Stroustrup's Appendix to the ARM,
ANSI/ISO Resolutions, r14.9 says:

"Duplicate explicit instantiation is technically an error"

1) Is this still valid?

2) If class A<T> uses class Other<T> in its implementation,
   is class Other<T> automatically instantiated when
   class A<T> is instantiated explicitely?

   a) If yes:
 What happens when a different class B<T> that
 also needs Other<T> is explicitely instantiated?

 Class Other<T> is then duplicately instantiated;
 is this an error?

   b) if no:
 Do I have to know all the implementation details
 of class A<T> to be able to explicitely
 instantiate explicitely all the needed templates?

By the way, is there a good reference how to instantiation
mechanism (automatically and explicit) is standardized?

Newsfeed unreliable, please answer also via E-mail

Helmut Zeisel  <helmut.zeisel@aon.at>
---
[ 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         ]
[ FAQ:      http://reality.sgi.com/employees/austern_mti/std-c++/faq.html    ]
[ Policy:   http://reality.sgi.com/employees/austern_mti/std-c++/policy.html ]
[ Comments? mailto:std-c++-request@ncar.ucar.edu                             ]

917312523 wrote:
> 2) If class A<T> uses class Other<T> in its implementation,
>    is class Other<T> automatically instantiated when
>    class A<T> is instantiated explicitely?

Yes, but the instantiation of Other<T> would be classified as an
implicit instantiation (even if Other<T> is a base class).

>  What happens when a different class B<T> that
>  also needs Other<T> is explicitely instantiated?

No problem, Other<T> is implicitly instantiated in both cases.

> By the way, is there a good reference how to instantiation
> mechanism (automatic and explicit) is standardized?

There is no standardization of the *mechanism*.  If you read the (draft)
standard carefully, you can get a feel for what plausible implementation
mechanisms people had in mind when the standard was cast.

As far as I know, there is no easily-digested source of information
about the mechanics of template instantiation.  The "SIGS C++ Report"
has frequent articles about STL and templates, which help.  Other then
that, get the latest (dec96) draft.

I have some ideas for a plausible implementation that I've been working
on -- drop me a line if you're interested in the ugly details.

john lilley

      [ Send an empty e-mail to c++-help@netlab.cs.rpi.edu for info ]
      [ about comp.lang.c++.moderated. First time posters: do this! ]
---
[ 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         ]
[ FAQ:      http://reality.sgi.com/employees/austern_mti/std-c++/faq.html    ]
[ Policy:   http://reality.sgi.com/employees/austern_mti/std-c++/policy.html ]
[ Comments? mailto:std-c++-request@ncar.ucar.edu                             ]