U_CAPI int32_t U_EXPORT2
uspoof_areConfusableUnicodeString(const USpoofChecker *sc,
const icu::UnicodeString &s1,
const icu::UnicodeString &s2,
UErrorCode *status) {
const UChar *u1 = s1.getBuffer();
int32_t length1 = s1.length();
const UChar *u2 = s2.getBuffer();
int32_t length2 = s2.length();
int32_t results = uspoof_areConfusable(sc, u1, length1, u2, length2, status);
return results;
}
void FStringConverter::ConvertString(const icu::UnicodeString& Source, const int32 SourceStartIndex, const int32 SourceLen, FString& Destination)
{
if (Source.length() > 0)
{
UErrorCode ICUStatus = U_ZERO_ERROR;
ucnv_reset(ICUConverter);
// Get the internal buffer of the string, we're going to use it as scratch space
TArray<TCHAR>& InternalStringBuffer = Destination.GetCharArray();
// Work out the maximum size required and resize the buffer so it can hold enough data
const int32_t DestinationCapacityBytes = UCNV_GET_MAX_BYTES_FOR_STRING(SourceLen, ucnv_getMaxCharSize(ICUConverter));
const int32 DestinationCapacityTCHARs = DestinationCapacityBytes / sizeof(TCHAR);
InternalStringBuffer.SetNumUninitialized(DestinationCapacityTCHARs);
// Perform the conversion into the string buffer, and then null terminate the FString and size it back down to the correct size
const int32_t DestinationSizeBytes = ucnv_fromUChars(ICUConverter, reinterpret_cast<char*>(InternalStringBuffer.GetData()), DestinationCapacityBytes, Source.getBuffer() + SourceStartIndex, SourceLen, &ICUStatus);
const int32 DestinationSizeTCHARs = DestinationSizeBytes / sizeof(TCHAR);
InternalStringBuffer[DestinationSizeTCHARs] = 0;
InternalStringBuffer.SetNum(DestinationSizeTCHARs + 1, /*bAllowShrinking*/false); // the array size includes null
check(U_SUCCESS(ICUStatus));
}
else
{
Destination.Empty();
}
}
U_CAPI int32_t U_EXPORT2
uspoof_checkUnicodeString(const USpoofChecker *sc,
const icu::UnicodeString &text,
int32_t *position,
UErrorCode *status) {
int32_t result = uspoof_check(sc, text.getBuffer(), text.length(), position, status);
return result;
}
int32 GetUnicodeStringLengthImpl(const TCHAR* Source, const int32 InSourceStartIndex, const int32 InSourceLength)
{
if (InSourceLength > 0)
{
const icu::UnicodeString TmpStr = ConvertString(Source, InSourceStartIndex, InSourceLength);
return TmpStr.length();
}
return 0;
}
/**
* Write an Unitex file content (to system filesystem or filespace)
* it write from two buffer (prefix and suffix). This is useful for writing both header and footer (or BOM and text...)
*/
UNITEX_FUNC int UNITEX_CALL WriteUnicodeUnitexFile(const char*filename, icu::UnicodeString const& uString)
{
UChar uBom = 0xfeff;
const UChar * uBuffer = uString.getBuffer();
int32_t uLength = uString.length();
bool result = WriteUnitexFile(filename, &uBom, sizeof(UChar), uBuffer, uLength * sizeof(UChar)) == 0;
return result;
}
static void
printLine(UChar32 start, UChar32 end, Status status, const icu::UnicodeString &mapping) {
if(start==end) {
printf("%04lX ", (long)start);
} else {
printf("%04lX..%04lX ", (long)start, (long)end);
}
printf("; %s", statusNames[status]);
if(status==MAPPED || status==DEVIATION || !mapping.isEmpty()) {
printf(" ;");
const UChar *buffer=mapping.getBuffer();
int32_t length=mapping.length();
int32_t i=0;
UChar32 c;
while(i<length) {
U16_NEXT(buffer, i, length, c);
printf(" %04lX", (long)c);
}
}
puts("");
}
// --------------------------------------------------------------------------
icu::UnicodeString
processor::normalize_enum (
icu::UnicodeString const& value
)
// --------------------------------------------------------------------------
{
icu::UnicodeString normalized;
int32_t pos;
bool space_before = false;
bool leading = true;
for (pos=0; pos<value.length(); ++pos)
{
if (value[pos] == ' ')
{
space_before = true;
}
else
{
if (space_before)
{
if (leading)
{
leading = false;
}
else
{
normalized += ' ';
}
space_before = false;
}
normalized += value[pos];
}
}
return normalized;
}
void alignedNormalizeUnicodeString(icu::UnicodeString const& u, IcuNormalizer2Ptr normalizer,
ITakeAlignedChars& out) {
// TODO: test
Position start = 0;
int32 len = u.length(), pos;
UErrorCode err = U_ZERO_ERROR;
int nfcPrefixLen = normalizer->spanQuickCheckYes(u, err);
assert(U_SUCCESS(err));
assert(len >= 0 && nfcPrefixLen >= 0);
TokenSpan span;
span.first = 0;
icu::StringCharacterIterator it(u);
while ((pos = it.getIndex()) < nfcPrefixLen) {
assert(it.hasNext());
Unicode c = it.next32PostInc();
span.second = span.first + 1;
out.takeWithSpan(c, span);
++span.first;
}
icu::UnicodeString remainder(u.tempSubString(nfcPrefixLen)), normalized;
CharsFromUnicodeStringImpl chars(remainder); // TODO: docs say normalizeSecondAndAppend
IcuNormalizeByChunks<CharsFromUnicodeStringImpl> norm(chars, normalizer);
norm.takeAllWithSpan(out);
}
jobject operator()(icu::UnicodeString const& value) const {
return env->NewString(value.getBuffer(), value.length());
}
inline
cxxopts::UnicodeStringIterator
end(const icu::UnicodeString& s)
{
return cxxopts::UnicodeStringIterator(&s, s.length());
}
void FStringConverter::ConvertString(const icu::UnicodeString& Source, FString& Destination)
{
return ConvertString(Source, 0, Source.length(), Destination);
}
int32 GetNativeStringLength(const icu::UnicodeString& Source)
{
return GetNativeStringLength(Source, 0, Source.length());
}
U_CAPI int32_t U_EXPORT2
uspoof_checkUnicodeString(const USpoofChecker *sc,
const icu::UnicodeString &id,
int32_t *position,
UErrorCode *status) {
const SpoofImpl *This = SpoofImpl::validateThis(sc, *status);
if (This == NULL) {
return 0;
}
int32_t result = 0;
IdentifierInfo *identifierInfo = NULL;
if ((This->fChecks) & (USPOOF_RESTRICTION_LEVEL | USPOOF_MIXED_NUMBERS)) {
identifierInfo = This->getIdentifierInfo(*status);
if (U_FAILURE(*status)) {
goto cleanupAndReturn;
}
identifierInfo->setIdentifier(id, *status);
identifierInfo->setIdentifierProfile(*This->fAllowedCharsSet);
}
if ((This->fChecks) & USPOOF_RESTRICTION_LEVEL) {
URestrictionLevel idRestrictionLevel = identifierInfo->getRestrictionLevel(*status);
if (idRestrictionLevel > This->fRestrictionLevel) {
result |= USPOOF_RESTRICTION_LEVEL;
}
if (This->fChecks & USPOOF_AUX_INFO) {
result |= idRestrictionLevel;
}
}
if ((This->fChecks) & USPOOF_MIXED_NUMBERS) {
const UnicodeSet *numerics = identifierInfo->getNumerics();
if (numerics->size() > 1) {
result |= USPOOF_MIXED_NUMBERS;
}
// TODO: ICU4J returns the UnicodeSet of the numerics found in the identifier.
// We have no easy way to do the same in C.
// if (checkResult != null) {
// checkResult.numerics = numerics;
// }
}
if (This->fChecks & (USPOOF_CHAR_LIMIT)) {
int32_t i;
UChar32 c;
int32_t length = id.length();
for (i=0; i<length ;) {
c = id.char32At(i);
i += U16_LENGTH(c);
if (!This->fAllowedCharsSet->contains(c)) {
result |= USPOOF_CHAR_LIMIT;
break;
}
}
}
if (This->fChecks &
(USPOOF_WHOLE_SCRIPT_CONFUSABLE | USPOOF_MIXED_SCRIPT_CONFUSABLE | USPOOF_INVISIBLE)) {
// These are the checks that need to be done on NFD input
UnicodeString nfdText;
gNfdNormalizer->normalize(id, nfdText, *status);
int32_t nfdLength = nfdText.length();
if (This->fChecks & USPOOF_INVISIBLE) {
// scan for more than one occurence of the same non-spacing mark
// in a sequence of non-spacing marks.
int32_t i;
UChar32 c;
UChar32 firstNonspacingMark = 0;
UBool haveMultipleMarks = FALSE;
UnicodeSet marksSeenSoFar; // Set of combining marks in a single combining sequence.
for (i=0; i<nfdLength ;) {
c = nfdText.char32At(i);
i += U16_LENGTH(c);
if (u_charType(c) != U_NON_SPACING_MARK) {
firstNonspacingMark = 0;
if (haveMultipleMarks) {
marksSeenSoFar.clear();
haveMultipleMarks = FALSE;
}
continue;
}
if (firstNonspacingMark == 0) {
firstNonspacingMark = c;
continue;
}
if (!haveMultipleMarks) {
marksSeenSoFar.add(firstNonspacingMark);
haveMultipleMarks = TRUE;
}
if (marksSeenSoFar.contains(c)) {
// report the error, and stop scanning.
// No need to find more than the first failure.
result |= USPOOF_INVISIBLE;
break;
}
marksSeenSoFar.add(c);
}
}
if (This->fChecks & (USPOOF_WHOLE_SCRIPT_CONFUSABLE | USPOOF_MIXED_SCRIPT_CONFUSABLE)) {
// The basic test is the same for both whole and mixed script confusables.
// Compute the set of scripts that every input character has a confusable in.
// For this computation an input character is always considered to be
// confusable with itself in its own script.
//
// If the number of such scripts is two or more, and the input consisted of
// characters all from a single script, we have a whole script confusable.
// (The two scripts will be the original script and the one that is confusable)
//
// If the number of such scripts >= one, and the original input contained characters from
// more than one script, we have a mixed script confusable. (We can transform
// some of the characters, and end up with a visually similar string all in
// one script.)
if (identifierInfo == NULL) {
identifierInfo = This->getIdentifierInfo(*status);
if (U_FAILURE(*status)) {
goto cleanupAndReturn;
}
identifierInfo->setIdentifier(id, *status);
}
int32_t scriptCount = identifierInfo->getScriptCount();
ScriptSet scripts;
This->wholeScriptCheck(nfdText, &scripts, *status);
int32_t confusableScriptCount = scripts.countMembers();
//printf("confusableScriptCount = %d\n", confusableScriptCount);
if ((This->fChecks & USPOOF_WHOLE_SCRIPT_CONFUSABLE) &&
confusableScriptCount >= 2 &&
scriptCount == 1) {
result |= USPOOF_WHOLE_SCRIPT_CONFUSABLE;
}
if ((This->fChecks & USPOOF_MIXED_SCRIPT_CONFUSABLE) &&
confusableScriptCount >= 1 &&
scriptCount > 1) {
result |= USPOOF_MIXED_SCRIPT_CONFUSABLE;
}
}
}
cleanupAndReturn:
This->releaseIdentifierInfo(identifierInfo);
if (position != NULL) {
*position = 0;
}
return result;
}
QString
EnabledLocalesModel::unicodeStringToQString( const icu::UnicodeString& sourceStr )
{
return QString( reinterpret_cast<const QChar*>( sourceStr.getBuffer() ),
sourceStr.length() );
}