In article <CLr7GH.1C8@cbnewse.cb.att.com>, grumpy@cbnewse.cb.att.com (Paul J Lucas) writes:
> From article <1994Feb24.165437.27844@us-es.sel.de>, by howland@us-es.sel.de (Gary Howland US/END 60/1/25):
> > Anyhow, the question is, why can't access specifiers be applied to
> > friend declarations?  I can't think of any reason against it.
>
>  That's ok, I can't think of any reason for it.  If you expect
>  some proposal to be taken seriously, then you have to at least
>  explain the semantics of it.
> --
>  - Paul J. Lucas
>    AT&T Bell Laboratories
>    Naperville, IL

It makes sense to me that we could use syntax such as

 class X
 {
  protected friend class Y;
  ...
 };

to allow class Y access to public & protected (but not private) members of
class X.  This would encapsulate the class a little more, which is surely a good thing.  I have heard of developers (and been tempted myself) who inherit from class X before adding the friend declaration in order to achieve the same results as above (although this makes the code less readable).

--
Gary Howland (howland@us-es.sel.de)

Be sure to use the above email address - sometimes the return address is broken.

Many apologies if this is a second posting, but I think my first posting wasn't to world.

Anyhow, the question is, why can't access specifiers be applied to friend declarations?  I can't think of any reason against it.

Gary

--
Gary Howland (howland@us-es.sel.de)

Be sure to use the above email address - more often than not the return address
is knackered - not my fault honest :-).