Re: [Fresco-devel] gcc 3.2 fun

Nathaniel Smith wrote:

> So, err, why do you need to define char_traits for unsigned char and
> the like?  Shouldn't gcc already have char_traits<unsigned char>?  Why
> do you need to redefine them?

Just in case its relevant, I remember reading somewhere that for the
purposes of specialisation, 'char' != 'unsigned char' != 'signed char'
!= 'char'. So perhaps gcc has char_traits<char> provided, but not the
modified versions?

-- 
Neil