From: Brian McNamara (lorgon_at_[hidden])
Date: 2003-11-20 01:35:41
On Wed, Nov 19, 2003 at 10:05:30PM -0500, David Abrahams wrote:
> Brian McNamara <lorgon_at_[hidden]> writes:
> > The main advantage to "intervals" over "iterators" is that I am not
> > constantly harassed into having to carry two separate objects around all
> > the time: instead I just carry one. It is tiring to deal with
> > interfaces which appear to have been designed with no attention to the
> > common case.
> I'd have thought you of all people would be irked at the inability to
> do functional composition with the results of algorithms.
Well, there is that too. But I think to really do that well, you may
have to dump the whole standard library and start again from scratch.
Imperative interfaces just don't mix very well with functional ones.
But I confess that I haven't thought deeply about topics like these
anytime in the last few years. Maybe adding appropriate layers like
"ranges" does make it possible to meaningfully do functional composition
of imperative algorithms. (Show me the money!)
-- -Brian McNamara (lorgon_at_[hidden])
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk