container interface for algorithms (Was: Boost algorithm library)

From: Philip Nash (philip.nash_at_[hidden])
Date: 2001-07-25 14:06:44

> > --- In boost_at_y..., "Fernando Cacciola" <fcacciola_at_g...> wrote:
> > >
> > > Do you mean that you just don't want to write (c.begin(),c.end())
> > each time?

[..]

> 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.

Well, I'll add my voice to the list of people who would like an alternative
container (wrapper) interface to the range algorithms. After all, they're
not very involved. Since they'd all be inline they would almost never add
any overhead, but could make the code more readable (I say "could" because I
do recognise that providing such "convenience" wrappers could mean you
become less accustomed to using the iterator range interface, so when you
have to use ranges instead it could become _less_ readable overall!).

But I think some implementations already have non-standard extensions for
container-based interfaces. It would be nice if these could be made portable
by adoption into a widely used and freely available library such as boost.

[)o
IhIL..