/*static*/
void WStringCodec::decode(DBusMessageIter& iter, std::wstring& str)
{
str.clear();
DBusMessageIter _iter;
simppl_dbus_message_iter_recurse(&iter, &_iter, DBUS_TYPE_ARRAY);
int count =
#if DBUS_MAJOR_VERSION == 1 && DBUS_MINOR_VERSION < 9
dbus_message_iter_get_array_len(&_iter) / sizeof(uint32_t);
#else
dbus_message_iter_get_element_count(&iter);
#endif
if (count > 0)
str.reserve(count);
while(dbus_message_iter_get_arg_type(&_iter) != 0)
{
uint32_t t;
Codec<uint32_t>::decode(_iter, t);
str.push_back((wchar_t)t);
}
// advance to next element
dbus_message_iter_next(&iter);
}
void FileReader::ReadRawWString(std::wstring &theWStr, int theNumChars)
{
theWStr.erase();
theWStr.reserve(theNumChars);
for(int i=0; i<theNumChars; i++)
theWStr += (wchar_t)ReadShort();
}
void load_file2(std::wstring& wstrOut, std::wistream& wistreamIn)
{
wstrOut.erase();
//bad() 如果出现错误则返回true
if(wistreamIn.bad())
return;
wstrOut.reserve(wistreamIn.rdbuf()->in_avail());
wchar_t c;
//get() 读取字符
while(wistreamIn.get(c))
{
if(wstrOut.capacity() == wstrOut.size())
wstrOut.reserve(wstrOut.capacity() * 3);
wstrOut.append(1, c);
}
}
explicit node(nodetype type, wchar_t op = '+', int num = 0)
{
t = type;
l_child = r_child = 0;
this->op = op;
this->num = num;
translist.reserve(10);
}
const std::wstring& getAlphabet() {
const wchar_t kMinCharCode = 32, kMaxCharCode = 126;
static std::wstring alphabet;
if (alphabet.empty()) {
alphabet.reserve(kMaxCharCode - kMinCharCode);
for (wchar_t c = kMinCharCode; c <= kMaxCharCode; ++c) {
alphabet.push_back(c);
}
}
return alphabet;
}
///-----------------------------------------------------------------------------
///! @brief TODO enter a description
///! @remark
///-----------------------------------------------------------------------------
void convertToUTF16String(const std::string& str, std::wstring& out)
{
wchar_t buffer[4096];
int numberOfWideChars = MultiByteToWideChar(CP_UTF8, MB_ERR_INVALID_CHARS, str.c_str(), (int)str.size(), buffer, 0);
if (numberOfWideChars > 0)
{
out.reserve(numberOfWideChars);
MultiByteToWideChar(CP_UTF8, MB_ERR_INVALID_CHARS, str.c_str(), (int)str.size(), buffer, numberOfWideChars);
buffer[numberOfWideChars] = 0;
out = buffer;
}
else
{
DWORD error = GetLastError();
HRESULT hr = HRESULT_FROM_WIN32(error);
MSG_TRACE_CHANNEL("String Conversion Error", "Failed to convert from UTF8 to MB with Hresult: 0x%08x, %s", hr, getLastErrorMessage(error));
}
}
bool extractWstring(std::istream & sourceFile, std::wstring & result) {
int stringLengthPrefix = extractInt(sourceFile);
unsigned char * buff = new unsigned char[2*stringLengthPrefix];
char16_t oneWchar;
sourceFile.read((char*)buff, 2*stringLengthPrefix);
result.clear();
result.reserve(stringLengthPrefix+1);
// .CEL files' "Unicode" wide-strings do not actually conform to
// the Unicode standard. Each wide-byte is big-endian instead
// of the required little-endian.
for (int i = 0; i < stringLengthPrefix; i++) {
oneWchar = buff[i*2] << 8;
oneWchar += buff[(i+1)*2-1];
result.append(1, oneWchar);
}
delete [] buff;
return true;
}
/**
* Extracts a JSON String as defined by the spec - "<some chars>"
* Any escaped characters are swapped out for their unescaped values
*
* @access protected
*
* @param wchar_t** data Pointer to a wchar_t* that contains the JSON text
* @param std::wstring& str Reference to a std::wstring to receive the extracted string
*
* @return bool Returns true on success, false on failure
*/
bool JSON::ExtractString(const wchar_t **data, std::wstring &str)
{
str = L"";
while (**data != 0)
{
// Save the char so we can change it if need be
wchar_t next_char = **data;
// Escaping something?
if (next_char == L'\\')
{
// Move over the escape char
(*data)++;
// Deal with the escaped char
switch (**data)
{
case L'"': next_char = L'"'; break;
case L'\\': next_char = L'\\'; break;
case L'/': next_char = L'/'; break;
case L'b': next_char = L'\b'; break;
case L'f': next_char = L'\f'; break;
case L'n': next_char = L'\n'; break;
case L'r': next_char = L'\r'; break;
case L't': next_char = L'\t'; break;
case L'u':
{
// We need 5 chars (4 hex + the 'u') or its not valid
if (!simplejson_wcsnlen(*data, 5))
return false;
// Deal with the chars
next_char = 0;
for (int i = 0; i < 4; i++)
{
// Do it first to move off the 'u' and leave us on the
// final hex digit as we move on by one later on
(*data)++;
next_char <<= 4;
// Parse the hex digit
if (**data >= '0' && **data <= '9')
next_char |= (**data - '0');
else if (**data >= 'A' && **data <= 'F')
next_char |= (10 + (**data - 'A'));
else if (**data >= 'a' && **data <= 'f')
next_char |= (10 + (**data - 'a'));
else
{
// Invalid hex digit = invalid JSON
return false;
}
}
break;
}
// By the spec, only the above cases are allowed
default:
return false;
}
}
// End of the string?
else if (next_char == L'"')
{
(*data)++;
str.reserve(); // Remove unused capacity
return true;
}
// Disallowed char?
else if (next_char < L' ' && next_char != L'\t')
{
// SPEC Violation: Allow tabs due to real world cases
return false;
}
// Add the next char
str += next_char;
// Move on
(*data)++;
}
// If we're here, the string ended incorrectly
return false;
}
BOOL DownloadWebPageUnicode(std::wstring& page, HINTERNET hInternet, LPCWSTR url)
{
TRACEST(_T("DownloadWebPageUnicode"), CW2CT(url));
BOOL bDownloadSuccess = FALSE;
if (hInternet == NULL)
return FALSE;
HINTERNET hFile = InternetOpenUrl(hInternet, url, NULL, 0, INTERNET_FLAG_NO_CACHE_WRITE, 0);
if (hFile)
{
page.clear();
DWORD dwSize = 0;
INT headerCP = 0;//Latin1
INT pageCP = 0;//Latin1
INT workingCP = 1252;//Latin1
if(!HttpQueryInfo(hFile, HTTP_QUERY_RAW_HEADERS_CRLF, 0, &dwSize, 0))
{
if (GetLastError()==ERROR_INSUFFICIENT_BUFFER)
{
SetLastError(0);
LPVOID lpOutBuffer = new CHAR[dwSize];
if (HttpQueryInfo(hFile, HTTP_QUERY_RAW_HEADERS_CRLF, lpOutBuffer, &dwSize, 0))
{
//TRACE((LPCTSTR)lpOutBuffer);
headerCP = GetWindowsCodePageW((LPCTSTR)lpOutBuffer);
}
delete[] lpOutBuffer;
}
}
DWORD pageSize = 0;
DWORD pageSizeLen = sizeof(pageSize);
if (::HttpQueryInfo(hFile, HTTP_QUERY_CONTENT_LENGTH, &pageSize, &pageSizeLen, NULL))
{
if (pageSize > 0)
page.reserve(pageSize + 10);
}
if (page.capacity() < 1000)
page.reserve(50000);
const int bufferSize = 8192;
const int bufferSizeU = 2 * bufferSize;
CHAR bf[bufferSize + 1];
TCHAR bfU[bufferSizeU];
unsigned long nSize = 0;
BOOL bReadSuccess = TRUE;
BOOL bFirstTime = TRUE;
while(bReadSuccess)
{
bReadSuccess = InternetReadFile(hFile, bf, bufferSize, &nSize);
if (bReadSuccess)
{
if (nSize == 0)
{
//TRACE(_T("@3 DownloadWebPage. InternetReadFile Finished\r\n"));
bDownloadSuccess = TRUE;
break;
}
if (bFirstTime)
{
bFirstTime = FALSE;
bf[nSize] = 0;
pageCP = GetWindowsCodePageA(bf);
if (headerCP != pageCP)
{
TRACE(_T("@3 DownloadWebPage. CodePage Differs (header:%d - page:%d)\r\n"),
headerCP, pageCP);
}
if (pageCP > 0) //===Choose the Codepage detected from page if available
workingCP = pageCP;
else if (headerCP > 0) //===Else select the Codepage detected from headers if available
workingCP = headerCP;
//===Otherwise keep the original 1252 (latin 1)
}
if (nSize > 0)
{
INT bfLen = MultiByteToWideChar(workingCP, 0, bf, nSize, 0, 0);
if (bfLen < bufferSizeU)
{
MultiByteToWideChar(workingCP, 0, bf, nSize, bfU, bfLen);
page.append(bfU, bfLen);
}
else
{
TRACE(_T("@1 DownloadWebPageUnicode. Unicode buffer not enough\r\n"));
bReadSuccess = FALSE;
}
}
}
else
HandleInternetError(_T("DownloadWebPageUnicode. InternetReadFile"));
}
InternetCloseHandle(hFile);
}
else
HandleInternetError(_T("DownloadWebPageUnicode. InternetOpenUrl"));
return bDownloadSuccess;
}
void
TextDecoder::Append(std::wstring& str, const uint8_t* bytes, size_t length, CharacterSet charset)
{
switch (charset)
{
case CharacterSet::Unknown:
case CharacterSet::ISO8859_1:
case CharacterSet::ASCII:
{
str.append(bytes, bytes + length);
break;
}
case CharacterSet::ISO8859_2:
case CharacterSet::ISO8859_3:
case CharacterSet::ISO8859_4:
case CharacterSet::ISO8859_5:
case CharacterSet::ISO8859_6:
case CharacterSet::ISO8859_7:
case CharacterSet::ISO8859_8:
case CharacterSet::ISO8859_9:
case CharacterSet::ISO8859_10:
case CharacterSet::ISO8859_11:
case CharacterSet::ISO8859_13:
case CharacterSet::ISO8859_14:
case CharacterSet::ISO8859_15:
case CharacterSet::ISO8859_16:
case CharacterSet::Cp437:
case CharacterSet::Cp1250:
case CharacterSet::Cp1251:
case CharacterSet::Cp1252:
case CharacterSet::Cp1256:
{
str.reserve(str.length() + length);
for (size_t i = 0; i < length; ++i) {
uint8_t c = bytes[i];
if (c < 128)
str.push_back(c);
else
str.push_back(Codecs::SINGLE_BYTE_CODEPAGES[((int)charset - (int)CharacterSet::ISO8859_2) * 128 + c - 128]);
}
break;
}
case CharacterSet::Shift_JIS:
{
std::vector<uint16_t> buf;
JPTextDecoder::AppendShiftJIS(buf, bytes, length);
TextUtfEncoding::AppendUtf16(str, buf.data(), buf.size());
break;
}
case CharacterSet::Big5:
{
std::vector<uint16_t> buf;
Big5TextDecoder::AppendBig5(buf, bytes, length);
TextUtfEncoding::AppendUtf16(str, buf.data(), buf.size());
break;
}
case CharacterSet::GB2312:
{
std::vector<uint16_t> buf;
GBTextDecoder::AppendGB2312(buf, bytes, length);
TextUtfEncoding::AppendUtf16(str, buf.data(), buf.size());
break;
}
case CharacterSet::GB18030:
{
std::vector<uint16_t> buf;
GBTextDecoder::AppendGB18030(buf, bytes, length);
TextUtfEncoding::AppendUtf16(str, buf.data(), buf.size());
break;
}
case CharacterSet::EUC_JP:
{
std::vector<uint16_t> buf;
JPTextDecoder::AppendEUCJP(buf, bytes, length);
TextUtfEncoding::AppendUtf16(str, buf.data(), buf.size());
break;
}
case CharacterSet::EUC_KR:
{
std::vector<uint16_t> buf;
KRTextDecoder::AppendEucKr(buf, bytes, length);
TextUtfEncoding::AppendUtf16(str, buf.data(), buf.size());
break;
break;
}
case CharacterSet::UnicodeBig:
{
str.reserve(str.length() + length / 2);
for (size_t i = 0; i + 1 < length; i += 2) {
str.push_back((static_cast<wchar_t>(bytes[i]) << 8) + bytes[i + 1]);
}
break;
}
case CharacterSet::UTF8:
{
TextUtfEncoding::AppendUtf8(str, bytes, length);
break;
}
default:
break;
}
}