Re: [Boost-users] MPL Map of types to integral values

From: Cromwell Enage (sponage_at_[hidden])
Date: 2007-11-17 14:27:16

--- Robert Dailey wrote:
> That's basically what I did in the previous post,
> the only differences are:
> 
> 1) You're using boosts version of integral (which
> I didn't know about), however I could make my
> integral struct a wrapper for integral_c (since
> integral_c specifies one additional template
> parameter that will always be the same in this
> case).

I'm guessing that you're looking for a map_c that kind
of wraps around a map almost the same way a vector_c
is wrapped around a vector.  Sorry, no such beast
exists yet.  You'll have to roll your own, either just
the way you did or with my suggestions.

> 2) You're specifying "using namespace" inside of
> namespace scope, which extends out into global
> namespace.  'using' does not respect namespace
> scope.  This conflicts with a lot of objects I
> have defined outside of the namespace.

We all know that "using namespace" declarations
outside class scope are a bad idea, especially in
header files.  (Not sure if they work in class scope
but outside function scope, haven't tried that yet.) 
They were placed there to keep the e-mail concise and
readable--eyes start to glaze after reading too many
boost::mpl:: prefixes and so on.

> Thanks for your help.

                              No problem.
                              Cromwell D. Enage

      ____________________________________________________________________________________
Never miss a thing.  Make Yahoo your home page. 
http://www.yahoo.com/r/hs