RE: Boost algorithm library (Was: [boost] New upload: algorithm.zip 1)

From: Fernando Cacciola (fcacciola_at_[hidden])
Date: 2001-07-25 14:58:46

----- Original Message -----
From: Jeremy Siek <jsiek_at_[hidden]>
To: <boost_at_[hidden]>
Sent: Wednesday, July 25, 2001 4:05 PM
Subject: RE: Boost algorithm library (Was: [boost] New upload: algorithm.zip
1)

> On Wed, 25 Jul 2001, Fernando Cacciola wrote:
> > I'm not sure if this should be provided by the library.
> > It is very common for library users to add all sorts of wrappers and
> > shortcuts, so this one in particular might be just something else that a
> > particular user can do upon the library.
> > Perhaps we should make a survey of the overall usage.
> > How many of you does prefer the container interface?
> > Not me.
>
> The wrappers are super easy to create, and do provide a convenience. I
> think enough people have already posted emails to prove the need for such
> things. In addition, its better to have a "standard" set of wrappers
> provided by boost than to have everyone using their own hacks.
>
> Also, providing the wrappers does no harm to those that like the iterator
> interface... just keep using the iterator interface if you like.
>

I am not sure if a *standarized* container-like interface will have a good
impact on a program design.
The range interface is a *very* powerful abstraction, much much better than
the abstraction of an (encapsulated) container.
IMHO, the decision to shortcut from (c.begin(),c.end()) to (c) should be
done consciously, not as the default practice; otherwise, a program might
end up stuck in a design that cannot exploit the flexibility of the range
abstraction.
Of course, this is just my personal view.
I'd like to know what the STL designers have to say about this matters. Do
they monitor this forum?

Fernando Cacciola
Sierra s.r.l.
fcacciola_at_[hidden]
www.gosierra.com