Re: Remote Types packages in distributed systems

[stt@houdini.camb.inmet.com (Tucker Taft)]

Michael F Brenner (mfb@mbunix.mitre.org) wrote:
: There are two contradictory statements in Appendix E of the Ada
: Language Reference Manual pertaining to remote-types. Which one
: takes precedence?

: E.2(9) Remote Types. The declaration of the library unit can depend
: only on other remote types library units, or [a shared passive or a pure
: unit]; the body of the library unit is unrestricted; ...


: E.2.2(4) A remote types library unit is a library unit to which
:          the pragma Remote_Types applies. The following restrictions
:          apply to the declaration of such a library unit:
:                  it shall be preelaborable; ...



: It cannot have a preelaborable declaration if its body is unrestricted
: and can import a non-preelaborable unit.

Saying that the *spec* (aka "declaration") of a Remote_Types unit shall be 
preelaborable says nothing about the preelaborability of its body
(aka "completion").

: Which of these two contradictory statements shall rule?

I don't see the contradiction.

: Mike Brenner               

--
-Tucker Taft   stt@inmet.com   http://www.inmet.com/~stt/
Intermetrics, Inc.  Burlington, MA  USA