/**
* Get a set containing the expansions defined by the collator. The set includes
* both the UCA expansions and the expansions defined by the tailoring
* @param coll collator
* @param conts the set to hold the result
* @param addPrefixes add the prefix contextual elements to contractions
* @param status to hold the error code
*
* @draft ICU 3.4
*/
U_CAPI void U_EXPORT2
ucol_getContractionsAndExpansions( const UCollator *coll,
USet *contractions,
USet *expansions,
UBool addPrefixes,
UErrorCode *status)
{
if(U_FAILURE(*status)) {
return;
}
if(coll == NULL) {
*status = U_ILLEGAL_ARGUMENT_ERROR;
return;
}
if(contractions) {
uset_clear(contractions);
}
if(expansions) {
uset_clear(expansions);
}
int32_t rulesLen = 0;
const UChar* rules = ucol_getRules(coll, &rulesLen);
UColTokenParser src;
ucol_tok_initTokenList(&src, rules, rulesLen, coll->UCA,
ucol_tok_getRulesFromBundle, NULL, status);
contContext c = { NULL, contractions, expansions, src.removeSet, addPrefixes, status };
// Add the UCA contractions
c.coll = coll->UCA;
utrie_enum(&coll->UCA->mapping, NULL, _processSpecials, &c);
// This is collator specific. Add contractions from a collator
c.coll = coll;
c.removedContractions = NULL;
utrie_enum(&coll->mapping, NULL, _processSpecials, &c);
ucol_tok_closeTokenList(&src);
}
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 bool __CFLocaleCopyExemplarCharSet(CFLocaleRef locale, bool user, CFTypeRef *cf, CFStringRef context) {
char localeID[ULOC_FULLNAME_CAPACITY+ULOC_KEYWORD_AND_VALUES_CAPACITY];
if (CFStringGetCString(locale->_identifier, localeID, sizeof(localeID)/sizeof(char), kCFStringEncodingASCII)) {
UErrorCode icuStatus = U_ZERO_ERROR;
ULocaleData* uld = ulocdata_open(localeID, &icuStatus);
USet *set = ulocdata_getExemplarSet(uld, NULL, USET_ADD_CASE_MAPPINGS, ULOCDATA_ES_STANDARD, &icuStatus);
ulocdata_close(uld);
if (U_FAILURE(icuStatus))
return false;
if (icuStatus == U_USING_DEFAULT_WARNING) // If default locale used, force to empty set
uset_clear(set);
*cf = (CFTypeRef) _CFCreateCharacterSetFromUSet(set);
uset_close(set);
return (*cf != NULL);
}
return false;
}
U_CAPI void U_EXPORT2
ucnv_getUnicodeSet(const UConverter *cnv,
USet *setFillIn,
UConverterUnicodeSet whichSet,
UErrorCode *pErrorCode) {
/* argument checking */
if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
return;
}
if(cnv==NULL || setFillIn==NULL || whichSet<UCNV_ROUNDTRIP_SET || UCNV_SET_COUNT<=whichSet) {
*pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
return;
}
/* does this converter support this function? */
if(cnv->sharedData->impl->getUnicodeSet==NULL) {
*pErrorCode=U_UNSUPPORTED_ERROR;
return;
}
{
USetAdder sa={
NULL,
uset_add,
uset_addRange,
uset_addString,
uset_remove,
uset_removeRange
};
sa.set=setFillIn;
/* empty the set */
uset_clear(setFillIn);
/* call the converter to add the code points it supports */
cnv->sharedData->impl->getUnicodeSet(cnv, &sa, whichSet, pErrorCode);
}
}
/**
* Basic API test for uset.x
*/
static void TestAPI() {
USet* set;
USet* set2;
UErrorCode ec;
/* [] */
set = uset_openEmpty();
expect(set, "", "abc{ab}", NULL);
uset_close(set);
set = uset_open(1, 0);
expect(set, "", "abc{ab}", NULL);
uset_close(set);
set = uset_open(1, 1);
uset_clear(set);
expect(set, "", "abc{ab}", NULL);
uset_close(set);
/* [ABC] */
set = uset_open(0x0041, 0x0043);
expect(set, "ABC", "DEF{ab}", NULL);
uset_close(set);
/* [a-c{ab}] */
ec = U_ZERO_ERROR;
set = uset_openPattern(PAT, PAT_LEN, &ec);
if(U_FAILURE(ec)) {
log_err("uset_openPattern([a-c{ab}]) failed - %s\n", u_errorName(ec));
return;
}
if(!uset_resemblesPattern(PAT, PAT_LEN, 0)) {
log_err("uset_resemblesPattern of PAT failed\n");
}
expect(set, "abc{ab}", "def{bc}", &ec);
/* [a-d{ab}] */
uset_add(set, 0x64);
expect(set, "abcd{ab}", "ef{bc}", NULL);
/* [acd{ab}{bc}] */
uset_remove(set, 0x62);
uset_addString(set, STR_bc, STR_bc_LEN);
expect(set, "acd{ab}{bc}", "bef{cd}", NULL);
/* [acd{bc}] */
uset_removeString(set, STR_ab, STR_ab_LEN);
expect(set, "acd{bc}", "bfg{ab}", NULL);
/* [^acd{bc}] */
uset_complement(set);
expect(set, "bef{bc}", "acd{ac}", NULL);
/* [a-e{bc}] */
uset_complement(set);
uset_addRange(set, 0x0062, 0x0065);
expect(set, "abcde{bc}", "fg{ab}", NULL);
/* [de{bc}] */
uset_removeRange(set, 0x0050, 0x0063);
expect(set, "de{bc}", "bcfg{ab}", NULL);
/* [g-l] */
uset_set(set, 0x0067, 0x006C);
expect(set, "ghijkl", "de{bc}", NULL);
if (uset_indexOf(set, 0x0067) != 0) {
log_err("uset_indexOf failed finding correct index of 'g'\n");
}
if (uset_charAt(set, 0) != 0x0067) {
log_err("uset_charAt failed finding correct char 'g' at index 0\n");
}
/* How to test this one...? */
uset_compact(set);
/* [g-i] */
uset_retain(set, 0x0067, 0x0069);
expect(set, "ghi", "dejkl{bc}", NULL);
/* UCHAR_ASCII_HEX_DIGIT */
uset_applyIntPropertyValue(set, UCHAR_ASCII_HEX_DIGIT, 1, &ec);
if(U_FAILURE(ec)) {
log_err("uset_applyIntPropertyValue([UCHAR_ASCII_HEX_DIGIT]) failed - %s\n", u_errorName(ec));
return;
}
expect(set, "0123456789ABCDEFabcdef", "GHIjkl{bc}", NULL);
/* [ab] */
uset_clear(set);
uset_addAllCodePoints(set, STR_ab, STR_ab_LEN);
expect(set, "ab", "def{ab}", NULL);
if (uset_containsAllCodePoints(set, STR_bc, STR_bc_LEN)){
log_err("set should not conatin all characters of \"bc\" \n");
}
/* [] */
set2 = uset_open(1, 1);
uset_clear(set2);
/* space */
uset_applyPropertyAlias(set2, PAT_lb, PAT_lb_LEN, VAL_SP, VAL_SP_LEN, &ec);
expect(set2, " ", "abcdefghi{bc}", NULL);
/* [a-c] */
uset_set(set2, 0x0061, 0x0063);
/* [g-i] */
uset_set(set, 0x0067, 0x0069);
/* [a-c g-i] */
if (uset_containsSome(set, set2)) {
log_err("set should not contain some of set2 yet\n");
}
uset_complementAll(set, set2);
if (!uset_containsSome(set, set2)) {
log_err("set should contain some of set2\n");
}
expect(set, "abcghi", "def{bc}", NULL);
/* [g-i] */
uset_removeAll(set, set2);
expect(set, "ghi", "abcdef{bc}", NULL);
/* [a-c g-i] */
uset_addAll(set2, set);
expect(set2, "abcghi", "def{bc}", NULL);
/* [g-i] */
uset_retainAll(set2, set);
expect(set2, "ghi", "abcdef{bc}", NULL);
uset_close(set);
uset_close(set2);
}
