Maybe I am making a fool of myself, but here it goes:

Suppose there was an implemention of (std::)vector that stores its elements
in the following order:

 "physical address":      0   1   2   3   4   5 ...
 "logical position":      0   2   1   3   4   5 ...
 content:                 a   c   b   d   e   f ...
                             ^^^^^^^

(The contents of this vector might have been created by
   v.push_back(a);
   v.push_back(b);
   v.push_back(c);
   v.push_back(d);
   v.push_back(e); etc)

Iterators (which obviously can't be simple pointers) shall work as usual,
i.e. incrementing  begin()  moves from logical position 0 to 1 to 2 ...
etc.

Now &v[0] would even give the address of an contiguous range of memory,
but we have e.g.

  &v[2] != &v[0] + 2

which is not really nice if one wants to use C-style libraries without
doing much copying.

Simple questions:

Is an implementation as outlined above forbidden by the Standard?
If so, where does it say so?  If not, shouldn't it be forbidden?

Andre'

--
Andr    P   nitz ........................................ poenitz@htwm.de

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.research.att.com/~austern/csc/faq.html                ]

Andr=E9 P=F6nitz wrote:
[Re: non-contiguous std::vector<>]
> Is an implementation as outlined above forbidden by the Standard?

See <http://anubis.dkuug.dk/jtc1/sc22/wg21/docs/lwg-defects.html#69>

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.research.att.com/~austern/csc/faq.html                ]