static UBool versionFilter(UChar32 ch, void* context) {
static const UVersionInfo none = { 0, 0, 0, 0 };
UVersionInfo v;
u_charAge(ch, v);
UVersionInfo* version = (UVersionInfo*)context;
return uprv_memcmp(&v, &none, sizeof(v)) > 0 && uprv_memcmp(&v, version, sizeof(v)) <= 0;
}
static int32_t
findUnicodeVersion(const UVersionInfo version) {
int32_t i;
for(i=0; /* while(version>unicodeVersions[i]) {} */
i<UNI_VER_COUNT && uprv_memcmp(version, unicodeVersions[i], 4)>0;
++i) {}
if(0<i && i<UNI_VER_COUNT && uprv_memcmp(version, unicodeVersions[i], 4)<0) {
--i; /* fix 4.0.2 to land before 4.1, for valid x>=ucdVersion comparisons */
}
return i; /* version>=unicodeVersions[i] && version<unicodeVersions[i+1]; possible: i==UNI_VER_COUNT */
}
UBool
CollationKey::operator==(const CollationKey& source) const
{
return getLength() == source.getLength() &&
(this == &source ||
uprv_memcmp(getBytes(), source.getBytes(), getLength()) == 0);
}
Collator *
Collator::createInstance(const Locale &loc,
UVersionInfo version,
UErrorCode &status)
{
Collator *collator;
UVersionInfo info;
collator=new RuleBasedCollator(loc, status);
/* test for NULL */
if (collator == 0) {
status = U_MEMORY_ALLOCATION_ERROR;
return 0;
}
if(U_SUCCESS(status)) {
collator->getVersion(info);
if(0!=uprv_memcmp(version, info, sizeof(UVersionInfo))) {
delete collator;
status=U_MISSING_RESOURCE_ERROR;
return 0;
}
}
return collator;
}
UBool
CollationKey::operator==(const CollationKey& source) const
{
return (this->fCount == source.fCount &&
(this->fBytes == source.fBytes ||
uprv_memcmp(this->fBytes, source.fBytes, this->fCount) == 0));
}
U_CAPI UBool U_EXPORT2 ulist_containsString(const UList * list, const char * data, int32_t length)
{
UBool result = FALSE;
const UListNode * pointer = NULL;
if (list != NULL && list->size != 0)
{
pointer = list->head;
while (pointer != NULL)
{
if (length == uprv_strlen(pointer->data))
{
if (uprv_memcmp(data, pointer->data, length) == 0)
{
result = TRUE;
break;
}
}
pointer = pointer->next;
}
}
return result;
}
//-----------------------------------------------------------------------------
//
// setAdd Set operation on UVector
// dest = dest union source
// Elements may only appear once and must be sorted.
//
//-----------------------------------------------------------------------------
void RBBITableBuilder::setAdd(UVector *dest, UVector *source) {
int32_t destOriginalSize = dest->size();
int32_t sourceSize = source->size();
int32_t di = 0;
MaybeStackArray<void *, 16> destArray, sourceArray; // Handle small cases without malloc
void **destPtr, **sourcePtr;
void **destLim, **sourceLim;
if (destOriginalSize > destArray.getCapacity()) {
if (destArray.resize(destOriginalSize) == NULL) {
return;
}
}
destPtr = destArray.getAlias();
destLim = destPtr + destOriginalSize; // destArray.getArrayLimit()?
if (sourceSize > sourceArray.getCapacity()) {
if (sourceArray.resize(sourceSize) == NULL) {
return;
}
}
sourcePtr = sourceArray.getAlias();
sourceLim = sourcePtr + sourceSize; // sourceArray.getArrayLimit()?
// Avoid multiple "get element" calls by getting the contents into arrays
(void) dest->toArray(destPtr);
(void) source->toArray(sourcePtr);
dest->setSize(sourceSize+destOriginalSize, *fStatus);
while (sourcePtr < sourceLim && destPtr < destLim) {
if (*destPtr == *sourcePtr) {
dest->setElementAt(*sourcePtr++, di++);
destPtr++;
}
// This check is required for machines with segmented memory, like i5/OS.
// Direct pointer comparison is not recommended.
else if (uprv_memcmp(destPtr, sourcePtr, sizeof(void *)) < 0) {
dest->setElementAt(*destPtr++, di++);
}
else { /* *sourcePtr < *destPtr */
dest->setElementAt(*sourcePtr++, di++);
}
}
// At most one of these two cleanup loops will execute
while (destPtr < destLim) {
dest->setElementAt(*destPtr++, di++);
}
while (sourcePtr < sourceLim) {
dest->setElementAt(*sourcePtr++, di++);
}
dest->setSize(di, *fStatus);
}
/* test one string with the ICU and the reference BOCU-1 implementations */
static void
roundtripBOCU1(UConverter *bocu1, int32_t number, const UChar *text, int32_t length) {
UChar *roundtripRef, *roundtripICU;
char *bocu1Ref, *bocu1ICU;
int32_t bocu1RefLength, bocu1ICULength, roundtripRefLength, roundtripICULength;
UErrorCode errorCode;
roundtripRef = malloc(DEFAULT_BUFFER_SIZE * sizeof(UChar));
roundtripICU = malloc(DEFAULT_BUFFER_SIZE * sizeof(UChar));
bocu1Ref = malloc(DEFAULT_BUFFER_SIZE);
bocu1ICU = malloc(DEFAULT_BUFFER_SIZE);
/* Unicode -> BOCU-1 */
bocu1RefLength=writeString(text, length, (uint8_t *)bocu1Ref);
errorCode=U_ZERO_ERROR;
bocu1ICULength=ucnv_fromUChars(bocu1, bocu1ICU, DEFAULT_BUFFER_SIZE, text, length, &errorCode);
if(U_FAILURE(errorCode)) {
log_err("ucnv_fromUChars(BOCU-1, text(%d)[%d]) failed: %s\n", number, length, u_errorName(errorCode));
goto cleanup;
}
if(bocu1RefLength!=bocu1ICULength || 0!=uprv_memcmp(bocu1Ref, bocu1ICU, bocu1RefLength)) {
log_err("Unicode(%d)[%d] -> BOCU-1: reference[%d]!=ICU[%d]\n", number, length, bocu1RefLength, bocu1ICULength);
goto cleanup;
}
/* BOCU-1 -> Unicode */
roundtripRefLength=readString((uint8_t *)bocu1Ref, bocu1RefLength, roundtripRef);
if(roundtripRefLength<0) {
goto cleanup; /* readString() found an error and reported it */
}
roundtripICULength=ucnv_toUChars(bocu1, roundtripICU, DEFAULT_BUFFER_SIZE, bocu1ICU, bocu1ICULength, &errorCode);
if(U_FAILURE(errorCode)) {
log_err("ucnv_toUChars(BOCU-1, text(%d)[%d]) failed: %s\n", number, length, u_errorName(errorCode));
goto cleanup;
}
if(length!=roundtripRefLength || 0!=u_memcmp(text, roundtripRef, length)) {
log_err("BOCU-1 -> Unicode: original(%d)[%d]!=reference[%d]\n", number, length, roundtripRefLength);
goto cleanup;
}
if(roundtripRefLength!=roundtripICULength || 0!=u_memcmp(roundtripRef, roundtripICU, roundtripRefLength)) {
log_err("BOCU-1 -> Unicode: reference(%d)[%d]!=ICU[%d]\n", number, roundtripRefLength, roundtripICULength);
goto cleanup;
}
cleanup:
free(roundtripRef);
free(roundtripICU);
free(bocu1Ref);
free(bocu1ICU);
}
void CollationRegressionTest::TestT7189() {
UErrorCode status = U_ZERO_ERROR;
UCollator *coll;
uint32_t i;
static const UChar text1[][CollationRegressionTest::MAX_TOKEN_LEN] = {
// "Achter De Hoven"
{ 0x41, 0x63, 0x68, 0x74, 0x65, 0x72, 0x20, 0x44, 0x65, 0x20, 0x48, 0x6F, 0x76, 0x65, 0x6E, 0x00 },
// "ABC"
{ 0x41, 0x42, 0x43, 0x00 },
// "HELLO world!"
{ 0x48, 0x45, 0x4C, 0x4C, 0x4F, 0x20, 0x77, 0x6F, 0x72, 0x6C, 0x64, 0x21, 0x00 }
};
static const UChar text2[][CollationRegressionTest::MAX_TOKEN_LEN] = {
// "Achter de Hoven"
{ 0x41, 0x63, 0x68, 0x74, 0x65, 0x72, 0x20, 0x64, 0x65, 0x20, 0x48, 0x6F, 0x76, 0x65, 0x6E, 0x00 },
// "abc"
{ 0x61, 0x62, 0x63, 0x00 },
// "hello world!"
{ 0x68, 0x65, 0x6C, 0x6C, 0x6F, 0x20, 0x77, 0x6F, 0x72, 0x6C, 0x64, 0x21, 0x00 }
};
// Open the collator
coll = ucol_openFromShortString("EO_S1", FALSE, NULL, &status);
if (U_FAILURE(status)) {
errln("Failed to create a collator for short string EO_S1");
return;
}
for (i = 0; i < sizeof(text1) / (CollationRegressionTest::MAX_TOKEN_LEN * sizeof(UChar)); i++) {
uint8_t key1[100], key2[100];
int32_t len1, len2;
len1 = calcKeyIncremental(coll, text1[i], -1, key1, sizeof(key1), status);
if (U_FAILURE(status)) {
errln(UnicodeString("Failed to get a partial collation key for ") + text1[i]);
break;
}
len2 = calcKeyIncremental(coll, text2[i], -1, key2, sizeof(key2), status);
if (U_FAILURE(status)) {
errln(UnicodeString("Failed to get a partial collation key for ") + text2[i]);
break;
}
if (len1 == len2 && uprv_memcmp(key1, key2, len1) == 0) {
errln(UnicodeString("Failed: Identical key\n") + " text1: " + text1[i] + "\n" + " text2: " + text2[i] + "\n" + " key : " + TestUtility::hex(key1, len1));
} else {
logln(UnicodeString("Keys produced -\n") + " text1: " + text1[i] + "\n" + " key1 : " + TestUtility::hex(key1, len1) + "\n" + " text2: " + text2[i] + "\n" + " key2 : "
+ TestUtility::hex(key2, len2));
}
}
ucol_close(coll);
}
UBool CollationElementIterator::operator==(
const CollationElementIterator& that) const
{
if (this == &that || m_data_ == that.m_data_) {
return TRUE;
}
// option comparison
if (m_data_->iteratordata_.coll != that.m_data_->iteratordata_.coll)
{
return FALSE;
}
// the constructor and setText always sets a length
// and we only compare the string not the contents of the normalization
// buffer
int thislength = (int)(m_data_->iteratordata_.endp - m_data_->iteratordata_.string);
int thatlength = (int)(that.m_data_->iteratordata_.endp - that.m_data_->iteratordata_.string);
if (thislength != thatlength) {
return FALSE;
}
if (uprv_memcmp(m_data_->iteratordata_.string,
that.m_data_->iteratordata_.string,
thislength * U_SIZEOF_UCHAR) != 0) {
return FALSE;
}
if (getOffset() != that.getOffset()) {
return FALSE;
}
// checking normalization buffer
if ((m_data_->iteratordata_.flags & UCOL_ITER_HASLEN) == 0) {
if ((that.m_data_->iteratordata_.flags & UCOL_ITER_HASLEN) != 0) {
return FALSE;
}
// both are in the normalization buffer
if (m_data_->iteratordata_.pos
- m_data_->iteratordata_.writableBuffer.getBuffer()
!= that.m_data_->iteratordata_.pos
- that.m_data_->iteratordata_.writableBuffer.getBuffer()) {
// not in the same position in the normalization buffer
return FALSE;
}
}
else if ((that.m_data_->iteratordata_.flags & UCOL_ITER_HASLEN) == 0) {
return FALSE;
}
// checking ce position
return (m_data_->iteratordata_.CEpos - m_data_->iteratordata_.CEs)
== (that.m_data_->iteratordata_.CEpos
- that.m_data_->iteratordata_.CEs);
}
//-----------------------------------------------------------------------------
//
// Operator == Consider two RBBIDataWrappers to be equal if they
// refer to the same underlying data. Although
// the data wrappers are normally shared between
// iterator instances, it's possible to independently
// open the same data twice, and get two instances, which
// should still be ==.
//
//-----------------------------------------------------------------------------
UBool RBBIDataWrapper::operator ==(const RBBIDataWrapper &other) const {
if (fHeader == other.fHeader) {
return TRUE;
}
if (fHeader->fLength != other.fHeader->fLength) {
return FALSE;
}
if (uprv_memcmp(fHeader, other.fHeader, fHeader->fLength) == 0) {
return TRUE;
}
return FALSE;
}
U_CAPI UBool U_EXPORT2 ulist_containsString(const UList *list, const char *data, int32_t length) {
if (list != NULL) {
const UListNode *pointer;
for (pointer = list->head; pointer != NULL; pointer = pointer->next) {
if (length == uprv_strlen(pointer->data)) {
if (uprv_memcmp(data, pointer->data, length) == 0) {
return TRUE;
}
}
}
}
return FALSE;
}
UBool SearchIterator::operator==(const SearchIterator &that) const
{
if (this == &that) {
return TRUE;
}
return (m_breakiterator_ == that.m_breakiterator_ &&
m_search_->isCanonicalMatch == that.m_search_->isCanonicalMatch &&
m_search_->isOverlap == that.m_search_->isOverlap &&
m_search_->matchedIndex == that.m_search_->matchedIndex &&
m_search_->matchedLength == that.m_search_->matchedLength &&
m_search_->textLength == that.m_search_->textLength &&
getOffset() == that.getOffset() &&
(uprv_memcmp(m_search_->text, that.m_search_->text,
m_search_->textLength * sizeof(UChar)) == 0));
}
U_EXPORT UBool U_EXPORT2
operator==(const StringPiece& x, const StringPiece& y) {
int32_t len = x.size();
if (len != y.size()) {
return false;
}
if (len == 0) {
return true;
}
const char* p = x.data();
const char* p2 = y.data();
// Test last byte in case strings share large common prefix
--len;
if (p[len] != p2[len]) return false;
// At this point we can, but don't have to, ignore the last byte.
return uprv_memcmp(p, p2, len) == 0;
}
// Bitwise comparison for the collation keys.
UCollationResult
CollationKey::compareTo(const CollationKey& target, UErrorCode &status) const
{
if(U_SUCCESS(status)) {
uint8_t *src = this->fBytes;
uint8_t *tgt = target.fBytes;
// are we comparing the same string
if (src == tgt)
return UCOL_EQUAL;
int minLength;
UCollationResult result;
// are we comparing different lengths?
if (this->fCount != target.fCount) {
if (this->fCount < target.fCount) {
minLength = this->fCount;
result = UCOL_LESS;
}
else {
minLength = target.fCount;
result = UCOL_GREATER;
}
}
else {
minLength = target.fCount;
result = UCOL_EQUAL;
}
if (minLength > 0) {
int diff = uprv_memcmp(src, tgt, minLength);
if (diff > 0) {
return UCOL_GREATER;
}
else
if (diff < 0) {
return UCOL_LESS;
}
}
return result;
} else {
return UCOL_EQUAL;
}
}
static void Test_aestrncpy(int32_t line, const uint8_t *expect, const uint8_t *src, int32_t len)
{
uint8_t str_buf[AESTRNCPY_SIZE] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
uint8_t *ret;
log_verbose("\n%s:%d: Beginning test of uprv_aestrncpy(dst, src, %d)\n", __FILE__, line, len);
ret = uprv_aestrncpy(str_buf, src, len);
if(ret != str_buf) {
log_err("\n%s:%d: FAIL: uprv_aestrncpy returned %p expected %p\n", __FILE__, line, (void*)ret, (void*)str_buf);
}
if(!uprv_memcmp(str_buf, expect, AESTRNCPY_SIZE)) {
log_verbose("\n%s:%d: OK - compared OK.", __FILE__, line);
log_verbose("\n%s:%d: expected: %s", __FILE__, line, dump_binline((uint8_t *)expect));
log_verbose("\n%s:%d: got : %s\n", __FILE__, line, dump_binline(str_buf));
} else {
log_err ("\n%s:%d: FAIL: uprv_aestrncpy output differs", __FILE__, line);
log_err ("\n%s:%d: expected: %s", __FILE__, line, dump_binline((uint8_t *)expect));
log_err ("\n%s:%d: got : %s\n", __FILE__, line, dump_binline(str_buf));
}
}
// Bitwise comparison for the collation keys.
UCollationResult
CollationKey::compareTo(const CollationKey& target, UErrorCode &status) const
{
if(U_SUCCESS(status)) {
const uint8_t *src = getBytes();
const uint8_t *tgt = target.getBytes();
// are we comparing the same string
if (src == tgt)
return UCOL_EQUAL;
UCollationResult result;
// are we comparing different lengths?
int32_t minLength = getLength();
int32_t targetLength = target.getLength();
if (minLength < targetLength) {
result = UCOL_LESS;
} else if (minLength == targetLength) {
result = UCOL_EQUAL;
} else {
minLength = targetLength;
result = UCOL_GREATER;
}
if (minLength > 0) {
int diff = uprv_memcmp(src, tgt, minLength);
if (diff > 0) {
return UCOL_GREATER;
}
else
if (diff < 0) {
return UCOL_LESS;
}
}
return result;
} else {
return UCOL_EQUAL;
}
}
UCAConformanceTest::UCAConformanceTest() :
rbUCA(NULL),
testFile(NULL),
status(U_ZERO_ERROR)
{
UCA = (RuleBasedCollator *)Collator::createInstance(Locale::getRoot(), status);
if(U_FAILURE(status)) {
dataerrln("Error - UCAConformanceTest: Unable to open UCA collator! - %s", u_errorName(status));
}
const char *srcDir = IntlTest::getSourceTestData(status);
if (U_FAILURE(status)) {
dataerrln("Could not open test data %s", u_errorName(status));
return;
}
uprv_strcpy(testDataPath, srcDir);
uprv_strcat(testDataPath, "CollationTest_");
UVersionInfo uniVersion;
static const UVersionInfo v62 = { 6, 2, 0, 0 };
u_getUnicodeVersion(uniVersion);
isAtLeastUCA62 = uprv_memcmp(uniVersion, v62, 4) >= 0;
}
static void
TestCaseTitle(void) {
static const UChar
beforeTitle[]= { 0x61, 0x42, 0x20, 0x69, 0x3c2, 0x20, 0xdf, 0x3c3, 0x2f, 0xfb03, 0xd93f, 0xdfff },
titleWord[]= { 0x41, 0x62, 0x20, 0x49, 0x3c2, 0x20, 0x53, 0x73, 0x3c3, 0x2f, 0x46, 0x66, 0x69, 0xd93f, 0xdfff },
titleChar[]= { 0x41, 0x42, 0x20, 0x49, 0x3a3, 0x20, 0x53, 0x73, 0x3a3, 0x2f, 0x46, 0x66, 0x69, 0xd93f, 0xdfff };
UChar buffer[32];
UBreakIterator *titleIterChars;
int32_t length;
UErrorCode errorCode;
errorCode=U_ZERO_ERROR;
titleIterChars=ubrk_open(UBRK_CHARACTER, "", beforeTitle, sizeof(beforeTitle)/U_SIZEOF_UCHAR, &errorCode);
if(U_FAILURE(errorCode)) {
log_err_status(errorCode, "error: ubrk_open(UBRK_CHARACTER)->%s\n", u_errorName(errorCode));
return;
}
/* titlecase with standard break iterator and in the same buffer */
uprv_memcpy(buffer, beforeTitle, sizeof(beforeTitle));
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
buffer, sizeof(beforeTitle)/U_SIZEOF_UCHAR,
NULL, "",
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(titleWord)/U_SIZEOF_UCHAR) ||
uprv_memcmp(titleWord, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strToTitle(standard iterator)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(titleWord, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no");
}
/* titlecase with UBRK_CHARACTERS and separate buffers */
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
beforeTitle, sizeof(beforeTitle)/U_SIZEOF_UCHAR,
titleIterChars, "",
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(titleChar)/U_SIZEOF_UCHAR) ||
uprv_memcmp(titleChar, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strToTitle(UBRK_CHARACTERS)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(titleChar, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no");
}
/* test preflighting */
errorCode=U_ZERO_ERROR;
length=u_strToTitle(NULL, 0,
beforeTitle, sizeof(beforeTitle)/U_SIZEOF_UCHAR,
titleIterChars, "",
&errorCode);
if( errorCode!=U_BUFFER_OVERFLOW_ERROR ||
length!=(sizeof(titleChar)/U_SIZEOF_UCHAR)
) {
log_err("error in u_strToTitle(UBRK_CHARACTERS pure preflighting)=%ld error=%s\n",
length,
u_errorName(errorCode));
}
/* test error handling */
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
NULL, sizeof(beforeTitle)/U_SIZEOF_UCHAR,
titleIterChars, "",
&errorCode);
if( errorCode!=U_ILLEGAL_ARGUMENT_ERROR ||
buffer[0]!=0xabcd
) {
log_err("error in u_strToTitle(UBRK_CHARACTERS src=NULL)=%ld error=%s buffer[0]==0x%lx\n",
length,
u_errorName(errorCode),
buffer[0]);
}
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
beforeTitle, -2,
titleIterChars, "",
&errorCode);
if( errorCode!=U_ILLEGAL_ARGUMENT_ERROR ||
buffer[0]!=0xabcd
) {
log_err("error in u_strToTitle(UBRK_CHARACTERS srcLength=-2)=%ld error=%s buffer[0]==0x%lx\n",
length,
u_errorName(errorCode),
buffer[0]);
}
ubrk_close(titleIterChars);
}
// Bitwise comparison for the collation keys.
// NOTE: this is somewhat messy 'cause we can't count
// on memcmp returning the exact values which match
// Collator::EComparisonResult
Collator::EComparisonResult
CollationKey::compareTo(const CollationKey& target) const
{
uint8_t *src = this->fBytes;
uint8_t *tgt = target.fBytes;
// are we comparing the same string
if (src == tgt)
return Collator::EQUAL;
/*
int count = (this->fCount < target.fCount) ? this->fCount : target.fCount;
if (count == 0)
{
// If count is 0, at least one of the keys is empty.
// An empty key is always LESS than a non-empty one
// and EQUAL to another empty
if (this->fCount < target.fCount)
{
return Collator::LESS;
}
if (this->fCount > target.fCount)
{
return Collator::GREATER;
}
return Collator::EQUAL;
}
*/
int minLength;
Collator::EComparisonResult result;
// are we comparing different lengths?
if (this->fCount != target.fCount) {
if (this->fCount < target.fCount) {
minLength = this->fCount;
result = Collator::LESS;
}
else {
minLength = target.fCount;
result = Collator::GREATER;
}
}
else {
minLength = target.fCount;
result = Collator::EQUAL;
}
if (minLength > 0) {
int diff = uprv_memcmp(src, tgt, minLength);
if (diff > 0) {
return Collator::GREATER;
}
else
if (diff < 0) {
return Collator::LESS;
}
}
return result;
/*
if (result < 0)
{
return Collator::LESS;
}
if (result > 0)
{
return Collator::GREATER;
}
return Collator::EQUAL;
*/
}
/**
* Main Windows time zone detection function. Returns the Windows
* time zone, translated to an ICU time zone, or NULL upon failure.
*/
U_CFUNC const char* U_EXPORT2
uprv_detectWindowsTimeZone() {
UErrorCode status = U_ZERO_ERROR;
UResourceBundle* bundle = NULL;
char* icuid = NULL;
char apiStdName[MAX_LENGTH_ID];
char regStdName[MAX_LENGTH_ID];
char tmpid[MAX_LENGTH_ID];
int32_t len;
int id;
int errorCode;
UChar ISOcodeW[3]; /* 2 letter iso code in UTF-16*/
char ISOcodeA[3]; /* 2 letter iso code in ansi */
LONG result;
TZI tziKey;
TZI tziReg;
TIME_ZONE_INFORMATION apiTZI;
BOOL isVistaOrHigher;
BOOL tryPreVistaFallback;
OSVERSIONINFO osVerInfo;
/* Obtain TIME_ZONE_INFORMATION from the API, and then convert it
to TZI. We could also interrogate the registry directly; we do
this below if needed. */
uprv_memset(&apiTZI, 0, sizeof(apiTZI));
uprv_memset(&tziKey, 0, sizeof(tziKey));
uprv_memset(&tziReg, 0, sizeof(tziReg));
GetTimeZoneInformation(&apiTZI);
tziKey.bias = apiTZI.Bias;
uprv_memcpy((char *)&tziKey.standardDate, (char*)&apiTZI.StandardDate,
sizeof(apiTZI.StandardDate));
uprv_memcpy((char *)&tziKey.daylightDate, (char*)&apiTZI.DaylightDate,
sizeof(apiTZI.DaylightDate));
/* Convert the wchar_t* standard name to char* */
uprv_memset(apiStdName, 0, sizeof(apiStdName));
wcstombs(apiStdName, apiTZI.StandardName, MAX_LENGTH_ID);
tmpid[0] = 0;
id = GetUserGeoID(GEOCLASS_NATION);
errorCode = GetGeoInfoW(id,GEO_ISO2,ISOcodeW,3,0);
u_strToUTF8(ISOcodeA, 3, NULL, ISOcodeW, 3, &status);
bundle = ures_openDirect(NULL, "windowsZones", &status);
ures_getByKey(bundle, "mapTimezones", bundle, &status);
/*
Windows Vista+ provides us with a "TimeZoneKeyName" that is not localized
and can be used to directly map a name in our bundle. Try to use that first
if we're on Vista or higher
*/
uprv_memset(&osVerInfo, 0, sizeof(osVerInfo));
osVerInfo.dwOSVersionInfoSize = sizeof(osVerInfo);
GetVersionEx(&osVerInfo);
isVistaOrHigher = osVerInfo.dwMajorVersion >= 6; /* actually includes Windows Server 2008 as well, but don't worry about it */
tryPreVistaFallback = TRUE;
if(isVistaOrHigher) {
result = getTZKeyName(regStdName, sizeof(regStdName));
if(ERROR_SUCCESS == result) {
UResourceBundle* winTZ = ures_getByKey(bundle, regStdName, NULL, &status);
if(U_SUCCESS(status)) {
const UChar* icuTZ = NULL;
if (errorCode != 0) {
icuTZ = ures_getStringByKey(winTZ, ISOcodeA, &len, &status);
}
if (errorCode==0 || icuTZ==NULL) {
/* fallback to default "001" and reset status */
status = U_ZERO_ERROR;
icuTZ = ures_getStringByKey(winTZ, "001", &len, &status);
}
if(U_SUCCESS(status)) {
int index=0;
while (! (*icuTZ == '\0' || *icuTZ ==' ')) {
tmpid[index++]=(char)(*icuTZ++); /* safe to assume 'char' is ASCII compatible on windows */
}
tmpid[index]='\0';
tryPreVistaFallback = FALSE;
}
}
ures_close(winTZ);
}
}
if(tryPreVistaFallback) {
/* Note: We get the winid not from static tables but from resource bundle. */
while (U_SUCCESS(status) && ures_hasNext(bundle)) {
UBool idFound = FALSE;
const char* winid;
UResourceBundle* winTZ = ures_getNextResource(bundle, NULL, &status);
if (U_FAILURE(status)) {
break;
}
winid = ures_getKey(winTZ);
result = getTZI(winid, &tziReg);
if (result == ERROR_SUCCESS) {
/* Windows alters the DaylightBias in some situations.
Using the bias and the rules suffices, so overwrite
these unreliable fields. */
tziKey.standardBias = tziReg.standardBias;
tziKey.daylightBias = tziReg.daylightBias;
if (uprv_memcmp((char *)&tziKey, (char*)&tziReg, sizeof(tziKey)) == 0) {
const UChar* icuTZ = NULL;
if (errorCode != 0) {
icuTZ = ures_getStringByKey(winTZ, ISOcodeA, &len, &status);
}
if (errorCode==0 || icuTZ==NULL) {
/* fallback to default "001" and reset status */
status = U_ZERO_ERROR;
icuTZ = ures_getStringByKey(winTZ, "001", &len, &status);
}
if (U_SUCCESS(status)) {
/* Get the standard name from the registry key to compare with
the one from Windows API call. */
uprv_memset(regStdName, 0, sizeof(regStdName));
result = getSTDName(winid, regStdName, sizeof(regStdName));
if (result == ERROR_SUCCESS) {
if (uprv_strcmp(apiStdName, regStdName) == 0) {
idFound = TRUE;
}
}
/* tmpid buffer holds the ICU timezone ID corresponding to the timezone ID from Windows.
* If none is found, tmpid buffer will contain a fallback ID (i.e. the time zone ID matching
* the current time zone information)
*/
if (idFound || tmpid[0] == 0) {
/* if icuTZ has more than one city, take only the first (i.e. terminate icuTZ at first space) */
int index=0;
while (! (*icuTZ == '\0' || *icuTZ ==' ')) {
tmpid[index++]=(char)(*icuTZ++); /* safe to assume 'char' is ASCII compatible on windows */
}
tmpid[index]='\0';
}
}
}
}
ures_close(winTZ);
if (idFound) {
break;
}
}
}
/*
* Copy the timezone ID to icuid to be returned.
*/
if (tmpid[0] != 0) {
len = uprv_strlen(tmpid);
icuid = (char*)uprv_calloc(len + 1, sizeof(char));
if (icuid != NULL) {
uprv_strcpy(icuid, tmpid);
}
}
ures_close(bundle);
return icuid;
}
U_CFUNC const char* U_EXPORT2
uprv_detectWindowsTimeZone() {
UErrorCode status = U_ZERO_ERROR;
UResourceBundle* bundle = NULL;
char* icuid = NULL;
LONG result;
TZI tziKey;
TZI tziReg;
TIME_ZONE_INFORMATION apiTZI;
/* Obtain TIME_ZONE_INFORMATION from the API, and then convert it
to TZI. We could also interrogate the registry directly; we do
this below if needed. */
uprv_memset(&apiTZI, 0, sizeof(apiTZI));
uprv_memset(&tziKey, 0, sizeof(tziKey));
uprv_memset(&tziReg, 0, sizeof(tziReg));
GetTimeZoneInformation(&apiTZI);
tziKey.bias = apiTZI.Bias;
uprv_memcpy((char *)&tziKey.standardDate, (char*)&apiTZI.StandardDate,
sizeof(apiTZI.StandardDate));
uprv_memcpy((char *)&tziKey.daylightDate, (char*)&apiTZI.DaylightDate,
sizeof(apiTZI.DaylightDate));
bundle = ures_openDirect(NULL, "windowsZones", &status);
ures_getByKey(bundle, "mapTimezones", bundle, &status);
/* Note: We get the winid not from static tables but from resource bundle. */
while (U_SUCCESS(status) && ures_hasNext(bundle)) {
const char* winid;
int32_t len;
UResourceBundle* winTZ = ures_getNextResource(bundle, NULL, &status);
if (U_FAILURE(status)) {
break;
}
winid = ures_getKey(winTZ);
result = getTZI(winid, &tziReg);
if (result == ERROR_SUCCESS) {
/* Windows alters the DaylightBias in some situations.
Using the bias and the rules suffices, so overwrite
these unreliable fields. */
tziKey.standardBias = tziReg.standardBias;
tziKey.daylightBias = tziReg.daylightBias;
if (uprv_memcmp((char *)&tziKey, (char*)&tziReg, sizeof(tziKey)) == 0) {
const UChar* icuTZ = ures_getStringByKey(winTZ, "001", &len, &status);
if (U_SUCCESS(status)) {
icuid = (char*)uprv_malloc(sizeof(char) * (len + 1));
uprv_memset(icuid, 0, len + 1);
u_austrncpy(icuid, icuTZ, len);
}
}
}
ures_close(winTZ);
if (icuid != NULL) {
break;
}
}
ures_close(bundle);
return icuid;
}
static void TestAppendChar(){
static const uint8_t s[11]={0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00};
static const uint32_t test[]={
/*append-position(unsafe), CHAR to be appended */
0, 0x10401,
2, 0x0028,
2, 0x007f,
3, 0xd801,
1, 0x20402,
8, 0x10401,
5, 0xc0,
5, 0xc1,
5, 0xfd,
6, 0x80,
6, 0x81,
6, 0xbf,
7, 0xfe,
/*append-position(safe), CHAR to be appended */
0, 0x10401,
2, 0x0028,
3, 0x7f,
3, 0xd801, /* illegal for UTF-8 starting with Unicode 3.2 */
1, 0x20402,
9, 0x10401,
5, 0xc0,
5, 0xc1,
5, 0xfd,
6, 0x80,
6, 0x81,
6, 0xbf,
7, 0xfe,
};
static const uint16_t movedOffset[]={
/*offset-moved-to(unsafe)*/
4, /*for append-pos: 0 , CHAR 0x10401*/
3,
3,
6,
5,
12,
7,
7,
7,
8,
8,
8,
9,
/*offset-moved-to(safe)*/
4, /*for append-pos: 0, CHAR 0x10401*/
3,
4,
6,
5,
11,
7,
7,
7,
8,
8,
8,
9,
};
static const uint8_t result[][11]={
/*unsafe*/
{0xF0, 0x90, 0x90, 0x81, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x28, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x7f, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0xed, 0xa0, 0x81, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0xF0, 0xa0, 0x90, 0x82, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0xF0, 0x90, 0x90},
{0x61, 0x62, 0x63, 0x64, 0x65, 0xc3, 0x80, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0xc3, 0x81, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0xc3, 0xbd, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0xc2, 0x80, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0xc2, 0x81, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0xc2, 0xbf, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0xc3, 0xbe, 0x6a, 0x00},
/*safe*/
{0xF0, 0x90, 0x90, 0x81, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x28, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x7f, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0xef, 0xbf, 0xbf, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0xF0, 0xa0, 0x90, 0x82, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0xc2, 0x9f}, /*gets UTF8_ERROR_VALUE_2 which takes 2 bytes 0xc0, 0x9f*/
{0x61, 0x62, 0x63, 0x64, 0x65, 0xc3, 0x80, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0xc3, 0x81, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0xc3, 0xbd, 0x68, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0xc2, 0x80, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0xc2, 0x81, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0xc2, 0xbf, 0x69, 0x6a, 0x00},
{0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0xc3, 0xbe, 0x6a, 0x00},
};
uint16_t i, count=0;
uint8_t str[12];
uint32_t offset;
/* UChar32 c=0;*/
uint16_t size=sizeof(s)/sizeof(s[0]);
for(i=0; i<sizeof(test)/sizeof(test[0]); i=(uint16_t)(i+2)){
uprv_memcpy(str, s, size);
offset=test[i];
if(count<13){
UTF8_APPEND_CHAR_UNSAFE(str, offset, test[i+1]);
if(offset != movedOffset[count]){
log_err("ERROR: UTF8_APPEND_CHAR_UNSAFE failed to move the offset correctly for count=%d.\nExpectedOffset=%d currentOffset=%d\n",
count, movedOffset[count], offset);
}
if(uprv_memcmp(str, result[count], size) !=0){
log_err("ERROR: UTF8_APPEND_CHAR_UNSAFE failed for count=%d. \nExpected:", count);
printUChars(result[count], size);
log_err("\nGot: ");
printUChars(str, size);
log_err("\n");
}
}else{
UTF8_APPEND_CHAR_SAFE(str, offset, size, test[i+1]);
if(offset != movedOffset[count]){
log_err("ERROR: UTF8_APPEND_CHAR_SAFE failed to move the offset correctly for count=%d.\nExpectedOffset=%d currentOffset=%d\n",
count, movedOffset[count], offset);
}
if(uprv_memcmp(str, result[count], size) !=0){
log_err("ERROR: UTF8_APPEND_CHAR_SAFE failed for count=%d. \nExpected:", count);
printUChars(result[count], size);
log_err("\nGot: ");
printUChars(str, size);
log_err("\n");
}
/*call the API instead of MACRO
uprv_memcpy(str, s, size);
offset=test[i];
c=test[i+1];
if((uint32_t)(c)<=0x7f) {
(str)[(offset)++]=(uint8_t)(c);
} else {
(offset)=utf8_appendCharSafeBody(str, (int32_t)(offset), (int32_t)(size), c);
}
if(offset != movedOffset[count]){
log_err("ERROR: utf8_appendCharSafeBody() failed to move the offset correctly for count=%d.\nExpectedOffset=%d currentOffset=%d\n",
count, movedOffset[count], offset);
}
if(uprv_memcmp(str, result[count], size) !=0){
log_err("ERROR: utf8_appendCharSafeBody() failed for count=%d. \nExpected:", count);
printUChars(result[count], size);
printf("\nGot: ");
printUChars(str, size);
printf("\n");
}
*/
}
count++;
}
}
static void U_CALLCONV
unicodeDataLineFn(void *context,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
uint32_t decomp[40];
Norm norm;
const char *s;
char *end;
uint32_t code, value;
int32_t length;
UBool isCompat, something=FALSE;
/* ignore First and Last entries for ranges */
if( *fields[1][0]=='<' &&
(length=(int32_t)(fields[1][1]-fields[1][0]))>=9 &&
(0==uprv_memcmp(", First>", fields[1][1]-8, 8) || 0==uprv_memcmp(", Last>", fields[1][1]-7, 7))
) {
return;
}
/* reset the properties */
uprv_memset(&norm, 0, sizeof(Norm));
/* get the character code, field 0 */
code=(uint32_t)uprv_strtoul(fields[0][0], &end, 16);
if(end<=fields[0][0] || end!=fields[0][1]) {
fprintf(stderr, "gennorm: syntax error in field 0 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
/* get canonical combining class, field 3 */
value=(uint32_t)uprv_strtoul(fields[3][0], &end, 10);
if(end<=fields[3][0] || end!=fields[3][1] || value>0xff) {
fprintf(stderr, "gennorm: syntax error in field 3 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
if(value>0) {
norm.udataCC=(uint8_t)value;
something=TRUE;
}
/* get the decomposition, field 5 */
if(fields[5][0]<fields[5][1]) {
if(*(s=fields[5][0])=='<') {
++s;
isCompat=TRUE;
/* skip and ignore the compatibility type name */
do {
if(s==fields[5][1]) {
/* missing '>' */
fprintf(stderr, "gennorm: syntax error in field 5 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
} while(*s++!='>');
} else {
isCompat=FALSE;
}
/* parse the decomposition string */
length=u_parseCodePoints(s, decomp, sizeof(decomp)/4, pErrorCode);
if(U_FAILURE(*pErrorCode)) {
fprintf(stderr, "gennorm error parsing UnicodeData.txt decomposition of U+%04lx - %s\n",
(long)code, u_errorName(*pErrorCode));
exit(*pErrorCode);
}
/* store the string */
if(length>0) {
something=TRUE;
if(isCompat) {
norm.lenNFKD=(uint8_t)length;
norm.nfkd=decomp;
} else {
if(length>2) {
fprintf(stderr, "gennorm: error - length of NFD(U+%04lx) = %ld >2 in UnicodeData - illegal\n",
(long)code, (long)length);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
norm.lenNFD=(uint8_t)length;
norm.nfd=decomp;
}
}
}
/* check for non-character code points */
if((code&0xfffe)==0xfffe || (uint32_t)(code-0xfdd0)<0x20 || code>0x10ffff) {
fprintf(stderr, "gennorm: error - properties for non-character code point U+%04lx\n",
(long)code);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
if(something) {
/* there are normalization values, so store them */
#if 0
if(beVerbose) {
printf("store values for U+%04lx: cc=%d, lenNFD=%ld, lenNFKD=%ld\n",
(long)code, norm.udataCC, (long)norm.lenNFD, (long)norm.lenNFKD);
}
#endif
storeNorm(code, &norm);
}
}
static void U_CALLCONV
derivedNormalizationPropertiesLineFn(void *context,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
UChar string[32];
char *s;
uint32_t start, end;
int32_t count;
uint8_t qcFlags;
/* get code point range */
count=u_parseCodePointRange(fields[0][0], &start, &end, pErrorCode);
if(U_FAILURE(*pErrorCode)) {
fprintf(stderr, "gennorm: error parsing DerivedNormalizationProperties.txt mapping at %s\n", fields[0][0]);
exit(*pErrorCode);
}
/* ignore hangul - handle explicitly */
if(start==0xac00) {
return;
}
/* get property - ignore unrecognized ones */
s=(char *)u_skipWhitespace(fields[1][0]);
if(*s=='N' && s[1]=='F') {
/* quick check flag */
qcFlags=0x11;
s+=2;
if(*s=='K') {
qcFlags<<=1;
++s;
}
if(*s=='C' && s[1]=='_') {
s+=2;
} else if(*s=='D' && s[1]=='_') {
qcFlags<<=2;
s+=2;
} else {
return;
}
if(0==uprv_memcmp(s, "NO", 2)) {
qcFlags&=0xf;
} else if(0==uprv_memcmp(s, "MAYBE", 5)) {
qcFlags&=0x30;
} else if(0==uprv_memcmp(s, "QC", 2) && *(s=(char *)u_skipWhitespace(s+2))==';') {
/*
* Unicode 4.0.1:
* changes single field "NFD_NO" -> two fields "NFD_QC; N" etc.
*/
/* start of the field */
s=(char *)u_skipWhitespace(s+1);
if(*s=='N') {
qcFlags&=0xf;
} else if(*s=='M') {
qcFlags&=0x30;
} else {
return; /* do nothing for "Yes" because it's the default value */
}
} else {
return; /* do nothing for "Yes" because it's the default value */
}
/* set this flag for all code points in this range */
while(start<=end) {
setQCFlags(start++, qcFlags);
}
} else if(0==uprv_memcmp(s, "Comp_Ex", 7) || 0==uprv_memcmp(s, "Full_Composition_Exclusion", 26)) {
/* full composition exclusion */
while(start<=end) {
setCompositionExclusion(start++);
}
} else if(
((0==uprv_memcmp(s, "FNC", 3) && *(s=(char *)u_skipWhitespace(s+3))==';') ||
(0==uprv_memcmp(s, "FC_NFKC", 7) && *(s=(char *)u_skipWhitespace(s+7))==';'))
) {
/* FC_NFKC_Closure, parse field 2 to get the string */
char *t;
/* start of the field */
s=(char *)u_skipWhitespace(s+1);
/* find the end of the field */
for(t=s; *t!=';' && *t!='#' && *t!=0 && *t!='\n' && *t!='\r'; ++t) {}
*t=0;
string[0]=(UChar)u_parseString(s, string+1, 31, NULL, pErrorCode);
if(U_FAILURE(*pErrorCode)) {
fprintf(stderr, "gennorm error: illegal FNC string at %s\n", fields[0][0]);
exit(*pErrorCode);
}
while(start<=end) {
setFNC(start++, string);
}
}
}
/* Try titlecasing with options. */
static void
TestUCaseMapToTitle(void) {
/* "a 'CaT. A 'dOg! 'eTc." where '=U+02BB */
/*
* Note: The sentence BreakIterator does not recognize a '.'
* as a sentence terminator if it is followed by lowercase.
* That is why the example has the '!'.
*/
static const UChar
beforeTitle[]= { 0x61, 0x20, 0x2bb, 0x43, 0x61, 0x54, 0x2e, 0x20, 0x41, 0x20, 0x2bb, 0x64, 0x4f, 0x67, 0x21, 0x20, 0x2bb, 0x65, 0x54, 0x63, 0x2e },
titleWord[]= { 0x41, 0x20, 0x2bb, 0x43, 0x61, 0x74, 0x2e, 0x20, 0x41, 0x20, 0x2bb, 0x44, 0x6f, 0x67, 0x21, 0x20, 0x2bb, 0x45, 0x74, 0x63, 0x2e },
titleWordNoAdjust[]={ 0x41, 0x20, 0x2bb, 0x63, 0x61, 0x74, 0x2e, 0x20, 0x41, 0x20, 0x2bb, 0x64, 0x6f, 0x67, 0x21, 0x20, 0x2bb, 0x65, 0x74, 0x63, 0x2e },
titleSentNoLower[]= { 0x41, 0x20, 0x2bb, 0x43, 0x61, 0x54, 0x2e, 0x20, 0x41, 0x20, 0x2bb, 0x64, 0x4f, 0x67, 0x21, 0x20, 0x2bb, 0x45, 0x54, 0x63, 0x2e };
UChar buffer[32];
UCaseMap *csm;
UBreakIterator *sentenceIter;
const UBreakIterator *iter;
int32_t length;
UErrorCode errorCode;
errorCode=U_ZERO_ERROR;
csm=ucasemap_open("", 0, &errorCode);
if(U_FAILURE(errorCode)) {
log_err("ucasemap_open(\"\") failed - %s\n", u_errorName(errorCode));
return;
}
iter=ucasemap_getBreakIterator(csm);
if(iter!=NULL) {
log_err("ucasemap_getBreakIterator() returns %p!=NULL before setting any iterator or titlecasing\n", iter);
}
/* Use default UBreakIterator: Word breaks. */
length=ucasemap_toTitle(csm, buffer, UPRV_LENGTHOF(buffer), beforeTitle, UPRV_LENGTHOF(beforeTitle), &errorCode);
if( U_FAILURE(errorCode) ||
length!=UPRV_LENGTHOF(titleWord) ||
0!=u_memcmp(buffer, titleWord, length) ||
buffer[length]!=0
) {
log_err_status(errorCode, "ucasemap_toTitle(default iterator)=%ld failed - %s\n", (long)length, u_errorName(errorCode));
}
if (U_SUCCESS(errorCode)) {
iter=ucasemap_getBreakIterator(csm);
if(iter==NULL) {
log_err("ucasemap_getBreakIterator() returns NULL after titlecasing\n");
}
}
/* Try U_TITLECASE_NO_BREAK_ADJUSTMENT. */
ucasemap_setOptions(csm, U_TITLECASE_NO_BREAK_ADJUSTMENT, &errorCode);
if(U_FAILURE(errorCode)) {
log_err_status(errorCode, "error: ucasemap_setOptions(U_TITLECASE_NO_BREAK_ADJUSTMENT) failed - %s\n", u_errorName(errorCode));
return;
}
length=ucasemap_toTitle(csm, buffer, UPRV_LENGTHOF(buffer), beforeTitle, UPRV_LENGTHOF(beforeTitle), &errorCode);
if( U_FAILURE(errorCode) ||
length!=UPRV_LENGTHOF(titleWordNoAdjust) ||
0!=u_memcmp(buffer, titleWordNoAdjust, length) ||
buffer[length]!=0
) {
log_err("ucasemap_toTitle(default iterator, no break adjustment)=%ld failed - %s\n", (long)length, u_errorName(errorCode));
}
/* Set a sentence break iterator. */
errorCode=U_ZERO_ERROR;
sentenceIter=ubrk_open(UBRK_SENTENCE, "", NULL, 0, &errorCode);
if(U_FAILURE(errorCode)) {
log_err("error: ubrk_open(UBRK_SENTENCE) failed - %s\n", u_errorName(errorCode));
ucasemap_close(csm);
return;
}
ucasemap_setBreakIterator(csm, sentenceIter, &errorCode);
if(U_FAILURE(errorCode)) {
log_err("error: ucasemap_setBreakIterator(sentence iterator) failed - %s\n", u_errorName(errorCode));
ubrk_close(sentenceIter);
ucasemap_close(csm);
return;
}
iter=ucasemap_getBreakIterator(csm);
if(iter!=sentenceIter) {
log_err("ucasemap_getBreakIterator() returns %p!=%p after setting the iterator\n", iter, sentenceIter);
}
ucasemap_setOptions(csm, U_TITLECASE_NO_LOWERCASE, &errorCode);
if(U_FAILURE(errorCode)) {
log_err("error: ucasemap_setOptions(U_TITLECASE_NO_LOWERCASE) failed - %s\n", u_errorName(errorCode));
return;
}
/* Use the sentence break iterator with the option. Preflight first. */
length=ucasemap_toTitle(csm, NULL, 0, beforeTitle, UPRV_LENGTHOF(beforeTitle), &errorCode);
if( errorCode!=U_BUFFER_OVERFLOW_ERROR ||
length!=UPRV_LENGTHOF(titleSentNoLower)
) {
log_err("ucasemap_toTitle(preflight sentence break iterator, no lowercasing)=%ld failed - %s\n", (long)length, u_errorName(errorCode));
}
errorCode=U_ZERO_ERROR;
buffer[0]=0;
length=ucasemap_toTitle(csm, buffer, UPRV_LENGTHOF(buffer), beforeTitle, UPRV_LENGTHOF(beforeTitle), &errorCode);
if( U_FAILURE(errorCode) ||
length!=UPRV_LENGTHOF(titleSentNoLower) ||
0!=u_memcmp(buffer, titleSentNoLower, length) ||
buffer[length]!=0
) {
log_err("ucasemap_toTitle(sentence break iterator, no lowercasing)=%ld failed - %s\n", (long)length, u_errorName(errorCode));
}
/* UTF-8 C API coverage. More thorough test via C++ intltest's StringCaseTest::TestCasing(). */
{
char utf8BeforeTitle[64], utf8TitleSentNoLower[64], utf8[64];
int32_t utf8BeforeTitleLength, utf8TitleSentNoLowerLength;
errorCode=U_ZERO_ERROR;
u_strToUTF8(utf8BeforeTitle, (int32_t)sizeof(utf8BeforeTitle), &utf8BeforeTitleLength, beforeTitle, UPRV_LENGTHOF(beforeTitle), &errorCode);
u_strToUTF8(utf8TitleSentNoLower, (int32_t)sizeof(utf8TitleSentNoLower), &utf8TitleSentNoLowerLength, titleSentNoLower, UPRV_LENGTHOF(titleSentNoLower), &errorCode);
length=ucasemap_utf8ToTitle(csm, utf8, (int32_t)sizeof(utf8), utf8BeforeTitle, utf8BeforeTitleLength, &errorCode);
if( U_FAILURE(errorCode) ||
length!=utf8TitleSentNoLowerLength ||
0!=uprv_memcmp(utf8, utf8TitleSentNoLower, length) ||
utf8[length]!=0
) {
log_err("ucasemap_utf8ToTitle(sentence break iterator, no lowercasing)=%ld failed - %s\n", (long)length, u_errorName(errorCode));
}
}
ucasemap_close(csm);
}
static void
TestCaseCompare(void) {
static const UChar
mixed[]= { 0x61, 0x42, 0x131, 0x3a3, 0xdf, 0xfb03, 0xd93f, 0xdfff, 0 },
otherDefault[]= { 0x41, 0x62, 0x131, 0x3c3, 0x73, 0x53, 0x46, 0x66, 0x49, 0xd93f, 0xdfff, 0 },
otherExcludeSpecialI[]={ 0x41, 0x62, 0x131, 0x3c3, 0x53, 0x73, 0x66, 0x46, 0x69, 0xd93f, 0xdfff, 0 },
different[]= { 0x41, 0x62, 0x131, 0x3c3, 0x73, 0x53, 0x46, 0x66, 0x49, 0xd93f, 0xdffd, 0 };
UVersionInfo unicodeVersion={ 0, 0, 17, 89 }, unicode_3_1={ 3, 1, 0, 0 };
int32_t result, lenMixed, lenOtherDefault, lenOtherExcludeSpecialI, lenDifferent;
UErrorCode errorCode;
UBool isUnicode_3_1;
errorCode=U_ZERO_ERROR;
lenMixed=u_strlen(mixed);
lenOtherDefault=u_strlen(otherDefault);
(void)lenOtherDefault; /* Suppress set but not used warning. */
lenOtherExcludeSpecialI=u_strlen(otherExcludeSpecialI);
lenDifferent=u_strlen(different);
/* if unicodeVersion()>=3.1 then test exclude-special-i cases as well */
u_getUnicodeVersion(unicodeVersion);
isUnicode_3_1= uprv_memcmp(unicodeVersion, unicode_3_1, 4)>=0;
(void)isUnicode_3_1; /* Suppress set but not used warning. */
/* test u_strcasecmp() */
result=u_strcasecmp(mixed, otherDefault, U_FOLD_CASE_DEFAULT);
if(result!=0) {
log_err("error: u_strcasecmp(mixed, other, default)=%ld instead of 0\n", result);
}
result=u_strCaseCompare(mixed, -1, otherDefault, -1, U_FOLD_CASE_DEFAULT, &errorCode);
if(result!=0) {
log_err("error: u_strCaseCompare(mixed, other, default)=%ld instead of 0\n", result);
}
/* test u_strcasecmp() - exclude special i */
result=u_strcasecmp(mixed, otherExcludeSpecialI, U_FOLD_CASE_EXCLUDE_SPECIAL_I);
if(result!=0) {
log_err("error: u_strcasecmp(mixed, other, exclude special i)=%ld instead of 0\n", result);
}
result=u_strCaseCompare(mixed, lenMixed, otherExcludeSpecialI, lenOtherExcludeSpecialI, U_FOLD_CASE_EXCLUDE_SPECIAL_I, &errorCode);
if(result!=0) {
log_err("error: u_strCaseCompare(mixed, other, exclude special i)=%ld instead of 0\n", result);
}
/* test u_strcasecmp() */
result=u_strcasecmp(mixed, different, U_FOLD_CASE_DEFAULT);
if(result<=0) {
log_err("error: u_strcasecmp(mixed, different, default)=%ld instead of positive\n", result);
}
result=u_strCaseCompare(mixed, -1, different, lenDifferent, U_FOLD_CASE_DEFAULT, &errorCode);
if(result<=0) {
log_err("error: u_strCaseCompare(mixed, different, default)=%ld instead of positive\n", result);
}
/* test u_strncasecmp() - stop before the sharp s (U+00df) */
result=u_strncasecmp(mixed, different, 4, U_FOLD_CASE_DEFAULT);
if(result!=0) {
log_err("error: u_strncasecmp(mixed, different, 4, default)=%ld instead of 0\n", result);
}
result=u_strCaseCompare(mixed, 4, different, 4, U_FOLD_CASE_DEFAULT, &errorCode);
if(result!=0) {
log_err("error: u_strCaseCompare(mixed, 4, different, 4, default)=%ld instead of 0\n", result);
}
/* test u_strncasecmp() - stop in the middle of the sharp s (U+00df) */
result=u_strncasecmp(mixed, different, 5, U_FOLD_CASE_DEFAULT);
if(result<=0) {
log_err("error: u_strncasecmp(mixed, different, 5, default)=%ld instead of positive\n", result);
}
result=u_strCaseCompare(mixed, 5, different, 5, U_FOLD_CASE_DEFAULT, &errorCode);
if(result<=0) {
log_err("error: u_strCaseCompare(mixed, 5, different, 5, default)=%ld instead of positive\n", result);
}
/* test u_memcasecmp() - stop before the sharp s (U+00df) */
result=u_memcasecmp(mixed, different, 4, U_FOLD_CASE_DEFAULT);
if(result!=0) {
log_err("error: u_memcasecmp(mixed, different, 4, default)=%ld instead of 0\n", result);
}
/* test u_memcasecmp() - stop in the middle of the sharp s (U+00df) */
result=u_memcasecmp(mixed, different, 5, U_FOLD_CASE_DEFAULT);
if(result<=0) {
log_err("error: u_memcasecmp(mixed, different, 5, default)=%ld instead of positive\n", result);
}
}
static void
TestCaseFolding(void) {
/*
* CaseFolding.txt says about i and its cousins:
* 0049; C; 0069; # LATIN CAPITAL LETTER I
* 0049; T; 0131; # LATIN CAPITAL LETTER I
*
* 0130; F; 0069 0307; # LATIN CAPITAL LETTER I WITH DOT ABOVE
* 0130; T; 0069; # LATIN CAPITAL LETTER I WITH DOT ABOVE
* That's all.
* See CaseFolding.txt and the Unicode Standard for how to apply the case foldings.
*/
static const UChar32
simple[]={
/* input, default, exclude special i */
0x61, 0x61, 0x61,
0x49, 0x69, 0x131,
0x130, 0x130, 0x69,
0x131, 0x131, 0x131,
0xdf, 0xdf, 0xdf,
0xfb03, 0xfb03, 0xfb03,
0x1040e,0x10436,0x10436,
0x5ffff,0x5ffff,0x5ffff
};
static const UChar
mixed[]= { 0x61, 0x42, 0x130, 0x49, 0x131, 0x3d0, 0xdf, 0xfb03, 0xd93f, 0xdfff },
foldedDefault[]= { 0x61, 0x62, 0x69, 0x307, 0x69, 0x131, 0x3b2, 0x73, 0x73, 0x66, 0x66, 0x69, 0xd93f, 0xdfff },
foldedExcludeSpecialI[]={ 0x61, 0x62, 0x69, 0x131, 0x131, 0x3b2, 0x73, 0x73, 0x66, 0x66, 0x69, 0xd93f, 0xdfff };
UVersionInfo unicodeVersion={ 0, 0, 17, 89 }, unicode_3_1={ 3, 1, 0, 0 };
const UChar32 *p;
int32_t i;
UChar buffer[32];
int32_t length;
UErrorCode errorCode;
UBool isUnicode_3_1;
/* if unicodeVersion()>=3.1 then test exclude-special-i cases as well */
u_getUnicodeVersion(unicodeVersion);
isUnicode_3_1= uprv_memcmp(unicodeVersion, unicode_3_1, 4)>=0;
/* test simple case folding */
p=simple;
for(i=0; i<sizeof(simple)/12; p+=3, ++i) {
if(u_foldCase(p[0], U_FOLD_CASE_DEFAULT)!=p[1]) {
log_err("error: u_foldCase(0x%04lx, default)=0x%04lx instead of 0x%04lx\n",
p[0], u_foldCase(p[0], U_FOLD_CASE_DEFAULT), p[1]);
return;
}
if(isUnicode_3_1 && u_foldCase(p[0], U_FOLD_CASE_EXCLUDE_SPECIAL_I)!=p[2]) {
log_err("error: u_foldCase(0x%04lx, exclude special i)=0x%04lx instead of 0x%04lx\n",
p[0], u_foldCase(p[0], U_FOLD_CASE_EXCLUDE_SPECIAL_I), p[2]);
return;
}
}
/* test full string case folding with default option and separate buffers */
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
mixed, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_DEFAULT,
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(foldedDefault)/U_SIZEOF_UCHAR) ||
uprv_memcmp(foldedDefault, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strFoldCase(default)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(foldedDefault, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no");
}
/* exclude special i */
if(isUnicode_3_1) {
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
mixed, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_EXCLUDE_SPECIAL_I,
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(foldedExcludeSpecialI)/U_SIZEOF_UCHAR) ||
uprv_memcmp(foldedExcludeSpecialI, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strFoldCase(exclude special i)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(foldedExcludeSpecialI, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no");
}
}
/* test full string case folding with default option and in the same buffer */
uprv_memcpy(buffer, mixed, sizeof(mixed));
buffer[sizeof(mixed)/U_SIZEOF_UCHAR]=0;
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
buffer, -1, /* implicit srcLength */
U_FOLD_CASE_DEFAULT,
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(foldedDefault)/U_SIZEOF_UCHAR) ||
uprv_memcmp(foldedDefault, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strFoldCase(default same buffer)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(foldedDefault, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no");
}
/* test full string case folding, exclude special i, in the same buffer */
if(isUnicode_3_1) {
uprv_memcpy(buffer, mixed, sizeof(mixed));
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
buffer, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_EXCLUDE_SPECIAL_I,
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(foldedExcludeSpecialI)/U_SIZEOF_UCHAR) ||
uprv_memcmp(foldedExcludeSpecialI, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strFoldCase(exclude special i same buffer)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(foldedExcludeSpecialI, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no");
}
}
/* test preflighting */
buffer[0]=buffer[2]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, 2, /* set destCapacity=2 */
mixed, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_DEFAULT,
&errorCode);
if( errorCode!=U_BUFFER_OVERFLOW_ERROR ||
length!=(sizeof(foldedDefault)/U_SIZEOF_UCHAR) ||
uprv_memcmp(foldedDefault, buffer, 2*U_SIZEOF_UCHAR)!=0 ||
buffer[2]!=0xabcd
) {
log_err("error in u_strFoldCase(default preflighting)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(foldedDefault, buffer, 2*U_SIZEOF_UCHAR)==0 && buffer[2]==0xabcd ? "yes" : "no");
}
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(NULL, 0,
mixed, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_DEFAULT,
&errorCode);
if( errorCode!=U_BUFFER_OVERFLOW_ERROR ||
length!=(sizeof(foldedDefault)/U_SIZEOF_UCHAR)
) {
log_err("error in u_strFoldCase(default pure preflighting)=%ld error=%s\n",
length,
u_errorName(errorCode));
}
/* test error handling */
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(NULL, sizeof(buffer)/U_SIZEOF_UCHAR,
mixed, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_DEFAULT,
&errorCode);
if(errorCode!=U_ILLEGAL_ARGUMENT_ERROR) {
log_err("error in u_strFoldCase(default dest=NULL)=%ld error=%s\n",
length,
u_errorName(errorCode));
}
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, -1,
mixed, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_DEFAULT,
&errorCode);
if( errorCode!=U_ILLEGAL_ARGUMENT_ERROR ||
buffer[0]!=0xabcd
) {
log_err("error in u_strFoldCase(default destCapacity=-1)=%ld error=%s buffer[0]==0x%lx\n",
length,
u_errorName(errorCode),
buffer[0]);
}
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
NULL, sizeof(mixed)/U_SIZEOF_UCHAR,
U_FOLD_CASE_EXCLUDE_SPECIAL_I,
&errorCode);
if( errorCode!=U_ILLEGAL_ARGUMENT_ERROR ||
buffer[0]!=0xabcd
) {
log_err("error in u_strFoldCase(exclude special i src=NULL)=%ld error=%s buffer[0]==0x%lx\n",
length,
u_errorName(errorCode),
buffer[0]);
}
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
mixed, -2,
U_FOLD_CASE_EXCLUDE_SPECIAL_I,
&errorCode);
if( errorCode!=U_ILLEGAL_ARGUMENT_ERROR ||
buffer[0]!=0xabcd
) {
log_err("error in u_strFoldCase(exclude special i srcLength=-2)=%ld error=%s buffer[0]==0x%lx\n",
length,
u_errorName(errorCode),
buffer[0]);
}
}
static void
TestCaseDutchTitle(void) {
static const UChar
beforeTitle[]= { 0x69, 0x6A, 0x73, 0x73, 0x45, 0x6c, 0x20, 0x69, 0x67, 0x6c, 0x4f, 0x6f , 0x20 , 0x49, 0x4A, 0x53, 0x53, 0x45, 0x4C },
titleRoot[]= { 0x49, 0x6A, 0x73, 0x73, 0x65, 0x6c, 0x20, 0x49, 0x67, 0x6c, 0x6f, 0x6f , 0x20 , 0x49, 0x6A, 0x73, 0x73, 0x65, 0x6C },
titleDutch[]= { 0x49, 0x4A, 0x73, 0x73, 0x65, 0x6c, 0x20, 0x49, 0x67, 0x6c, 0x6f, 0x6f , 0x20 , 0x49, 0x4A, 0x73, 0x73, 0x65, 0x6C };
UChar buffer[32];
UBreakIterator *titleIterWord;
int32_t length;
UErrorCode errorCode;
errorCode=U_ZERO_ERROR;
titleIterWord=ubrk_open(UBRK_WORD, "", beforeTitle, sizeof(beforeTitle)/U_SIZEOF_UCHAR, &errorCode);
if(U_FAILURE(errorCode)) {
log_err_status(errorCode, "error: ubrk_open(UBRK_WORD)->%s\n", u_errorName(errorCode));
return;
}
/* titlecase with default locale */
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
beforeTitle, sizeof(beforeTitle)/U_SIZEOF_UCHAR,
titleIterWord, "",
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(titleRoot)/U_SIZEOF_UCHAR) ||
uprv_memcmp(titleRoot, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
char charsOut[21];
u_UCharsToChars(buffer,charsOut,sizeof(charsOut));
log_err("error in u_strToTitle(UBRK_CHARACTERS)=%ld error=%s root locale string matches: %s\noutput buffer is {%s}\n",
length,
u_errorName(errorCode),
uprv_memcmp(titleRoot, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no", charsOut);
}
/* titlecase with Dutch locale */
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
beforeTitle, sizeof(beforeTitle)/U_SIZEOF_UCHAR,
titleIterWord, "nl",
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(titleDutch)/U_SIZEOF_UCHAR) ||
uprv_memcmp(titleDutch, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
char charsOut[21];
u_UCharsToChars(buffer,charsOut,sizeof(charsOut));
log_err("error in u_strToTitle(UBRK_CHARACTERS)=%ld error=%s dutch locale string matches: %s\noutput buffer is {%s}\n",
length,
u_errorName(errorCode),
uprv_memcmp(titleDutch, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no", charsOut);
}
ubrk_close(titleIterWord);
}
static void
TestCaseLower(void) {
static const UChar
beforeLower[]= { 0x61, 0x42, 0x49, 0x3a3, 0xdf, 0x3a3, 0x2f, 0xd93f, 0xdfff },
lowerRoot[]= { 0x61, 0x62, 0x69, 0x3c3, 0xdf, 0x3c2, 0x2f, 0xd93f, 0xdfff },
lowerTurkish[]={ 0x61, 0x62, 0x131, 0x3c3, 0xdf, 0x3c2, 0x2f, 0xd93f, 0xdfff };
UChar buffer[32];
int32_t length;
UErrorCode errorCode;
/* lowercase with root locale and separate buffers */
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToLower(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
beforeLower, sizeof(beforeLower)/U_SIZEOF_UCHAR,
"",
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(lowerRoot)/U_SIZEOF_UCHAR) ||
uprv_memcmp(lowerRoot, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strToLower(root locale)=%ld error=%s string matches: %s\t\nlowerRoot=%s\t\nbuffer=%s\n",
length,
u_errorName(errorCode),
uprv_memcmp(lowerRoot, buffer, length*U_SIZEOF_UCHAR)==0 &&
buffer[length]==0 ? "yes" : "no",
aescstrdup(lowerRoot,-1),
aescstrdup(buffer,-1));
}
/* lowercase with turkish locale and in the same buffer */
uprv_memcpy(buffer, beforeLower, sizeof(beforeLower));
buffer[sizeof(beforeLower)/U_SIZEOF_UCHAR]=0;
errorCode=U_ZERO_ERROR;
length=u_strToLower(buffer, sizeof(buffer)/U_SIZEOF_UCHAR,
buffer, -1, /* implicit srcLength */
"tr",
&errorCode);
if( U_FAILURE(errorCode) ||
length!=(sizeof(lowerTurkish)/U_SIZEOF_UCHAR) ||
uprv_memcmp(lowerTurkish, buffer, length*U_SIZEOF_UCHAR)!=0 ||
buffer[length]!=0
) {
log_err("error in u_strToLower(turkish locale)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(lowerTurkish, buffer, length*U_SIZEOF_UCHAR)==0 && buffer[length]==0 ? "yes" : "no");
}
/* test preflighting */
buffer[0]=buffer[2]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToLower(buffer, 2, /* set destCapacity=2 */
beforeLower, sizeof(beforeLower)/U_SIZEOF_UCHAR,
"",
&errorCode);
if( errorCode!=U_BUFFER_OVERFLOW_ERROR ||
length!=(sizeof(lowerRoot)/U_SIZEOF_UCHAR) ||
uprv_memcmp(lowerRoot, buffer, 2*U_SIZEOF_UCHAR)!=0 ||
buffer[2]!=0xabcd
) {
log_err("error in u_strToLower(root locale preflighting)=%ld error=%s string matches: %s\n",
length,
u_errorName(errorCode),
uprv_memcmp(lowerRoot, buffer, 2*U_SIZEOF_UCHAR)==0 && buffer[2]==0xabcd ? "yes" : "no");
}
/* test error handling */
errorCode=U_ZERO_ERROR;
length=u_strToLower(NULL, sizeof(buffer)/U_SIZEOF_UCHAR,
beforeLower, sizeof(beforeLower)/U_SIZEOF_UCHAR,
"",
&errorCode);
if(errorCode!=U_ILLEGAL_ARGUMENT_ERROR) {
log_err("error in u_strToLower(root locale dest=NULL)=%ld error=%s\n",
length,
u_errorName(errorCode));
}
buffer[0]=0xabcd;
errorCode=U_ZERO_ERROR;
length=u_strToLower(buffer, -1,
beforeLower, sizeof(beforeLower)/U_SIZEOF_UCHAR,
"",
&errorCode);
if( errorCode!=U_ILLEGAL_ARGUMENT_ERROR ||
buffer[0]!=0xabcd
) {
log_err("error in u_strToLower(root locale destCapacity=-1)=%ld error=%s buffer[0]==0x%lx\n",
length,
u_errorName(errorCode),
buffer[0]);
}
}
