/**
* Resets the cursor to the beginning of the string.
*/
void CollationElementIterator::reset()
{
ucol_reset(m_data_);
}
//static void NativeCollation_reset(JNIEnv*, jclass, jint address) {
JNIEXPORT void JNICALL
Java_com_ibm_icu4jni_text_NativeCollation_reset(JNIEnv*, jclass, jint address) {
ucol_reset(toCollationElements(address));
}
static void NativeCollation_reset(JNIEnv*, jclass, jlong address) {
ucol_reset(toCollationElements(address));
}
void
backAndForth(UCollationElements *iter)
{
/* Run through the iterator forwards and stick it into an array */
int32_t index, o;
UErrorCode status = U_ZERO_ERROR;
int32_t orderLength = 0;
int32_t *orders;
orders= getOrders(iter, &orderLength);
/* Now go through it backwards and make sure we get the same values */
index = orderLength;
ucol_reset(iter);
/* synwee : changed */
while ((o = ucol_previous(iter, &status)) != UCOL_NULLORDER)
{
if (o != orders[-- index])
{
if (o == 0)
index ++;
else
{
while (index > 0 && orders[-- index] == 0)
{
}
if (o != orders[index])
{
log_err("Mismatch at index : 0x%x\n", index);
return;
}
}
}
}
while (index != 0 && orders[index - 1] == 0) {
index --;
}
if (index != 0)
{
log_err("Didn't get back to beginning - index is %d\n", index);
ucol_reset(iter);
log_err("\nnext: ");
if ((o = ucol_next(iter, &status)) != UCOL_NULLORDER)
{
log_err("Error at %x\n", o);
}
log_err("\nprev: ");
if ((o = ucol_previous(iter, &status)) != UCOL_NULLORDER)
{
log_err("Error at %x\n", o);
}
log_verbose("\n");
}
free(orders);
}
void
backAndForth(UCollationElements *iter)
{
/* Run through the iterator forwards and stick it into an array */
int32_t idx, o;
UErrorCode status = U_ZERO_ERROR;
int32_t orderLength = 0;
OrderAndOffset *orders = getOrders(iter, &orderLength);
/* Now go through it backwards and make sure we get the same values */
idx = orderLength;
ucol_reset(iter);
/* synwee : changed */
while ((o = ucol_previous(iter, &status)) != UCOL_NULLORDER) {
#if TEST_OFFSETS
int32_t offset =
#endif
ucol_getOffset(iter);
idx -= 1;
if (o != orders[idx].order) {
if (o == 0)
idx ++;
else {
while (idx > 0 && orders[-- idx].order == 0) {
/* nothing... */
}
if (o != orders[idx].order) {
log_err("Mismatched order at index %d: 0x%8.8X vs. 0x%8.8X\n", idx,
orders[idx].order, o);
goto bail;
}
}
}
#if TEST_OFFSETS
if (offset != orders[idx].offset) {
log_err("Mismatched offset at index %d: %d vs. %d\n", idx,
orders[idx].offset, offset);
goto bail;
}
#endif
}
while (idx != 0 && orders[idx - 1].order == 0) {
idx -= 1;
}
if (idx != 0) {
log_err("Didn't get back to beginning - index is %d\n", idx);
ucol_reset(iter);
log_err("\nnext: ");
if ((o = ucol_next(iter, &status)) != UCOL_NULLORDER) {
log_err("Error at %x\n", o);
}
log_err("\nprev: ");
if ((o = ucol_previous(iter, &status)) != UCOL_NULLORDER) {
log_err("Error at %x\n", o);
}
log_verbose("\n");
}
bail:
free(orders);
}
/**
* Testing iterators with extremely small buffers
*/
static void TestSmallBuffer()
{
UErrorCode status = U_ZERO_ERROR;
UCollator *coll;
UCollationElements *testiter,
*iter;
int32_t count = 0;
OrderAndOffset *testorders,
*orders;
UChar teststr[500];
UChar str[] = {0x300, 0x31A, 0};
/*
creating a long string of decomposable characters,
since by default the writable buffer is of size 256
*/
while (count < 500) {
if ((count & 1) == 0) {
teststr[count ++] = 0x300;
}
else {
teststr[count ++] = 0x31A;
}
}
coll = ucol_open("th_TH", &status);
if(U_SUCCESS(status) && coll) {
testiter = ucol_openElements(coll, teststr, 500, &status);
iter = ucol_openElements(coll, str, 2, &status);
orders = getOrders(iter, &count);
if (count != 2) {
log_err("Error collation elements size is not 2 for \\u0300\\u031A\n");
}
/*
this will rearrange the string data to 250 characters of 0x300 first then
250 characters of 0x031A
*/
testorders = getOrders(testiter, &count);
if (count != 500) {
log_err("Error decomposition does not give the right sized collation elements\n");
}
while (count != 0) {
/* UCA collation element for 0x0F76 */
if ((count > 250 && testorders[-- count].order != orders[1].order) ||
(count <= 250 && testorders[-- count].order != orders[0].order)) {
log_err("Error decomposition does not give the right collation element at %d count\n", count);
break;
}
}
free(testorders);
free(orders);
ucol_reset(testiter);
/* ensures closing of elements done properly to clear writable buffer */
ucol_next(testiter, &status);
ucol_next(testiter, &status);
ucol_closeElements(testiter);
ucol_closeElements(iter);
ucol_close(coll);
} else {
log_err_status(status, "Couldn't open collator -> %s\n", u_errorName(status));
}
}
static void TestBug672() {
UErrorCode status = U_ZERO_ERROR;
UChar pattern[20];
UChar text[50];
int i;
int result[3][3];
u_uastrcpy(pattern, "resume");
u_uastrcpy(text, "Time to resume updating my resume.");
for (i = 0; i < 3; ++ i) {
UCollator *coll = ucol_open(LOCALES[i], &status);
UCollationElements *pitr = ucol_openElements(coll, pattern, -1,
&status);
UCollationElements *titer = ucol_openElements(coll, text, -1,
&status);
if (U_FAILURE(status)) {
log_err_status(status, "ERROR: in creation of either the collator or the collation iterator :%s\n",
myErrorName(status));
return;
}
log_verbose("locale tested %s\n", LOCALES[i]);
while (ucol_next(pitr, &status) != UCOL_NULLORDER &&
U_SUCCESS(status)) {
}
if (U_FAILURE(status)) {
log_err("ERROR: reversing collation iterator :%s\n",
myErrorName(status));
return;
}
ucol_reset(pitr);
ucol_setOffset(titer, u_strlen(pattern), &status);
if (U_FAILURE(status)) {
log_err("ERROR: setting offset in collator :%s\n",
myErrorName(status));
return;
}
result[i][0] = ucol_getOffset(titer);
log_verbose("Text iterator set to offset %d\n", result[i][0]);
/* Use previous() */
ucol_previous(titer, &status);
result[i][1] = ucol_getOffset(titer);
log_verbose("Current offset %d after previous\n", result[i][1]);
/* Add one to index */
log_verbose("Adding one to current offset...\n");
ucol_setOffset(titer, ucol_getOffset(titer) + 1, &status);
if (U_FAILURE(status)) {
log_err("ERROR: setting offset in collator :%s\n",
myErrorName(status));
return;
}
result[i][2] = ucol_getOffset(titer);
log_verbose("Current offset in text = %d\n", result[i][2]);
ucol_closeElements(pitr);
ucol_closeElements(titer);
ucol_close(coll);
}
if (uprv_memcmp(result[0], result[1], 3) != 0 ||
uprv_memcmp(result[1], result[2], 3) != 0) {
log_err("ERROR: Different locales have different offsets at the same character\n");
}
}
/**
* Test for getOffset() and setOffset()
*/
static void TestOffset()
{
UErrorCode status= U_ZERO_ERROR;
UCollator *en_us=NULL;
UCollationElements *iter, *pristine;
int32_t offset;
OrderAndOffset *orders;
int32_t orderLength=0;
int count = 0;
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");
en_us = ucol_open("en_US", &status);
log_verbose("Testing getOffset and setOffset for collations\n");
iter = ucol_openElements(en_us, test1, u_strlen(test1), &status);
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(en_us);
return;
}
/* testing boundaries */
ucol_setOffset(iter, 0, &status);
if (U_FAILURE(status) || ucol_previous(iter, &status) != UCOL_NULLORDER) {
log_err("Error: After setting offset to 0, we should be at the end "
"of the backwards iteration");
}
ucol_setOffset(iter, u_strlen(test1), &status);
if (U_FAILURE(status) || ucol_next(iter, &status) != UCOL_NULLORDER) {
log_err("Error: After setting offset to end of the string, we should "
"be at the end of the backwards iteration");
}
/* Run all the way through the iterator, then get the offset */
orders = getOrders(iter, &orderLength);
offset = ucol_getOffset(iter);
if (offset != u_strlen(test1))
{
log_err("offset at end != length %d vs %d\n", offset,
u_strlen(test1) );
}
/* Now set the offset back to the beginning and see if it works */
pristine=ucol_openElements(en_us, test1, u_strlen(test1), &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;
}
status = U_ZERO_ERROR;
ucol_setOffset(iter, 0, &status);
if (U_FAILURE(status))
{
log_err("setOffset failed. %s\n", myErrorName(status));
}
else
{
assertEqual(iter, pristine);
}
ucol_closeElements(pristine);
ucol_closeElements(iter);
free(orders);
/* testing offsets in normalization buffer */
test1[0] = 0x61;
test1[1] = 0x300;
test1[2] = 0x316;
test1[3] = 0x62;
test1[4] = 0;
ucol_setAttribute(en_us, UCOL_NORMALIZATION_MODE, UCOL_ON, &status);
iter = ucol_openElements(en_us, test1, 4, &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;
}
count = 0;
while (ucol_next(iter, &status) != UCOL_NULLORDER &&
U_SUCCESS(status)) {
switch (count) {
case 0:
if (ucol_getOffset(iter) != 1) {
log_err("ERROR: Offset of iteration should be 1\n");
}
break;
case 3:
if (ucol_getOffset(iter) != 4) {
log_err("ERROR: Offset of iteration should be 4\n");
}
break;
default:
if (ucol_getOffset(iter) != 3) {
log_err("ERROR: Offset of iteration should be 3\n");
}
}
count ++;
}
ucol_reset(iter);
count = 0;
while (ucol_previous(iter, &status) != UCOL_NULLORDER &&
U_SUCCESS(status)) {
switch (count) {
case 0:
case 1:
if (ucol_getOffset(iter) != 3) {
log_err("ERROR: Offset of iteration should be 3\n");
}
break;
case 2:
if (ucol_getOffset(iter) != 1) {
log_err("ERROR: Offset of iteration should be 1\n");
}
break;
default:
if (ucol_getOffset(iter) != 0) {
log_err("ERROR: Offset of iteration should be 0\n");
}
}
count ++;
}
if(U_FAILURE(status)){
log_err("ERROR: in iterating collation elements %s\n",
myErrorName(status));
}
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);
}
