Regression in AdaControl?

Regression in AdaControl?

[Markus Schöpflin]

Before raising an issue in the AdaControl bug tracker I would like to be sure 
that I'm indeed facing an AdaControl bug here, and not some issue on my side.

Given the follwing code:

   package FOO
   is
      type T;
      type CALLBACK_T is access procedure (P : T);
      type T is access INTEGER;
   end FOO;

and the following AdaControl rules file:

   Name_Clashes: check naming_convention (all, not "example");

When executing this with an old version of GnatPro and AdaControl "ADACTL v. 
1.14r9 with ASIS 2.0.R for GNAT Pro 6.3.2 (20100607)" everything is fine:

---%<---
 > adactl -v -f foo.aru foo
Controlling FOO specification

Total execution time: 0.0s.
--->%---

When executing the same command with a later version of GnatPro and the latest 
version of AdaControl "ADACTL v. 1.16r11 with ASIS 2.0.R for GNAT Pro 7.3.1 
(20150118)", I get the following:

---%<---
 > adactl -v -f foo.aru foo
Controlling FOO specification
============= Phase: Processing =============
AdaCtl version: 1.16r11 with ASIS 2.0.R for GNAT Pro 7.3.1 (20150118)
ASIS error: ASIS.EXCEPTIONS.ASIS_INAPPROPRIATE_ELEMENT
    In rule: NAMING_CONVENTION
   For unit: FOO
Status   : VALUE_ERROR
Diagnosis: Inappropriate Element Kind in 
Asis.Declarations.Corresponding_First_Subtype (AN_INCOMPLETE_TYPE_DECLARATION)
called in Actual procedure for Pre_Operation
with the argument : A_DEFINING_IDENTIFIER
located in FOO (spec, Unit_Id = 2, Context_Id = 1)
text position : foo.ads:4:41
    Nodes:
       Node            : 2043 - N_DEFINING_IDENTIFIER
       R_Node          : 2043 - N_DEFINING_IDENTIFIER
       Node_Field_1    : 0 - N_EMPTY
       Node_Field_2    : 0 - N_EMPTY
    Rel_Sloc           : 58
    obtained from the tree foo.adt (Tree_Id = 1)
called in Asis.Iterator.Traverse_Element
with the argument : AN_ORDINARY_TYPE_DECLARATION
located in FOO (spec, Unit_Id = 2, Context_Id = 1)
text position : foo.ads:4:4
    Nodes:
       Node            : 2051 - N_FULL_TYPE_DECLARATION
       R_Node          : 2051 - N_FULL_TYPE_DECLARATION
       Node_Field_1    : 0 - N_EMPTY
       Node_Field_2    : 0 - N_EMPTY
    Rel_Sloc           : 21
    obtained from the tree foo.adt (Tree_Id = 1)
called in Actual procedure for Pre_Operation
with the argument : A_PACKAGE_DECLARATION
located in FOO (spec, Unit_Id = 2, Context_Id = 1)
text position : foo.ads:1:1
    Nodes:
       Node            : 2062 - N_PACKAGE_DECLARATION
       R_Node          : 2062 - N_PACKAGE_DECLARATION
       Node_Field_1    : 0 - N_EMPTY
       Node_Field_2    : 0 - N_EMPTY
    Rel_Sloc           :-8
    obtained from the tree foo.adt (Tree_Id = 1)
called in Asis.Iterator.Traverse_Element
with the argument : A_PACKAGE_DECLARATION
located in FOO (spec, Unit_Id = 2, Context_Id = 1)
text position : foo.ads:1:1
    Nodes:
       Node            : 2062 - N_PACKAGE_DECLARATION
       R_Node          : 2062 - N_PACKAGE_DECLARATION
       Node_Field_1    : 0 - N_EMPTY
       Node_Field_2    : 0 - N_EMPTY
    Rel_Sloc           :-8
    obtained from the tree foo.adt (Tree_Id = 1)
Total execution time: 0.3s.
--->%---

Can anyone reproduce this with the latest version of AdaControl? I want to be 
sure that the error doesn't come from any problem on my side, e.g. due to 
incorrectly compiling AdaControl with the incorrect ASIS version or some other 
stupidity.

Thanks in advance,
Markus

Re: Regression in AdaControl?

[J-P. Rosen]

Le 15/10/2015 16:23, Markus Schöpflin a écrit :
> Before raising an issue in the AdaControl bug tracker I would like to be
> sure that I'm indeed facing an AdaControl bug here, and not some issue
> on my side.
....
If you have an AdaControl bug box, there IS a problem! Now the problem
may come from AdaControl, from Gnat proper, or from ASIS. That's for me
to determine.

So please, fill in a ticket. That makes it easier for me to follow-up.

PS: crazy^H^H^H^H^H new sophisticated access types have created a number
of problems in the development version of AdaControl. This may be one of
them.

-- 
J-P. Rosen
Adalog
2 rue du Docteur Lombard, 92441 Issy-les-Moulineaux CEDEX
Tel: +33 1 45 29 21 52, Fax: +33 1 45 29 25 00
http://www.adalog.fr

Re: Regression in AdaControl?

[Markus Schöpflin]

Am 15.10.2015 um 18:08 schrieb J-P. Rosen:
> Le 15/10/2015 16:23, Markus Schöpflin a écrit :
>> Before raising an issue in the AdaControl bug tracker I would like to be
>> sure that I'm indeed facing an AdaControl bug here, and not some issue
>> on my side.
> ....
> If you have an AdaControl bug box, there IS a problem! Now the problem
> may come from AdaControl, from Gnat proper, or from ASIS. That's for me
> to determine.
>
> So please, fill in a ticket. That makes it easier for me to follow-up.

OK, done. See http://sourceforge.net/p/adacontrol/tickets/41/.