On Wed, Oct 7, 2009 at 11:58 AM, Igor R wrote:

Hello,

I try to accomplish the subj with help of boost's utf8_codecvt_facet. I based my code on this example: http://www.boost.org/doc/libs/1_40_0/libs/serialization/doc/codecvt.html . The only difference is that my utf8 text resides in std::string:

#include <sstream> #include <iostream> #include "boost/archive/detail/utf8_codecvt_facet.hpp" // link with boost/libs/serialization/src/utf8_codecvt_facet.cpp

int main() {  std::string utf;  utf.resize(11);  // hardcode some utf8 text  utf[0] = 0xd7;  utf[1] = 0x90;  utf[2] = 0xd7;  utf[3] = 0x99;  utf[4] = 0xd7;  utf[5] = 0x92;  utf[6] = 0xd7;  utf[7] = 0x95;  utf[8] = 0xd7;  utf[9] = 0xa8;  utf[10] = 0x0;  std::locale old_locale;  std::locale utf8_locale(old_locale, new boost::archive::detail::utf8_codecvt_facet());  std::locale::global(utf8_locale);  std::stringstream in;  in.imbue(utf8_locale);  in.str(utf);  std::wstringstream out;  out << in;  std::wcout << out.str() << std::endl; }

The above code doesn't work: "out" buffer doesn't contain correct unicode interpretation of the string. Actually, all i want is a c++ equivalent to the following WinAPI:

#include "windows.h" int main() {  std::string utf;  utf.resize(11);  // hardcode some utf8 text  utf[0] = 0xd7;  utf[1] = 0x90;  utf[2] = 0xd7;  utf[3] = 0x99;  utf[4] = 0xd7;  utf[5] = 0x92;  utf[6] = 0xd7;  utf[7] = 0x95;  utf[8] = 0xd7;  utf[9] = 0xa8;  utf[10] = 0x0;  wchar_t outBuff[11];  MultiByteToWideChar(CP_UTF8, 0, utf.c_str(), -1, outBuff, 10); }

...which works well.

Any idea would be greatly appreciated!

Er... I thought UTF8 *is* a form of Unicode? Looks like you are trying to convert UTF8 to UTF16, for what reason?

boost-users-bounces@lists.boost.org wrote:

Hello,

I try to accomplish the subj with help of boost's utf8_codecvt_facet.

[snip]

std::locale utf8_locale(old_locale, new boost::archive::detail::utf8_codecvt_facet()); std::locale::global(utf8_locale); std::stringstream in; in.imbue(utf8_locale); in.str(utf); std::wstringstream out; out << in; std::wcout << out.str() << std::endl; }

The above code doesn't work: "out" buffer doesn't contain correct unicode interpretation of the string. Actually, all i want is a c++ equivalent to the following WinAPI:

I *think* that this invokes operator<<(const void*) on "out", passing the results of operator void*() on "in". What may have had a chance at producing something near the desired behavior would be to invoke operator<<(wostream&, const char*), like: out << in.str().c_str() Sadly, this won't work either as this uses ctype::widen, not codecvt::do_in to make the conversion. AFAIK, the only use of codedcvt in the iostreams library is for the conversion between the "external" and "internal" encodings done by basic_filebuf. To get the desired behavior, I think you'll have to write your own operator<<(wostream&, const string&) and use the codecvt facet inside it. IIRC, C++0x includes things like wstring_convert to do that.