Boost logo

Ublas :

From: Michael Stevens (mail_at_[hidden])
Date: 2005-06-29 17:13:13

On Wednesday 29 June 2005 23:26, Gunter Winkler wrote:
> Hallo,
> currently it is not possible to call index1() or index2() of a matrix
> iterator if the iterator is at the end. But is this really necessary?
> Why shouldn't I be able to access iterator1.index2() if this iterator is
> at the end of the column? The column index is independent of the
> current position. Most (all?) iterators store this index anyway.

Interesting! The other dimensions index must always be valid else this
iterator instance could not have been created.

        iterator1.index2() and iterator2.index1() have no preconditions
        iterator1.index1() and iterator2.index2() have the existing 'must be
dereferenceable' precondition

Make sense to me!