From: David Abrahams (abrahams_at_[hidden])
Date: 2001-06-07 13:45:31
----- Original Message -----
From: "Daryle Walker" <darylew_at_[hidden]>
> > I am not sure that "void value_type" requirement is a reasonable one,
> > After reading
> > r, it seems to me that it may be a defect as well. Thoughts?
> Shouldn't we keep the decision as defaulted template arguments, so the
> can override the arguments if necessary?
As long as the standard mandates void value_type and difference_type, I
think we have to respect it. The only point in having template arguments for
these would be to allow users to make "iterators" that don't conform to the
standard. Simpler is (much, much) better in this case. IMO.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk