
That's the point. I don't remember exactly what I have done during the process of installing ICU. But one thing clear is that I haven't modified the solution configuration that comes with the original package.
In that case, you are restricted to using the same msvc runtime that ICU uses (assuming you are committed to using regex with ICU - regex and Win32 provide passable Unicode support without it via boost::wregex).
That means I can get unicode support without ICU in the Win32 environment? If that's true, then what is ICU for?
It depends how complete you want the Unicode support, have you seen: http://www.boost.org/doc/libs/1_39_0/libs/regex/doc/html/boost_regex/unicode... ? Basically on Win32 character classification and upper/lower case conversion works OK in wregex, but Unicode specific character classes like [[:lu:]] can't be used, likewise nothing in icu.hpp (so UTF8 support for example). HTH, John.