void SetMonkey::append(UnicodeString &test, UnicodeString &alternate)
{
int32_t size = uset_size(set);
int32_t index = m_rand() % size;
UChar32 ch = uset_charAt(set, index);
UnicodeString str(ch);
test.append(str);
alternate.append(str); // flip case, or some junk?
}
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 void expectItems(const USet* set,
const char* items) {
const char* p = items;
UChar ustr[4096], itemStr[4096];
char buf[4096];
char *pat;
UErrorCode ec;
int32_t expectedSize = 0;
int32_t itemCount = uset_getItemCount(set);
int32_t itemIndex = 0;
UChar32 start = 1, end = 0;
int32_t itemLen = 0, length;
ec = U_ZERO_ERROR;
length = uset_toPattern(set, ustr, sizeof(ustr), TRUE, &ec);
if (U_FAILURE(ec)) {
log_err("FAIL: uset_toPattern => %s\n", u_errorName(ec));
return;
}
pat=aescstrdup(ustr, length);
if (uset_isEmpty(set) != (strlen(items)==0)) {
log_data_err("FAIL: %s should return %s from isEmpty (Are you missing data?)\n",
pat,
strlen(items)==0 ? "TRUE" : "FALSE");
}
/* Don't test patterns starting with "[^" */
if (u_strlen(ustr) > 2 && ustr[1] == 0x5e /*'^'*/) {
return;
}
while (*p) {
++expectedSize;
if (start > end || start == -1) {
/* Fetch our next item */
if (itemIndex >= itemCount) {
log_data_err("FAIL: ran out of items iterating %s (Are you missing data?)\n", pat);
return;
}
itemLen = uset_getItem(set, itemIndex, &start, &end,
itemStr, sizeof(itemStr), &ec);
if (U_FAILURE(ec) || itemLen < 0) {
log_err("FAIL: uset_getItem => %s\n", u_errorName(ec));
return;
}
if (itemLen == 0) {
log_verbose("Ok: %s item %d is %c-%c\n", pat,
itemIndex, oneUCharToChar(start),
oneUCharToChar(end));
} else {
itemStr[itemLen] = 0;
u_UCharsToChars(itemStr, buf, itemLen+1);
log_verbose("Ok: %s item %d is \"%s\"\n", pat, itemIndex, buf);
}
++itemIndex;
}
if (*p=='{') {
const char* stringStart = ++p;
int32_t stringLength = 0;
char strCopy[64];
while (*p++ != '}') {
}
stringLength = (int32_t)(p - stringStart - 1);
strncpy(strCopy, stringStart, stringLength);
strCopy[stringLength] = 0;
u_charsToUChars(stringStart, ustr, stringLength);
ustr[stringLength] = 0;
if (itemLen == 0) {
log_err("FAIL: for %s expect \"%s\" next, but got a char\n",
pat, strCopy);
return;
}
if (u_strcmp(ustr, itemStr) != 0) {
log_err("FAIL: for %s expect \"%s\" next\n",
pat, strCopy);
return;
}
}
else {
UChar32 c;
u_charsToUChars(p, ustr, 1);
c = ustr[0];
if (itemLen != 0) {
log_err("FAIL: for %s expect '%c' next, but got a string\n",
pat, *p);
return;
}
if (c != start++) {
log_err("FAIL: for %s expect '%c' next\n",
pat, *p);
return;
}
++p;
}
}
if (uset_size(set) == expectedSize) {
log_verbose("Ok: %s size is %d\n", pat, expectedSize);
} else {
log_err("FAIL: %s size is %d, expected %d\n",
pat, uset_size(set), expectedSize);
}
}
