Re: Generic-Package Elaboration Question / Possible GNAT Bug.

[Robert A Duff <bobduff@shell01.TheWorld.com>]

"Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> writes:

> On Sun, 20 Nov 2011 23:25:39 -0800 (PST), AdaMagica wrote:
>
>> On 20 Nov., 10:55, "Dmitry A. Kazakov" <mail...@dmitry-kazakov.de>
>> wrote:
>>>> On Nov 19, 4:12�pm, Robert A Duff <bobd...@shell01.TheWorld.com>
>>>> wrote:
>>>>> In general, the design of Ada is that if a generic body is legal,
>>>>> then every possible instance is legal (so the compiler need not
>>>>> check the instances). �The design of C++ templates is very different
>>>>> in this regard!
>>>
>>> I don't know if Ada keeps that promise for generic bodies, ...

I think it does.  I'd be interested in any counterexamples.
(Implementation defined cases don't count.)

>...but it certainly
>>> does not for the specifications.

Right.

>...Generic contracts are too weak for that.

Well, it's not that the contracts are too weak.  You just have to
consider the entire generic spec (including the private part,
unfortunately), as part of the "contract".

Also, the generic contract model applies only to compile-time rules.
The fact that the instance body is guaranteed to be legal doesn't
mean that it will work properly!  And there are a few cases where we
deliberately "cheat" in the RM, by making something raise an exception
in a generic, when it would normally be a legality rule.

>> Note that Bob said this holds true for *bodies*, here you have a
>> *spec* that might produce an illegal instance.
>
> If the specification becomes illegal so does the body. 

OK, but my comment was the other way around:  "if a generic body is
legal, then ..."

- Bob