Re: [boost] [MPL] new run-time algorithm: for_all

Subject: Re: [boost] [MPL] new run-time algorithm: for_all
From: Larry Evans (cppljevans_at_[hidden])
Date: 2008-11-23 13:39:24

On 11/23/08 11:43, George van Venrooij wrote:
> Larry Evans wrote:
> 
>> Judging from the "what is it?" on that tinyurl, it is the cross product
>> of the mplbook's exercise 7-8 mentioned in my previous post.
> 
> It indeed has similar behavior when feeding it 2 sequences as the cross 
> product, or to be more precise, an outer product (as you mentioned 
> earlier) since both sequences need not have the same length.
> 
> Looking at Wikipedia's definition: 
> http://en.wikipedia.org/wiki/Outer_product then my algorithm produces 
> the outer product when called with 2 sequences. When called with 3 
> sequences, it generates some kind of 3-way outer product that I am not 
> able to find a good name or definition for.
In the cross.zip file mentioned in my previous post and located here:

http://www.boostpro.com/vault/index.php?PHPSESSID=ab51206c9d980155d142f5bcef8e00ee&direction=0&order=&directory=Template%20Metaprogramming

you can see cross.out, which contains the output from a sort of 3 way
outer product which seems like just an outer product of of a sequence
and a previous outer product.  IOW, a fold of cross over a sequence of 
sequences should result in an n-way outer product.  I haven't tried it
yet; however, it should be very simple to do in cross.hs since haskell
already has a fold function.