libcmo21/LibCmo/VTEncoding.cpp

258 lines
6.7 KiB
C++
Raw Normal View History

2023-02-11 15:29:51 +08:00
#include "VTEncoding.hpp"
2023-02-26 21:48:03 +08:00
namespace LibCmo::EncodingHelper {
2023-02-11 15:29:51 +08:00
#pragma region assist functions
#if defined(LIBCMO_OS_WIN32)
#define LIBCMO_STR_EQUAL(a, b) strcmp(reinterpret_cast<const char*>(a), reinterpret_cast<const char*>(b)) == 0
2023-02-26 21:48:03 +08:00
bool GetWindowsCodePage(const char* u8_encoding_spec, UINT* result) {
if (LIBCMO_STR_EQUAL(u8_encoding_spec, u8"CP_ACP")) *result = CP_ACP;
else if (LIBCMO_STR_EQUAL(u8_encoding_spec, u8"CP_MACCP")) *result = CP_MACCP;
else if (LIBCMO_STR_EQUAL(u8_encoding_spec, u8"CP_OEMCP")) *result = CP_OEMCP;
else if (LIBCMO_STR_EQUAL(u8_encoding_spec, u8"CP_THREAD_ACPP")) *result = CP_THREAD_ACP;
else if (LIBCMO_STR_EQUAL(u8_encoding_spec, u8"CP_UTF8")) *result = CP_UTF8;
else {
char* pend = nullptr;
errno = 0;
uint64_t v = std::strtoull(u8_encoding_spec, &pend, 10);
if (pend == u8_encoding_spec || errno == ERANGE) return false;
*result = static_cast<UINT>(v);
2023-02-11 15:29:51 +08:00
}
2023-02-26 21:48:03 +08:00
return true;
}
2023-02-11 15:29:51 +08:00
#undef LIBCMO_STR_EQUAL
2023-02-26 21:48:03 +08:00
bool WcharToChar(const wchar_t* src, std::string& dest, UINT codepage) {
int count, write_result;
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
//converter to CHAR
2023-03-03 11:06:26 +08:00
count = WideCharToMultiByte(codepage, 0, src, -1, NULL, 0, NULL, NULL);
2023-02-26 21:48:03 +08:00
if (count <= 0) return false;
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
dest.resize(count);
2023-03-03 11:06:26 +08:00
write_result = WideCharToMultiByte(codepage, 0, src, -1, dest.data(), count, NULL, NULL);
2023-02-26 21:48:03 +08:00
if (write_result <= 0) return false;
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
return true;
}
bool WcharToChar(std::wstring& src, std::string& dest, UINT codepage) {
return WcharToChar(src.c_str(), dest, codepage);
}
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
bool CharToWchar(const char* src, std::wstring& dest, UINT codepage) {
int wcount, write_result;
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
// convert to WCHAR
wcount = MultiByteToWideChar(codepage, 0, src, -1, NULL, 0);
if (wcount <= 0) return false;
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
dest.resize(wcount);
2023-03-03 11:06:26 +08:00
write_result = MultiByteToWideChar(codepage, 0, src, -1, dest.data(), wcount);
2023-02-26 21:48:03 +08:00
if (write_result <= 0) return false;
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
return true;
}
bool CharToWchar(std::string& src, std::wstring& dest, UINT codepage) {
return CharToWchar(src.c_str(), dest, codepage);
}
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
bool CharToChar(const char* src, std::string& dest, UINT src_codepage, UINT dest_codepage) {
std::wstring intermediary;
if (!CharToWchar(src, intermediary, src_codepage)) return false;
if (!WcharToChar(intermediary, dest, dest_codepage)) return false;
return true;
}
bool CharToChar(std::string& src, std::string& dest, UINT src_codepage, UINT dest_codepage) {
return CharToChar(src.c_str(), dest, src_codepage, dest_codepage);
}
2023-02-25 22:58:28 +08:00
2023-02-11 15:29:51 +08:00
#else
2023-03-03 11:06:26 +08:00
static constexpr const size_t IconvInc = 16;
bool DoIconv(const char* enc_from, const char* enc_to, std::string& str_from, std::string& str_to) {
2023-03-01 15:51:56 +08:00
iconv_t cd;
char *inbuf = nullptr, *outbuf = nullptr;
size_t inbytesleft, outbytesleft, nchars, result_len;
// check empty
if (str_from.empty()) {
str_to.clear();
return true;
}
// create iconv descriptor
cd = iconv_open(enc_to, enc_from);
if (cd == (iconv_t) -1) {
// fail to create iconv descriptor
return false;
}
// pre-resize
str_to.resize(str_from.size() + IconvInc);
// setup some variables
inbytesleft = str_from.size();
inbuf = str_from.data();
outbytesleft = str_to_size();
outbuf = str_to.data();
result_len = str_to.size();
// conv core
nchars = iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
while (nchars == (size_t)-1 && errno == E2BIG) {
// record the length has been converted
size_t len = outbuf - str_to.data();
// resize for variables
result_len += IconvInc;
outbytesleft += IconvInc;
// resize for container
str_to.resize(result_len);
// assign new outbuf from failed position
outbuf = str_to.data() + len;
nchars = iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
}
// close iconv descriptor
iconv_close(cd);
// check error
if (nchars == (size_t)-1) {
// failed
return false;
} else {
// success
// resize result to get correct data
str_to.resize(result_len - outbytesleft);
return true;
}
}
2023-02-11 15:29:51 +08:00
#endif
#pragma endregion
#pragma region core functions
#if defined(LIBCMO_OS_WIN32)
2023-02-26 21:48:03 +08:00
ENCODING_TOKEN CreateEncodingToken(std::string& token_string) {
ENCODING_TOKEN token = new(std::nothrow) UINT();
if (token == nullptr) return ENCODING_TOKEN_DEFAULT;
2023-02-25 22:58:28 +08:00
2023-02-26 21:48:03 +08:00
if (!GetWindowsCodePage(token_string.c_str(), token)) {
*token = CP_ACP;
2023-02-25 22:58:28 +08:00
}
2023-02-26 21:48:03 +08:00
return token;
}
void DestroyEncodingToken(ENCODING_TOKEN token) {
if (token != ENCODING_TOKEN_DEFAULT) {
delete token;
2023-02-25 22:58:28 +08:00
}
2023-02-26 21:48:03 +08:00
}
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
void GetUtf8VirtoolsName(std::string& native_name, std::string& u8_name, ENCODING_TOKEN token) {
if (token == ENCODING_TOKEN_DEFAULT) {
u8_name = native_name.c_str();
return;
2023-02-25 22:58:28 +08:00
}
2023-02-26 21:48:03 +08:00
// convert with fallback
if (!CharToChar(native_name, u8_name, *token, CP_UTF8)) {
u8_name = native_name.c_str();
}
}
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
void GetNativeVirtoolsName(std::string& u8_name, std::string& native_name, ENCODING_TOKEN token) {
if (token == ENCODING_TOKEN_DEFAULT) {
native_name = u8_name.c_str();
return;
2023-02-25 22:58:28 +08:00
}
2023-02-26 21:48:03 +08:00
// convert with fallback
if (!CharToChar(u8_name, native_name, CP_UTF8, *token)) {
native_name = u8_name.c_str();
2023-02-11 15:29:51 +08:00
}
2023-02-26 21:48:03 +08:00
}
2023-02-11 15:29:51 +08:00
2023-02-26 21:48:03 +08:00
void SetStdPathFromU8Path(std::filesystem::path& stdpath, const char* u8_path) {
std::wstring intermediary;
if (CharToWchar(u8_path, intermediary, CP_UTF8)) {
stdpath = intermediary.c_str();
} else {
// fallback
stdpath = u8_path;
2023-02-12 18:08:29 +08:00
}
2023-02-26 21:48:03 +08:00
}
FILE* OpenStdPathFile(std::filesystem::path& u8_filepath, bool is_read) {
return _wfopen(u8_filepath.wstring().c_str(), is_read ? L"rb" : L"wb");
}
2023-02-12 18:08:29 +08:00
2023-02-11 15:29:51 +08:00
#else
2023-02-25 22:58:28 +08:00
2023-03-03 11:06:26 +08:00
static constexpr const char UTF8_SYMBOL[] = "UTF-8";
2023-03-01 15:51:56 +08:00
ENCODING_TOKEN CreateEncodingToken(std::string& token_string) {
ENCODING_TOKEN token = new(std::nothrow) char[token_string.size() + 1];
if (token == nullptr) return ENCODING_TOKEN_DFAULT;
std::memcpy(token, token_string.c_str(), token_string.size() + 1);
return token;
}
void DestroyEncodingToken(ENCODING_TOKEN token) {
if (token != ENCODING_TOKEN_DEFAULT) {
delete[] token;
}
}
void GetUtf8VirtoolsName(std::string& native_name, std::string& u8_name, ENCODING_TOKEN token) {
if (token == ENCODING_TOKEN_DEFAULT) {
u8_name = native_name.c_str();
return;
}
// convert with fallback
if (!DoIconv(token, UTF8_SYMBOL, native_name, u8_name)) {
u8_name = native_name.c_str();
}
}
void GetNativeVirtoolsName(std::string& u8_name, std::string& native_name, ENCODING_TOKEN token) {
if (token == ENCODING_TOKEN_DEFAULT) {
native_name = u8_name.c_str();
return;
}
// convert with fallback
if (!DoIconv(UTF8_SYMBOL, token, u8_name, native_name)) {
native_name = u8_name.c_str();
}
}
void SetStdPathFromU8Path(std::filesystem::path& stdpath, const char* u8_path) {
stdpath = u8_path;
}
FILE* OpenStdPathFile(std::filesystem::path& u8_filepath, bool is_read) {
return fopen(u8_filepath.string().c_str(), is_read ? "rb" : "wb");
}
2023-02-25 22:58:28 +08:00
2023-02-11 15:29:51 +08:00
#endif
#pragma endregion
}