U_CAPI USet* U_EXPORT2
ulocdata_getExemplarSet(ULocaleData *uld, USet *fillIn,
uint32_t options, ULocaleDataExemplarSetType extype, UErrorCode *status){
static const char* const exemplarSetTypes[] = { "ExemplarCharacters", "AuxExemplarCharacters" };
const UChar *exemplarChars = NULL;
int32_t len = 0;
UErrorCode localStatus = U_ZERO_ERROR;
if (U_FAILURE(*status))
return NULL;
exemplarChars = ures_getStringByKey(uld->bundle, exemplarSetTypes[extype], &len, &localStatus);
if ( (localStatus == U_USING_DEFAULT_WARNING) && uld->noSubstitute ) {
localStatus = U_MISSING_RESOURCE_ERROR;
}
if (localStatus != U_ZERO_ERROR) {
*status = localStatus;
}
if (U_FAILURE(*status))
return NULL;
if(fillIn != NULL)
uset_applyPattern(fillIn, exemplarChars, len,
USET_IGNORE_SPACE | options, status);
else
fillIn = uset_openPatternOptions(exemplarChars, len,
USET_IGNORE_SPACE | options, status);
return fillIn;
}
U_CAPI int32_t U_EXPORT2
ucol_getUnsafeSet( const UCollator *coll,
USet *unsafe,
UErrorCode *status)
{
UChar buffer[internalBufferSize];
int32_t len = 0;
uset_clear(unsafe);
// cccpattern = "[[:^tccc=0:][:^lccc=0:]]", unfortunately variant
static const UChar cccpattern[25] = { 0x5b, 0x5b, 0x3a, 0x5e, 0x74, 0x63, 0x63, 0x63, 0x3d, 0x30, 0x3a, 0x5d,
0x5b, 0x3a, 0x5e, 0x6c, 0x63, 0x63, 0x63, 0x3d, 0x30, 0x3a, 0x5d, 0x5d, 0x00 };
// add chars that fail the fcd check
uset_applyPattern(unsafe, cccpattern, 24, USET_IGNORE_SPACE, status);
// add Thai/Lao prevowels
uset_addRange(unsafe, 0xe40, 0xe44);
uset_addRange(unsafe, 0xec0, 0xec4);
// add lead/trail surrogates
uset_addRange(unsafe, 0xd800, 0xdfff);
USet *contractions = uset_open(0,0);
int32_t i = 0, j = 0;
int32_t contsSize = ucol_getContractions(coll, contractions, status);
UChar32 c = 0;
// Contraction set consists only of strings
// to get unsafe code points, we need to
// break the strings apart and add them to the unsafe set
for(i = 0; i < contsSize; i++) {
len = uset_getItem(contractions, i, NULL, NULL, buffer, internalBufferSize, status);
if(len > 0) {
j = 0;
while(j < len) {
U16_NEXT(buffer, j, len, c);
if(j < len) {
uset_add(unsafe, c);
}
}
}
}
uset_close(contractions);
return uset_size(unsafe);
}
static int32_t
u_scanf_scanset_handler(UFILE *input,
u_scanf_spec_info *info,
ufmt_args *args,
const UChar *fmt,
int32_t *fmtConsumed,
int32_t *argConverted)
{
USet *scanset;
UErrorCode status = U_ZERO_ERROR;
int32_t chLeft = INT32_MAX;
UChar32 c;
UChar *alias = (UChar*) (args[0].ptrValue);
UBool isNotEOF = FALSE;
UBool readCharacter = FALSE;
/* Create an empty set */
scanset = uset_open(0, -1);
/* Back up one to get the [ */
fmt--;
/* truncate to the width, if specified and alias the target */
if(info->fWidth >= 0) {
chLeft = info->fWidth;
}
/* parse the scanset from the fmt string */
*fmtConsumed = uset_applyPattern(scanset, fmt, -1, 0, &status);
/* verify that the parse was successful */
if (U_SUCCESS(status)) {
c=0;
/* grab characters one at a time and make sure they are in the scanset */
while(chLeft > 0) {
if ((isNotEOF = ufile_getch32(input, &c)) && uset_contains(scanset, c)) {
readCharacter = TRUE;
if (!info->fSkipArg) {
int32_t idx = 0;
UBool isError = FALSE;
U16_APPEND(alias, idx, chLeft, c, isError);
if (isError) {
break;
}
alias += idx;
}
chLeft -= (1 + U_IS_SUPPLEMENTARY(c));
}
else {
/* if the character's not in the scanset, break out */
break;
}
}
/* put the final character we read back on the input */
if(isNotEOF && chLeft > 0) {
u_fungetc(c, input);
}
}
uset_close(scanset);
/* if we didn't match at least 1 character, fail */
if(!readCharacter)
return -1;
/* otherwise, add the terminator */
else if (!info->fSkipArg) {
*alias = 0x00;
}
/* we converted 1 arg */
*argConverted = !info->fSkipArg;
return (info->fWidth >= 0 ? info->fWidth : INT32_MAX) - chLeft;
}