On Thu, Nov 30, 2000 at 01:33:41PM -0200, Alexandre Oliva wrote:
> On Nov 30, 2000, Falk Hueffner <falk.hueffner@student.uni-tuebingen.de> wrote:
>
> >> IIRC, vectors have to be contiguous, even though the Standard doesn't
> >> explicitly state that.
>
> > How do you mean that? If the standard doesn't explicitely state it, they
> > don't have to, I would think.
>
> As you pointed out, this was decided/concluded after the Standard was
> published. IIRC, the requirements for vector can't be met with a
> non-contiguous implementation.
More than that. Alex Stepanov's original design used contiguous storage for
vectors, and that design decision was so "obvious" that the ISO committee
never discussed it until after the standard was frozen. Then finally
somebody pointed out the assumption, and the committee decided to codify
it with the next TC.
So yes, it's a safe assumption, and will soon be "valid code."
Phil
-- pedwards at disaster dot jaj dot com | pme at sources dot redhat dot com devphil at several other less interesting addresses in various dot domains The gods do not protect fools. Fools are protected by more capable fools.--------------------- Object Oriented Numerics List -------------------------- * To subscribe/unsubscribe: use the handy web form at http://oonumerics.org/oon/ * If this doesn't work, please send a note to owner-oon-list@oonumerics.org
This archive was generated by hypermail 2b29 : Wed Feb 20 2002 - 03:20:14 EST