|
Boost : |
From: David Abrahams (david.abrahams_at_[hidden])
Date: 2001-08-24 09:58:17
----- Original Message -----
From: "Peter Dimov" <pdimov_at_[hidden]>
> My first reaction is that the compile-time list support should probably be
> factored out into a submodule. Many of the tuple_* operations are actually
> list operations; a compile-time list library would be useful for template
> metaprogramming, although I don't know how this will be affected by the
> upcoming boost::mpl.
I've been using Aleksey's MPL stuff. The more I use it, the more amazing it
looks.
> A possible alternative is to define a tuple namespace and a nested class
> template named 'type'; a tuple will then be referred to as
>
> tuple::type<X, Y> t;
That is fiendishly clever!
> > Tuple has a typedef 'inherited' which gives the underlying cons list
> > type, e.g.:
> > tuple<A, B, C>::inherited equals cons<A, cons<B, cons<C,null_type> > >
>
> Have you considered tuple<...>::list_type?
Actually, I've been hoping that tuple<...>::type would give me the cons
list, making it a uniform type generator which can be used in metaprograms.
-Dave
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk