int u_memcasecmp(const UChar* a, const UChar* b, int length, unsigned options)
{
for (int i = 0; i < length; ++i) {
UChar c1 = u_foldCase(a[i], options);
UChar c2 = u_foldCase(b[i], options);
if (c1 != c2)
return c1 - c2;
}
return 0;
}
int ICUTransService::compareNIString(const XMLCh* const comp1
, const XMLCh* const comp2
, const unsigned int maxChars)
{
if (maxChars > 0)
{
// Note that this function has somewhat broken semantics, as it's
// possible for two strings of different lengths to compare as equal
// in a case-insensitive manner, since one character could be
// represented as a surrogate pair.
size_t i = 0;
size_t j = 0;
for(;;)
{
UChar32 ch1;
UChar32 ch2;
U16_NEXT_UNSAFE(comp1, i, ch1);
U16_NEXT_UNSAFE(comp2, j, ch2);
const UChar32 folded1 =
u_foldCase(ch1, U_FOLD_CASE_DEFAULT);
const UChar32 folded2 =
u_foldCase(ch2, U_FOLD_CASE_DEFAULT);
if (folded1 != folded2)
{
return folded1 - folded2;
}
else if (i == maxChars)
{
// If we're at the end of both strings, return 0.
// Otherwise, we've run out of characters in the
// left string, so return -1.
return j == maxChars ? 0 : -1;
}
else if (j == maxChars)
{
// We've run out of characters in the right string,
// but not the left, so return 1.
return 1;
}
}
}
return 0;
}
std::string foldcase(const std::string& str)
{
return transform(str, [](uint32_t cp)
{
return u_foldCase(static_cast<UChar32>(cp),
U_FOLD_CASE_DEFAULT);
});
}
Variant HHVM_STATIC_METHOD(IntlChar, foldCase,
const Variant& arg, int64_t options) {
GETCP(arg, cp);
auto ret = u_foldCase(cp, options);
if (arg.isString()) {
char buffer[5];
int buffer_len = 0;
U8_APPEND_UNSAFE(buffer, buffer_len, ret);
return String(buffer, buffer_len, CopyString);
} else {
return ret;
}
}
// ---------------------------------------------------------------------------
// ICUTransService: The virtual transcoding service API
// ---------------------------------------------------------------------------
int ICUTransService::compareIString(const XMLCh* const comp1
, const XMLCh* const comp2)
{
size_t i = 0;
size_t j = 0;
for(;;)
{
UChar32 ch1;
UChar32 ch2;
U16_NEXT_UNSAFE(comp1, i, ch1);
U16_NEXT_UNSAFE(comp2, j, ch2);
const UChar32 folded1 =
u_foldCase(ch1, U_FOLD_CASE_DEFAULT);
const UChar32 folded2 =
u_foldCase(ch2, U_FOLD_CASE_DEFAULT);
if (folded1 !=
folded2)
{
return folded1 - folded2;
}
else if (ch1 == 0)
{
// If ch1 is 0, the ch2 must also be
// 0. Otherwise, the previous if
// would have failed.
break;
}
}
return 0;
}
/*
** Prepare to begin tokenizing a particular string. The input
** string to be tokenized is pInput[0..nBytes-1]. A cursor
** used to incrementally tokenize this string is returned in
** *ppCursor.
*/
static int icuOpen(
sqlite3_tokenizer *pTokenizer, /* The tokenizer */
const char *zInput, /* Input string */
int nInput, /* Length of zInput in bytes */
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
){
IcuTokenizer *p = (IcuTokenizer *)pTokenizer;
IcuCursor *pCsr;
const int32_t opt = U_FOLD_CASE_DEFAULT;
UErrorCode status = U_ZERO_ERROR;
int nChar;
UChar32 c;
int iInput = 0;
int iOut = 0;
*ppCursor = 0;
if( nInput<0 ){
nInput = strlen(zInput);
}
nChar = nInput+1;
pCsr = (IcuCursor *)sqlite3_malloc(
sizeof(IcuCursor) + /* IcuCursor */
nChar * sizeof(UChar) + /* IcuCursor.aChar[] */
(nChar+1) * sizeof(int) /* IcuCursor.aOffset[] */
);
if( !pCsr ){
return SQLITE_NOMEM;
}
memset(pCsr, 0, sizeof(IcuCursor));
pCsr->aChar = (UChar *)&pCsr[1];
pCsr->aOffset = (int *)&pCsr->aChar[nChar];
pCsr->aOffset[iOut] = iInput;
U8_NEXT(zInput, iInput, nInput, c);
while( c>0 ){
int isError = 0;
c = u_foldCase(c, opt);
U16_APPEND(pCsr->aChar, iOut, nChar, c, isError);
if( isError ){
sqlite3_free(pCsr);
return SQLITE_ERROR;
}
pCsr->aOffset[iOut] = iInput;
if( iInput<nInput ){
U8_NEXT(zInput, iInput, nInput, c);
}else{
c = 0;
}
}
pCsr->pIter = ubrk_open(UBRK_WORD, p->zLocale, pCsr->aChar, iOut, &status);
if( !U_SUCCESS(status) ){
sqlite3_free(pCsr);
return SQLITE_ERROR;
}
pCsr->nChar = iOut;
ubrk_first(pCsr->pIter);
*ppCursor = (sqlite3_tokenizer_cursor *)pCsr;
return SQLITE_OK;
}
/*
** Compare two UTF-8 strings for equality where the first string is
** a "LIKE" expression. Return true (1) if they are the same and
** false (0) if they are different.
*/
static int icuLikeCompare(
const uint8_t *zPattern, /* LIKE pattern */
const uint8_t *zString, /* The UTF-8 string to compare against */
const UChar32 uEsc /* The escape character */
){
static const int MATCH_ONE = (UChar32)'_';
static const int MATCH_ALL = (UChar32)'%';
int iPattern = 0; /* Current byte index in zPattern */
int iString = 0; /* Current byte index in zString */
int prevEscape = 0; /* True if the previous character was uEsc */
while( zPattern[iPattern]!=0 ){
/* Read (and consume) the next character from the input pattern. */
UChar32 uPattern;
U8_NEXT_UNSAFE(zPattern, iPattern, uPattern);
assert(uPattern!=0);
/* There are now 4 possibilities:
**
** 1. uPattern is an unescaped match-all character "%",
** 2. uPattern is an unescaped match-one character "_",
** 3. uPattern is an unescaped escape character, or
** 4. uPattern is to be handled as an ordinary character
*/
if( !prevEscape && uPattern==MATCH_ALL ){
/* Case 1. */
uint8_t c;
/* Skip any MATCH_ALL or MATCH_ONE characters that follow a
** MATCH_ALL. For each MATCH_ONE, skip one character in the
** test string.
*/
while( (c=zPattern[iPattern]) == MATCH_ALL || c == MATCH_ONE ){
if( c==MATCH_ONE ){
if( zString[iString]==0 ) return 0;
U8_FWD_1_UNSAFE(zString, iString);
}
iPattern++;
}
if( zPattern[iPattern]==0 ) return 1;
while( zString[iString] ){
if( icuLikeCompare(&zPattern[iPattern], &zString[iString], uEsc) ){
return 1;
}
U8_FWD_1_UNSAFE(zString, iString);
}
return 0;
}else if( !prevEscape && uPattern==MATCH_ONE ){
/* Case 2. */
if( zString[iString]==0 ) return 0;
U8_FWD_1_UNSAFE(zString, iString);
}else if( !prevEscape && uPattern==uEsc){
/* Case 3. */
prevEscape = 1;
}else{
/* Case 4. */
UChar32 uString;
U8_NEXT_UNSAFE(zString, iString, uString);
uString = u_foldCase(uString, U_FOLD_CASE_DEFAULT);
uPattern = u_foldCase(uPattern, U_FOLD_CASE_DEFAULT);
if( uString!=uPattern ){
return 0;
}
prevEscape = 0;
}
}
return zString[iString]==0;
}
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]);
}
}
/*
* imp: common/ucase.c
* hdr: common/unicode/uchar.h
* @stable ICU 2.0
*/
U_CAPI UChar32 U_EXPORT2
u_foldCase_4_0(UChar32 c, uint32_t options) {
return u_foldCase(c, options);
}
/*
** Compare two UTF-8 strings for equality where the first string is
** a "LIKE" expression. Return true (1) if they are the same and
** false (0) if they are different.
*/
static int icuLikeCompare(
const uint8_t *zPattern, /* LIKE pattern */
const uint8_t *zString, /* The UTF-8 string to compare against */
const UChar32 uEsc /* The escape character */
){
static const int MATCH_ONE = (UChar32)'_';
static const int MATCH_ALL = (UChar32)'%';
int prevEscape = 0; /* True if the previous character was uEsc */
while( 1 ){
/* Read (and consume) the next character from the input pattern. */
UChar32 uPattern;
SQLITE_ICU_READ_UTF8(zPattern, uPattern);
if( uPattern==0 ) break;
/* There are now 4 possibilities:
**
** 1. uPattern is an unescaped match-all character "%",
** 2. uPattern is an unescaped match-one character "_",
** 3. uPattern is an unescaped escape character, or
** 4. uPattern is to be handled as an ordinary character
*/
if( !prevEscape && uPattern==MATCH_ALL ){
/* Case 1. */
uint8_t c;
/* Skip any MATCH_ALL or MATCH_ONE characters that follow a
** MATCH_ALL. For each MATCH_ONE, skip one character in the
** test string.
*/
while( (c=*zPattern) == MATCH_ALL || c == MATCH_ONE ){
if( c==MATCH_ONE ){
if( *zString==0 ) return 0;
SQLITE_ICU_SKIP_UTF8(zString);
}
zPattern++;
}
if( *zPattern==0 ) return 1;
while( *zString ){
if( icuLikeCompare(zPattern, zString, uEsc) ){
return 1;
}
SQLITE_ICU_SKIP_UTF8(zString);
}
return 0;
}else if( !prevEscape && uPattern==MATCH_ONE ){
/* Case 2. */
if( *zString==0 ) return 0;
SQLITE_ICU_SKIP_UTF8(zString);
}else if( !prevEscape && uPattern==uEsc){
/* Case 3. */
prevEscape = 1;
}else{
/* Case 4. */
UChar32 uString;
SQLITE_ICU_READ_UTF8(zString, uString);
uString = u_foldCase(uString, U_FOLD_CASE_DEFAULT);
uPattern = u_foldCase(uPattern, U_FOLD_CASE_DEFAULT);
if( uString!=uPattern ){
return 0;
}
prevEscape = 0;
}
}
return *zString==0;
}
static void demoCaseMapInC() {
/*
* input=
* "aB<capital sigma>"
* "iI<small dotless i><capital dotted I> "
* "<sharp s> <small lig. ffi>"
* "<small final sigma><small sigma><capital sigma>"
*/
static const UChar input[]={
0x61, 0x42, 0x3a3,
0x69, 0x49, 0x131, 0x130, 0x20,
0xdf, 0x20, 0xfb03,
0x3c2, 0x3c3, 0x3a3, 0
};
UChar buffer[32];
UErrorCode errorCode;
UChar32 c;
int32_t i, j, length;
UBool isError;
printf("\n* demoCaseMapInC() ----------------- ***\n\n");
/*
* First, use simple case mapping functions which provide
* 1:1 code point mappings without context/locale ID.
*
* Note that some mappings will not be "right" because some "real"
* case mappings require context, depend on the locale ID,
* and/or result in a change in the number of code points.
*/
printUString("input string: ", input, -1);
/* uppercase */
isError=FALSE;
for(i=j=0; j<UPRV_LENGTHOF(buffer) && !isError; /* U16_NEXT post-increments */) {
U16_NEXT(input, i, INT32_MAX, c); /* without length because NUL-terminated */
if(c==0) {
break; /* stop at terminating NUL, no need to terminate buffer */
}
c=u_toupper(c);
U16_APPEND(buffer, j, UPRV_LENGTHOF(buffer), c, isError);
}
printUString("simple-uppercased: ", buffer, j);
/* lowercase */
isError=FALSE;
for(i=j=0; j<UPRV_LENGTHOF(buffer) && !isError; /* U16_NEXT post-increments */) {
U16_NEXT(input, i, INT32_MAX, c); /* without length because NUL-terminated */
if(c==0) {
break; /* stop at terminating NUL, no need to terminate buffer */
}
c=u_tolower(c);
U16_APPEND(buffer, j, UPRV_LENGTHOF(buffer), c, isError);
}
printUString("simple-lowercased: ", buffer, j);
/* titlecase */
isError=FALSE;
for(i=j=0; j<UPRV_LENGTHOF(buffer) && !isError; /* U16_NEXT post-increments */) {
U16_NEXT(input, i, INT32_MAX, c); /* without length because NUL-terminated */
if(c==0) {
break; /* stop at terminating NUL, no need to terminate buffer */
}
c=u_totitle(c);
U16_APPEND(buffer, j, UPRV_LENGTHOF(buffer), c, isError);
}
printUString("simple-titlecased: ", buffer, j);
/* case-fold/default */
isError=FALSE;
for(i=j=0; j<UPRV_LENGTHOF(buffer) && !isError; /* U16_NEXT post-increments */) {
U16_NEXT(input, i, INT32_MAX, c); /* without length because NUL-terminated */
if(c==0) {
break; /* stop at terminating NUL, no need to terminate buffer */
}
c=u_foldCase(c, U_FOLD_CASE_DEFAULT);
U16_APPEND(buffer, j, UPRV_LENGTHOF(buffer), c, isError);
}
printUString("simple-case-folded/default: ", buffer, j);
/* case-fold/Turkic */
isError=FALSE;
for(i=j=0; j<UPRV_LENGTHOF(buffer) && !isError; /* U16_NEXT post-increments */) {
U16_NEXT(input, i, INT32_MAX, c); /* without length because NUL-terminated */
if(c==0) {
break; /* stop at terminating NUL, no need to terminate buffer */
}
c=u_foldCase(c, U_FOLD_CASE_EXCLUDE_SPECIAL_I);
U16_APPEND(buffer, j, UPRV_LENGTHOF(buffer), c, isError);
}
printUString("simple-case-folded/Turkic: ", buffer, j);
/*
* Second, use full case mapping functions which provide
* 1:n code point mappings (n can be 0!) and are sensitive to context and locale ID.
*
* Note that lower/upper/titlecasing take a locale ID while case-folding
* has bit flag options instead, by design of the Unicode SpecialCasing.txt UCD file.
*
* Also, string titlecasing requires a BreakIterator to find starts of words.
* The sample code here passes in a NULL pointer; u_strToTitle() will open and close a default
* titlecasing BreakIterator automatically.
* For production code where many strings are titlecased it would be more efficient
* to open a BreakIterator externally and pass it in.
*/
printUString("\ninput string: ", input, -1);
/* lowercase/English */
errorCode=U_ZERO_ERROR;
length=u_strToLower(buffer, UPRV_LENGTHOF(buffer), input, -1, "en", &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-lowercased/en: ", buffer, length);
} else {
printf("error in u_strToLower(en)=%ld error=%s\n", length, u_errorName(errorCode));
}
/* lowercase/Turkish */
errorCode=U_ZERO_ERROR;
length=u_strToLower(buffer, UPRV_LENGTHOF(buffer), input, -1, "tr", &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-lowercased/tr: ", buffer, length);
} else {
printf("error in u_strToLower(tr)=%ld error=%s\n", length, u_errorName(errorCode));
}
/* uppercase/English */
errorCode=U_ZERO_ERROR;
length=u_strToUpper(buffer, UPRV_LENGTHOF(buffer), input, -1, "en", &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-uppercased/en: ", buffer, length);
} else {
printf("error in u_strToUpper(en)=%ld error=%s\n", length, u_errorName(errorCode));
}
/* uppercase/Turkish */
errorCode=U_ZERO_ERROR;
length=u_strToUpper(buffer, UPRV_LENGTHOF(buffer), input, -1, "tr", &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-uppercased/tr: ", buffer, length);
} else {
printf("error in u_strToUpper(tr)=%ld error=%s\n", length, u_errorName(errorCode));
}
/* titlecase/English */
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, UPRV_LENGTHOF(buffer), input, -1, NULL, "en", &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-titlecased/en: ", buffer, length);
} else {
printf("error in u_strToTitle(en)=%ld error=%s\n", length, u_errorName(errorCode));
}
/* titlecase/Turkish */
errorCode=U_ZERO_ERROR;
length=u_strToTitle(buffer, UPRV_LENGTHOF(buffer), input, -1, NULL, "tr", &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-titlecased/tr: ", buffer, length);
} else {
printf("error in u_strToTitle(tr)=%ld error=%s\n", length, u_errorName(errorCode));
}
/* case-fold/default */
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, UPRV_LENGTHOF(buffer), input, -1, U_FOLD_CASE_DEFAULT, &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-case-folded/default: ", buffer, length);
} else {
printf("error in u_strFoldCase(default)=%ld error=%s\n", length, u_errorName(errorCode));
}
/* case-fold/Turkic */
errorCode=U_ZERO_ERROR;
length=u_strFoldCase(buffer, UPRV_LENGTHOF(buffer), input, -1, U_FOLD_CASE_EXCLUDE_SPECIAL_I, &errorCode);
if(U_SUCCESS(errorCode)) {
printUString("full-case-folded/Turkic: ", buffer, length);
} else {
printf("error in u_strFoldCase(Turkic)=%ld error=%s\n", length, u_errorName(errorCode));
}
}