owen nirvana wrote:
> if compiling with _UNICODE, all string must be std:wstring & const wchar*
Really? How do the custom* examples work under VC++, then? They're
built with _UNICODE defined.
I don't believe there is any requirement that you must only use wide
character strings a VC++ Unicode build. You can still use narrow
strings to handle the UTF-8 strings returned by MySQL. Then you
translate those to the UCS-2 character set used by Windows NT-derived
OSes as shown in the util module from the examples. This is clearly
explained in chapter 6 in the user manual.