Re: [boost] [mpl] is there a or_seq like logical metafunction?

Subject: Re: [boost] [mpl] is there a or_seq like logical metafunction?
From: vicente.botet (vicente.botet_at_[hidden])
Date: 2009-02-07 14:29:29

----- Original Message ----- 
From: "Larry Evans" <cppljevans_at_[hidden]>
To: <boost_at_[hidden]>
Sent: Saturday, February 07, 2009 8:24 PM
Subject: Re: [boost] [mpl] is there a or_seq like logical metafunction?

> On 04/27/08 20:50, David Abrahams wrote:
>> on Fri Mar 07 2008, Steven Watanabe <watanabesj-AT-gmail.com> wrote:
>> 
>>> AMDG
>>>
>>> Giovanni Piero Deretta wrote:
>>>> It seems a perfect application for fold:
>>>>
>>>> template<typename Seq>
>>>> struct or_seq
>>>>     : mpl::apply<
>>>>           mpl::fold<mpl::_1, mpl::false_,  mpl::quote2<mpl::or_> >
>>>>         , Seq
>>>>       > {};
>>>>
>>>> template<typename Seq>
>>>> struct and_seq
>>>>     : mpl::apply<
>>>>           mpl::fold<mpl::_1, mpl::true_,  mpl::quote2<mpl::and_> >
>>>>         , Seq
>>>>       > {};
>>>>   
>>> Not quite, because we want it to short circuit.
>>> Not to mention that quote2 will not work on and_ and or_.
>> 
>> Yeah, these should be rewritten in terms of mpl::find.
>> Not sure why mpl::apply was used here either, though maybe I'm missign
>> something.
>> 
> 
> The curren mpl::find uses and_ indirectly (in iter_fold_if.hpp:46).
> Would such a recursive implementation still work?
> 
> Anyway, J.Fletcher is now attempting a variadic template rewrite of
> mpl:;find:
> 
> http://sourceforge.net/mailarchive/message.php?msg_name=F0DEE5AE9F6027488C064CD31A5B26AC07BA6EDC78%40EXCHANGEMB.campus.aston.ac.uk
> 
> which doesn't use and_; however, another alternative which does 
> short-circuit is attached.  It uses eval_if to short-circuit the
> evaluation.  Fletcher's find, OTOH, used if_; however, I think
> that's just because he's not interested in optimization yet.
> 
> A very similar or_ is done by simply exchanging the Zero and One
> parameters.  Interestingly enough, the current and_  uses
> a similar pattern in that it recursively calls a template (and.hpp:23).
> 
> The attached (with the similar or.hpp) passes the logical.cpp test.
 
 

Hi,

Wow, I didn't think that this thread was yet alive.

After suggestion of I don't remember who, I have implemented it as

/* Calculate the mlp::and_ of a sequence of nullary logical metafunctions.*/    
template <typename Seq>
struct and_seq : boost::is_same<typename mpl::find_if<Seq, mpl::not_<mpl::_> >::type, 
           typename mpl::end<Seq>::type> 
     {};

/* Calculate the mlp::or_ of a sequence of nullary logical metafunctions. */    
template <typename Seq>
struct or_seq : mpl::not_<boost::is_same<typename mpl::find_if<Seq, mpl::_>::type, 
           typename mpl::end<Seq>::type> 
    > {};

If there are better ways to do that, you don't tink that this must be encapsulated in the library? Why not add these metafunctions in Boost.MPL?

Thanks,

Vicente