entries and access parameters

entries and access parameters

[Alex R. Mosteo]

Hello, 

I'm trying to find in the 95 rationale the explanation for this point

http://www.adahome.com/rm95/rm9x-09-05-02.html

(13) An entry_declaration in a task declaration shall not contain a
specification for an access parameter (See 3.10)

3.10 is simply the section on access types and no mention of entries is
made. 

I'm curious about that prohibition, mainly to know if workarounding it with
a type like

type Obj_Access is access all ...

is legal or will give further problems. Gnat compiles these without problem.

I've failed to find anything in the rationale (though I have remembrances of
having read something about this in the past relating to accessibility
checks being the problem). I've found, however, this paragraph in a doc
about Gnat Runtime:

"The entry declaration must be placed inside the task specification. (...)
Access parameters are not permitted, though parameters of any access type
are, of course, allowed."
http://www.iuma.ulpgc.es/users/jmiranda/gnat-rts/node21.htm

So I guess the above workaround is legit.

Any comments on why the restriction are welcome! BTW, if the workaround is
legit, I've found a bug in gnatgpl06 that I'll post in a separate message.

Re: entries and access parameters

[Stuart]

"Alex R. Mosteo" <devnull@mailinator.com> wrote in message 
news:4hn8k3FdjobU1@individual.net...
> I'm trying to find in the 95 rationale the explanation for this point
>
> http://www.adahome.com/rm95/rm9x-09-05-02.html
>
> (13) An entry_declaration in a task declaration shall not contain a
> specification for an access parameter (See 3.10)
> I'm curious about that prohibition, mainly to know if workarounding it 
> with
> a type like

I can't help with the reference to 3.10, but the annotated ARM gives the 
following after 9.5.2 (13)

<quote>
Reason: Access parameters for task entries would require a complex 
implementation. For example:

task T is
   entry E(Z : access Integer); -- Illegal!
end T;

task body T is
begin
   declare
      type A is access all Integer;
      X : A;
      Int : aliased Integer;
      task Inner;
      task body Inner is
      begin
         T.E(Int'Access);
      end Inner;
   begin
      accept E(Z : access Integer) do
         X := A(Z); -- Accessibility_Check
      end E;
   end;
end T;

Implementing the Accessibility_Check inside the accept_statement for E is 
difficult, since one does not know whether the entry caller is calling from 
inside the immediately enclosing declare block or from outside it. This 
means that the lexical nesting level associated with the designated object 
is not sufficient to determine whether the Accessibility_Check should pass 
or fail.

Note that such problems do not arise with protected entries, because 
entry_bodies are always nested immediately within the protected_body; they 
cannot be further nested as can accept_statements, nor can they be called 
from within the protected_body (since no entry calls are permitted inside a 
protected_body).
</quote>
-- 
Stuart 

Re: entries and access parameters

[Alex R. Mosteo]

Stuart wrote:

> "Alex R. Mosteo" <devnull@mailinator.com> wrote in message
> news:4hn8k3FdjobU1@individual.net...
>> I'm trying to find in the 95 rationale the explanation for this point
>>
>> http://www.adahome.com/rm95/rm9x-09-05-02.html
>>
>> (13) An entry_declaration in a task declaration shall not contain a
>> specification for an access parameter (See 3.10)
>> I'm curious about that prohibition, mainly to know if workarounding it
>> with
>> a type like
> 
> I can't help with the reference to 3.10, but the annotated ARM gives the
> following after 9.5.2 (13)

Ahhh thanks, so it was there where I read about it. Now it's coming back.

(removed the relevant part of the aarm).

Re: entries and access parameters

[Jeffrey R. Carter]

Alex R. Mosteo wrote:
> 
> I'm curious about that prohibition, mainly to know if workarounding it with
> a type like
> 
> type Obj_Access is access all ...
> 
> is legal or will give further problems. Gnat compiles these without problem.

That is legal.

-- 
Jeff Carter
"When Roman engineers built a bridge, they had to stand under it
while the first legion marched across. If programmers today
worked under similar ground rules, they might well find
themselves getting much more interested in Ada!"
Robert Dewar
62