Re: [boost] AlRangeExandrescu?

Subject: Re: [boost] AlRangeExandrescu?
From: David Abrahams (dave_at_[hidden])
Date: 2009-07-27 07:12:51

on Sat Jul 25 2009, Thorsten Ottosen <thorsten.ottosen-AT-dezide.com> wrote:

> David Abrahams skrev:
>> on Fri Jul 24 2009, Thorsten Ottosen <thorsten.ottosen-AT-dezide.com> wrote:
>>
>>> David Abrahams skrev:
>>>
>>>> Regardless, there could still be a single correct answer from the
>>>> standpoint of aesthetics, purity, and orthogonality.  If
>>>>
>>>>     drop_front(find(range, x))
>>>>
>>>> does the same thing as
>>>>
>>>>     find[_f+1,_e]( range, x )
>>>>
>>>> then you don't need the complicated meta-goop in there.
>
> So how do we evaluate that aesthetics, purity and orthogonality
> of the two different approaches? 

Lots of hard thinking.

> The find interface is one interface with goop, and the other
> alternative is now at least 5 different functions: find, until,
> before, drop_front, drop_back.

No, that's not a fair assessment.  drop_front and drop_back are useful
on their own and in fact I think Andrei is already calling them
pop_front and pop_back(?)

So we're left with "find, until, and before."  What's "before?"

>>> Then how would you safely implement
>>>
>>>    find[_f-1,_e]( range, x );
>>
>> At first glance, you can't.  Is there a known use case for that?  If not, it's moot.
>
> Who knows. Maybe you want the element before the found one to be there for some
> reason.

"Maybe, for some reason" arguments don't hold much water with me,
especially in a library design context.

-- 
Dave Abrahams
BoostPro Computing
http://www.boostpro.com