ANSCENTER/ANSCORE
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Support UTF8 to UTF16 LE.

Browse Source 
Support Unicode helper
Fix ANSFR to show 2 same faces in 1 image
This commit is contained in:
Tuan Nghia Nguyen
2026-03-31 21:52:47 +11:00
parent 70be68d0fc
commit ccfc5964d4
21 changed files with 379 additions and 104 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
modules/ANSOCR/dllmain.cpp
View File

@@ -427,33 +427,25 @@ extern "C" ANSOCR_API int SetANSOCRALPRFormat(ANSCENTER::ANSOCRBase** Handle, co
// - JSON Unicode escapes (\uXXXX) from ensure_ascii=true output
// - Raw UTF-8 encoded strings
// Pure ASCII input is passed through directly (no conversion overhead).
extern "C" ANSOCR_API int ANSOCR_ConvertUTF8ToUTF16LE(const char* utf8Str, LStrHandle result) {
extern "C" ANSOCR_API int ANSOCR_ConvertUTF8ToUTF16LE(const char* utf8Str, LStrHandle result, int includeBOM) {
try {
if (!utf8Str || !result) return -1;
int len = (int)strlen(utf8Str);
if (len == 0) return 0;
// Check if input contains \uXXXX escapes or non-ASCII bytes
// Always output UTF-16LE (required for LabVIEW "Force Unicode Text" indicators)
const char bom[2] = { '\xFF', '\xFE' };
// Check if input contains \uXXXX escapes
bool hasUnicodeEscapes = false;
bool hasNonAscii = false;
for (int i = 0; i < len; i++) {
if ((unsigned char)utf8Str[i] >= 0x80) hasNonAscii = true;
if (i + 1 < len && utf8Str[i] == '\\' && utf8Str[i + 1] == 'u') hasUnicodeEscapes = true;
for (int i = 0; i + 1 < len; i++) {
if (utf8Str[i] == '\\' && utf8Str[i + 1] == 'u') { hasUnicodeEscapes = true; break; }
}
// Pure ASCII with no escapes — pass through directly
if (!hasNonAscii && !hasUnicodeEscapes) {
MgErr error = DSSetHandleSize(result, sizeof(int32) + len * sizeof(uChar));
if (error != noErr) return -2;
(*result)->cnt = len;
memcpy((*result)->str, utf8Str, len);
return 1;
}
// If contains \uXXXX escapes, decode them to UTF-16LE directly
if (hasUnicodeEscapes) {
std::string utf16le;
utf16le.reserve(len * 2);
if (includeBOM) utf16le.assign(bom, 2);
utf16le.reserve(len * 2 + 2);
for (int i = 0; i < len; ) {
if (i + 5 < len && utf8Str[i] == '\\' && utf8Str[i + 1] == 'u') {
char hex[5] = { utf8Str[i + 2], utf8Str[i + 3], utf8Str[i + 4], utf8Str[i + 5], 0 };
@@ -462,7 +454,6 @@ extern "C" ANSOCR_API int ANSOCR_ConvertUTF8ToUTF16LE(const char* utf8Str, LStrH
utf16le += static_cast<char>((cp >> 8) & 0xFF);
i += 6;
} else {
// ASCII or raw UTF-8 byte — convert as single char
utf16le += utf8Str[i];
utf16le += '\0';
i++;
@@ -476,17 +467,19 @@ extern "C" ANSOCR_API int ANSOCR_ConvertUTF8ToUTF16LE(const char* utf8Str, LStrH
return 1;
}
// Raw UTF-8 — convert via Windows API
#ifdef _WIN32
int wideLen = MultiByteToWideChar(CP_UTF8, 0, utf8Str, len, nullptr, 0);
if (wideLen <= 0) return 0;
std::wstring wideStr(wideLen, 0);
MultiByteToWideChar(CP_UTF8, 0, utf8Str, len, &wideStr[0], wideLen);
int size = wideLen * (int)sizeof(wchar_t);
MgErr error = DSSetHandleSize(result, sizeof(int32) + size * sizeof(uChar));
int dataSize = wideLen * (int)sizeof(wchar_t);
int bomSize = includeBOM ? 2 : 0;
int totalSize = bomSize + dataSize;
MgErr error = DSSetHandleSize(result, sizeof(int32) + totalSize * sizeof(uChar));
if (error != noErr) return -2;
(*result)->cnt = size;
memcpy((*result)->str, wideStr.data(), size);
(*result)->cnt = totalSize;
if (includeBOM) memcpy((*result)->str, bom, 2);
memcpy((*result)->str + bomSize, wideStr.data(), dataSize);
return 1;
#else
return 0;