Elaboration order [was: cross linking packages]

[jgv@swl.msd.ray.com (John Volan)]

dewar@cs.nyu.edu (Robert Dewar) writes:

>Matt a brief answer to your question is that order of elaboration of
>types is critical, and in particular if package a with's package b,
                                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>it is essential that the spec of b be elaborated before the spec of a.
                           
Did you mean to say:

(1) "if the *spec* of package A with's package B, then the spec of B
must be elaborated before the spec of A"

or did you mean:

(2) "if package A (anywhere, either in its spec or its body) with's
package B, then the spec of B must be elaborated before the spec of A"

Without hunting through the RM9X for a relevant chapter and verse, it
seems to me that Ada only requires (1) to be enforced.  If (2) were
the case, then it would be impossible to have "mutually-coupled"
packages, where the bodies of A and B both "with" each other's specs.
But clearly this is allowed.  (Moreover, it seems to work very nicely
in GNAT.)  Of course, the spec of A must be elaborated before the body
of B, and the spec of B must be elaborated before the body of A.  But
does Ada require any ordering of the elaborations of the two bodies?
Or perhaps I should ask, under what circumstances would Ada require
a particular order of elaboration between two package bodies?

					-- John Volan

--------------------------------------------------------------------------------
--  Me : Person := (Name                => "John Volan",
--                  Company             => "Raytheon Missile Systems Division",
--                  E_Mail_Address      => "jgv@swl.msd.ray.com",
--                  Affiliation         => "Enthusiastic member of Team Ada!",
--                  Humorous_Disclaimer => "These opinions are undefined " &
--                                         "by my employer and therefore " &
--                                         "any use of them would be "     &
--                                         "totally erroneous.");
--------------------------------------------------------------------------------