Bool
UnicodeSanityCheck(const void *buffer, // IN
ssize_t lengthInBytes, // IN
StringEncoding encoding) // IN
{
ASSERT(Unicode_IsEncodingValid(encoding));
/*
* Sanity check US-ASCII here, so we can fast-path its conversion
* to Unicode later.
*/
if (encoding == STRING_ENCODING_US_ASCII) {
const uint8 *asciiBytes = (const uint8 *) buffer;
if (lengthInBytes == -1) {
for (; *asciiBytes != '\0'; asciiBytes++) {
if (*asciiBytes >= 0x80) {
return FALSE;
}
}
} else {
ssize_t i;
for (i = 0; i < lengthInBytes; i++) {
if (asciiBytes[i] >= 0x80) {
return FALSE;
}
}
}
}
return TRUE;
}
Unicode
UnicodeAllocInternal(const void *buffer, // IN
ssize_t lengthInBytes, // IN
StringEncoding encoding, // IN
Bool strict) // IN
{
char *utf8Result = NULL;
ASSERT(buffer != NULL);
ASSERT(lengthInBytes >= 0);
ASSERT(Unicode_IsEncodingValid(encoding));
if (!strict) {
CodeSet_GenericToGeneric(Unicode_EncodingEnumToName(encoding),
buffer, lengthInBytes,
"UTF-8", CSGTG_TRANSLIT, &utf8Result, NULL);
return utf8Result;
}
switch (encoding) {
case STRING_ENCODING_US_ASCII:
case STRING_ENCODING_UTF8:
if (Unicode_IsBufferValid(buffer, lengthInBytes, encoding)) {
utf8Result = Util_SafeStrndup(buffer, lengthInBytes);
}
break;
case STRING_ENCODING_UTF16_LE:
// utf8Result will be left NULL on failure.
CodeSet_Utf16leToUtf8((const char *)buffer,
lengthInBytes,
&utf8Result,
NULL);
break;
default:
CodeSet_GenericToGeneric(Unicode_EncodingEnumToName(encoding),
buffer, lengthInBytes,
"UTF-8", 0, &utf8Result, NULL);
break;
}
return (Unicode)utf8Result;
}
