[scicatur]

Do you know what would be the best/most portable solution to handle UTF-8 characters in C++ code.

WinAPI seems to define a bunch of types like : WCHAR and TCHAR ... but I don't want to use anything MS specific.

wchar_t seems to be standard ... but its 16 bit unicode while UTF-8 is variable length encoding where one character takes space 1 - 6 bytes.

Also it would be nice to have something like the STL <string> class to use but with UTF-8.

Please if you know the solution or have ideas I would be most pleased to hear about it.

[Advertisements]

haven't used this but it may be somewhat 'portable':
http://www.utilityco...tr/default.aspx

unicode is probably more portable(no guarentee in C++, however)

sales pitch: consider using java on your next project if being portable is a requirement

[bdlt]

haven't used this but it may be somewhat 'portable':
http://www.utilityco...tr/default.aspx

unicode is probably more portable(no guarentee in C++, however)

sales pitch: consider using java on your next project if being portable is a requirement

[scicatur]

Thanks for answer. It looks good except it is commercial.

I think I tought up kind of a work-around solution to the problem.
Because UTF-8 its fundamentally just bytes so I read from file to
a 'unsigned char' array after which I use custom function to convert into
STL <string>. That custom function between ansi and utf-8 I maybe write myself BUT it is never perfect because utf-8 is fully unicode while ansi is not fully. Fortunately the utf-8 characters that cannot be converted are used only in very exotic languages. Converting japanese to ansi shouldn't be problem. Then of course I must make ansi to utf-8 function as well.