I'm developing a windows application using regex++3.0 to solve regulare expression, but >i don't know how to add unicode support to it, anyone can give me some advice about it, >thanks a lot.

Depends what you mean by unicode support: on windows if you use boost::wregex then you'll get a certain amount of Unicode support (character classification works for example, but surrogate pairs do not). Alternatively in Boost-1.33 there is optional support for ICU (http://www-306.ibm.com/software/globalization/icu/index.jsp), which together with Boost.Regex provides level 1 Unicode regex support (so surrogate pairs do the right thing, and character classifications like [[:Lu:]] work). HTH, John.