[boost] Re: Review Request: FC++

From: David Abrahams (dave_at_[hidden])
Date: 2003-10-08 14:52:02

"David B. Held" <dheld_at_[hidden]> writes:

> "David B. Held" <dheld_at_[hidden]> wrote in message
> news:bm1nvf$6eb$1_at_sea.gmane.org...
>> [...]
>> And I second the use of MPL conventions.  I would fully
>> expect FC++ to interoperate with MPL where it is easy for
>> it to do so.  Failure to interoperate only because of naming
>> differences would be most unfortunate.
>
> Actually, let me put it this way.  The great thing about MPL is
> not the meta-algorithms or type containers.  The great thing
> about MPL is the MetaConcepts that attempt to do for
> metaprogramming what the STL does for runtime
> programming.  A meta-algorithm that doesn't use ::type and
> ::apply<> is like a generic algorithm that doesn't take iterators
> or functors.  A type container that doesn't define begin<> and
> end<> is like a generic container that doesn't produce iterators.
> The STL is extensible precisely because it relies on Concepts,
> and the same should be true of MPL.

Nicely said.

[There are other great things about MPL, like its ability to reduce
the amount of nasty syntax in and increase understandability of
metaprograms, increased portability, efficiency, and fun... but Dave
hit the most significant nail on its head ;-)]

-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com