|
Boost : |
From: David A. Greene (greened_at_[hidden])
Date: 2001-11-26 23:51:20
David A. Greene wrote:
> The question I always ask when facing these sort of issues is: what
> is the purpose of the thing I'm defining? The name should reflect
> that. Now, I'm just starting out learning about template
> metaprogramming, but from Andrei's book it looks like a common
> use is to enable convenient overloading/specialization. Perhaps
> the name should reflect that. Of course other uses should be
> considered as well.
After continuing in the thread and seeing the other uses of
int_t<N>, it strikes me that it's quite overloaded. One
of these is unique_type<N>. Another is int_value<N>.
type2type<T> can maybe be called type_wrapper<T> or even
degenerate_typedef<T> in some cases. :)
Does it at all make sense to provide "naming" subclasses so
that programmers can name the thing based on its use context?
I'm thinking something along the lines of:
template<int N>
class int_value : public int_t<N> {};
etc. Maybe it isn't practical -- just a thought I had.
-Dave
-- "Some little people have music in them, but Fats, he was all music, and you know how big he was." -- James P. Johnson
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk