backAndForth(iter);
ucol_closeElements(iter);
/* null termination test */
iter=ucol_openElements(en_us, source, -1, &status);
if(U_FAILURE(status)){
log_err("ERROR: in creation of collation element iterator using ucol_openElements()\n %s\n",
myErrorName(status));
ucol_close(en_us);
return;
}
/* A basic test to see if it's working at all */
backAndForth(iter);
ucol_closeElements(iter);
}
ucol_close(en_us);
}
/**
* Testing the discontigous contractions
*/
static void TestDiscontiguos() {
const char *rulestr =
"&z < AB < X\\u0300 < ABC < X\\u0300\\u0315";
UChar rule[50];
int rulelen = u_unescape(rulestr, rule, 50);
const char *src[] = {
"ADB", "ADBC", "A\\u0315B", "A\\u0315BC",
/* base character blocked */
"XD\\u0300", "XD\\u0300\\u0315",
/* non blocking combining character */
"X\\u0319\\u0300", "X\\u0319\\u0300\\u0315",
/* blocking combining character */
"X\\u0314\\u0300", "X\\u0314\\u0300\\u0315",
/* contraction prefix */
"ABDC", "AB\\u0315C","X\\u0300D\\u0315", "X\\u0300\\u0319\\u0315",
"X\\u0300\\u031A\\u0315",
/* ends not with a contraction character */
"X\\u0319\\u0300D", "X\\u0319\\u0300\\u0315D", "X\\u0300D\\u0315D",
"X\\u0300\\u0319\\u0315D", "X\\u0300\\u031A\\u0315D"
};
const char *tgt[] = {
/* non blocking combining character */
"A D B", "A D BC", "A \\u0315 B", "A \\u0315 BC",
/* base character blocked */
"X D \\u0300", "X D \\u0300\\u0315",
/* non blocking combining character */
"X\\u0300 \\u0319", "X\\u0300\\u0315 \\u0319",
/* blocking combining character */
"X \\u0314 \\u0300", "X \\u0314 \\u0300\\u0315",
/* contraction prefix */
"AB DC", "AB \\u0315 C","X\\u0300 D \\u0315", "X\\u0300\\u0315 \\u0319",
"X\\u0300 \\u031A \\u0315",
/* ends not with a contraction character */
"X\\u0300 \\u0319D", "X\\u0300\\u0315 \\u0319D", "X\\u0300 D\\u0315D",
"X\\u0300\\u0315 \\u0319D", "X\\u0300 \\u031A\\u0315D"
};
int size = 20;
UCollator *coll;
UErrorCode status = U_ZERO_ERROR;
int count = 0;
UCollationElements *iter;
UCollationElements *resultiter;
coll = ucol_openRules(rule, rulelen, UCOL_OFF, UCOL_DEFAULT_STRENGTH,NULL, &status);
iter = ucol_openElements(coll, rule, 1, &status);
resultiter = ucol_openElements(coll, rule, 1, &status);
if (U_FAILURE(status)) {
log_err_status(status, "Error opening collation rules -> %s\n", u_errorName(status));
return;
}
while (count < size) {
UChar str[20];
UChar tstr[20];
int strLen = u_unescape(src[count], str, 20);
UChar *s;
ucol_setText(iter, str, strLen, &status);
if (U_FAILURE(status)) {
log_err("Error opening collation iterator\n");
return;
}
u_unescape(tgt[count], tstr, 20);
s = tstr;
log_verbose("count %d\n", count);
for (;;) {
uint32_t ce;
UChar *e = u_strchr(s, 0x20);
if (e == 0) {
e = u_strchr(s, 0);
}
ucol_setText(resultiter, s, (int32_t)(e - s), &status);
ce = ucol_next(resultiter, &status);
if (U_FAILURE(status)) {
log_err("Error manipulating collation iterator\n");
return;
}
while (ce != UCOL_NULLORDER) {
if (ce != (uint32_t)ucol_next(iter, &status) ||
U_FAILURE(status)) {
log_err("Discontiguos contraction test mismatch\n");
return;
}
ce = ucol_next(resultiter, &status);
if (U_FAILURE(status)) {
log_err("Error getting next collation element\n");
return;
}
}
s = e + 1;
if (*e == 0) {
break;
}
}
ucol_reset(iter);
backAndForth(iter);
count ++;
}
ucol_closeElements(resultiter);
ucol_closeElements(iter);
ucol_close(coll);
}
void CollationIteratorTest::TestPrevious(/* char* par */)
{
UErrorCode status = U_ZERO_ERROR;
CollationElementIterator *iter = en_us->createCollationElementIterator(test1);
// A basic test to see if it's working at all
backAndForth(*iter);
delete iter;
// Test with a contracting character sequence
UnicodeString source;
RuleBasedCollator *c1 = NULL;
c1 = new RuleBasedCollator(
(UnicodeString)"&a,A < b,B < c,C, d,D < z,Z < ch,cH,Ch,CH", status);
if (c1 == NULL || U_FAILURE(status))
{
errln("Couldn't create a RuleBasedCollator with a contracting sequence.");
delete c1;
return;
}
source = "abchdcba";
iter = c1->createCollationElementIterator(source);
backAndForth(*iter);
delete iter;
delete c1;
// Test with an expanding character sequence
RuleBasedCollator *c2 = NULL;
c2 = new RuleBasedCollator((UnicodeString)"&a < b < c/abd < d", status);
if (c2 == NULL || U_FAILURE(status))
{
errln("Couldn't create a RuleBasedCollator with an expanding sequence.");
delete c2;
return;
}
source = "abcd";
iter = c2->createCollationElementIterator(source);
backAndForth(*iter);
delete iter;
delete c2;
// Now try both
RuleBasedCollator *c3 = NULL;
c3 = new RuleBasedCollator((UnicodeString)"&a < b < c/aba < d < z < ch", status);
if (c3 == NULL || U_FAILURE(status))
{
errln("Couldn't create a RuleBasedCollator with both an expanding and a contracting sequence.");
delete c3;
return;
}
source = "abcdbchdc";
iter = c3->createCollationElementIterator(source);
backAndForth(*iter);
delete iter;
delete c3;
status=U_ZERO_ERROR;
source= CharsToUnicodeString("\\u0e41\\u0e02\\u0e41\\u0e02\\u0e27abc");
Collator *c4 = Collator::createInstance(Locale("th", "TH", ""), status);
if(U_FAILURE(status)){
errln("Couldn't create a collator");
}
iter = ((RuleBasedCollator*)c4)->createCollationElementIterator(source);
backAndForth(*iter);
delete iter;
delete c4;
source= CharsToUnicodeString("\\u0061\\u30CF\\u3099\\u30FC");
Collator *c5 = Collator::createInstance(Locale("ja", "JP", ""), status);
iter = ((RuleBasedCollator*)c5)->createCollationElementIterator(source);
if(U_FAILURE(status)){
errln("Couldn't create Japanese collator\n");
}
backAndForth(*iter);
delete iter;
delete c5;
}
