win32ada with Object Ada

win32ada with Object Ada

[Saif-Ali Butt]

I have got Object Ada7 (Beta relaease) and Gnat 3.09 and both of the
compilers are failed to compile win32ada bindings.  Object Ada does not
recognize the project files available in the bindings.  However I opened a
new project and included some files in the project.  Both of the compilers
have problems with the code in winnt.ads.  

If the code in winnt.ads really illegal, I ask, "How  could it be compiled
on Gnat 3.02 (as is mentioned in the readme file)?" Normally all the
validated compilers should be able to compile a code or not to compile a
code.  If some compilers can compile it and other not, then one cannot
talk about compatibility of compilers.

With best wishes

Saif Ali Butt 

Re: win32ada with Object Ada

[Tom Moran]

> If some compilers can compile it and other not, then one cannot
> talk about compatibility of compilers.
We're talking about compiler bugs here, and the set of bugs in one
compiler is unlikely to match the set in another (unless of course they
are both the empty set ;).

Re: win32ada with Object Ada

[Martin C. Carlisle]

>I have got Object Ada7 (Beta relaease) and Gnat 3.09 and both of the
>compilers are failed to compile win32ada bindings.  Object Ada does not
>recognize the project files available in the bindings.  However I opened a
>new project and included some files in the project.  Both of the compilers
>have problems with the code in winnt.ads.  

Object Ada should already have the win32ada binding built in (This is
certainly true for v. 7.0.269).  You merely need to add 
c:\objectada\win32ada\binding\lib to the search path and some
subset of the following to Link/Pass to Linker (depending on your app):

c:\objectada\apilib\comdlg32.lib
c:\objectada\apilib\advapi32.lib
c:\objectada\apilib\comctl32.lib

How to get win32ada to compile with gnat 3.09 has been asked enough
times it should probably make this newsgroup's FAQ.  Move offensive
pragmas from where they appear to the procedure spec they modify.  Also,
use the ObjectAda file for win32-winmain rather than the gnat one.

>If the code in winnt.ads really illegal, I ask, "How  could it be compiled
>on Gnat 3.02 (as is mentioned in the readme file)?" Normally all the
>validated compilers should be able to compile a code or not to compile a
>code.  If some compilers can compile it and other not, then one cannot
>talk about compatibility of compilers.

Actually two compilers that are fully compliant with the LRM may compile
the same code such that only one gives an error.  For example, Object Ada
specifies System.Address as an access type, thus making NULL a legal
value.  Gnat does not, thus requiring the use of System.Null_Address
rather than NULL.

--Martin

-- 
Martin C. Carlisle, Computer Science, US Air Force Academy
mcc@cs.usafa.af.mil, http://www.usafa.af.mil/dfcs/bios/carlisle.html
DISCLAIMER:  This content in no way reflects the opinions, standard or 
policy of the US Air Force Academy or the United States Government.

Re: win32ada with Object Ada

[Ed Falis]

There were some early problems compiling the binding, which I beleive were
related to the presence of 64 bit integer types.

These problems have been corrected in one of the compiler or the bindings,
and the current version works fine with the binding.  Since OA 7.0 was
released in September, two patches have been made available on our web site
(http://www.aonix.com/).

OA 7.1 for Windows is due to start shipping within a week or so.

Also note that the bindings are already precompiled in the product, and you
only need to add them to the project search list using the project.settings.
search tab.

- Ed

Ed Falis
Aonix
(617) 221-7341

Argue for your limitations ... and they're yours
- R. Bach
 Saif-Ali Butt wrote in article ...
>
>I have got Object Ada7 (Beta relaease) and Gnat 3.09 and both of the
>compilers are failed to compile win32ada bindings.  Object Ada does not
>recognize the project files available in the bindings.  However I opened
a
>new project and included some files in the project.  Both of the
compilers
>have problems with the code in winnt.ads. 
>
>If the code in winnt.ads really illegal, I ask, "How  could it be
compiled
>on Gnat 3.02 (as is mentioned in the readme file)?" Normally all the
>validated compilers should be able to compile a code or not to compile a
>code.  If some compilers can compile it and other not, then one cannot
>talk about compatibility of compilers.
>
>With best wishes
>
>Saif Ali Butt
>
> 

Re: win32ada with Object Ada

[Robert Dewar]

Saif-Ali asks 

<<If the code in winnt.ads really illegal, I ask, "How  could it be compiled
on Gnat 3.02 (as is mentioned in the readme file)?" Normally all the
validated compilers should be able to compile a code or not to compile a
code.  If some compilers can compile it and other not, then one cannot
talk about compatibility of compilers.
>>

First, the version of GNAT mentioned, 3.02, was not validated, that is a very
old version of GNAT.  

Second, validation does not ensure 100% conformance (no testing can ever
ensure 100% conformance).

It is indeed the validated version of GNAT (4.00) and versions immediately
preceding it (3.09) that correctly reject the illegalities.

But always remember, it is the RM that defines what is legal, NOT any
compiler, and NOT the ACVC tests. Programmer's should not rely entirely
on a compiler to tell them what is legal and what is not. For one thing,
they may accidentally write non-portable code if they do this (it is
perfectly possible to write code that is legal on one compiler and illegal
on another, e.g. somethingt like Integer(System.Max_Int).

Re: win32ada with Object Ada

[Pascal Obry]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2419 bytes --]




I have made modifications to the Win32Ada binding for GNAT 3.09 and
GNAT 3.10 (a prerealease not yet publicly released). I could distribute a
patch if you are interested.

The only question is "have I the right to distribute a patch ?" Looking at
the copyright it seem that I can't, what do you think :

----------------------------------------------------------------------------
---
--
-- THIS FILE AND ANY ASSOCIATED DOCUMENTATION IS PROVIDED WITHOUT CHARGE
-- "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED,
INCLUDING
-- BUT NOT LIMITED TO THE IMPLIED WARRANTIES OF MERCHANTABILITY AND/OR
-- FITNESS FOR A PARTICULAR PURPOSE.  The user assumes the entire risk as
to
-- the accuracy and the use of this file.  This file may be used, copied,
-- modified and distributed only by licensees of Microsoft Corporation's
-- WIN32 Software Development Kit in accordance with the terms of the 
-- licensee's End-User License Agreement for Microsoft Software for the
-- WIN32 Development Kit.
--
-- Copyright (c) Intermetrics, Inc. 1995
-- Portions (c) 1985-1994 Microsoft Corporation with permission.
-- Microsoft is a registered trademark and Windows and Windows NT are
-- trademarks of Microsoft Corporation.
--
----------------------------------------------------------------------------
---

It say that it may be used only by licensees of Microsoft Corporation's
WIN32 Software Development Kit. With GNAT 3.09 it is possible to use
the binding without the Microsoft librairies... so it seem that this
copyright
forbid this kind of use and that you must buy the SKD to use it !

Is my understanding right ?

Pascal.
--

--|------------------------------------------------------------
--| Pascal Obry                               Team-Ada Member |
--|                                                           |
--| EDF-DER-IPN-SID- Ing�nierie des Syst�mes d'Informations   |
--|                                                           |
--| Bureau G1-010           e-mail: pascal.obry@der.edfgdf.fr |
--| 1 Av G�n�ral de Gaulle  voice : +33-1-47.65.50.91         |
--| 92141 Clamart CEDEX     fax   : +33-1-47.65.50.07         |
--| FRANCE                                                    |
--|------------------------------------------------------------
--|
--|   http://ourworld.compuserve.com/homepages/pascal_obry
--|
--|   "The best way to travel is by means of imagination"

 

Re: win32ada with Object Ada

[Robert Dewar]

pascal asks

<<I have made modifications to the Win32Ada binding for GNAT 3.09 and
GNAT 3.10 (a prerealease not yet publicly released). I could distribute a
patch if you are interested.
 
The only question is "have I the right to distribute a patch ?" Looking at
the copyright it seem that I can't, what do you think :>>


Just to be clear, you can distribute the patch, you absolutely cannot
distribute the patched version. Any copying of the patched or unpatched
versoins of the Win32 bindings not explicitly permitted by Microsoft is
of course a violation of their copyright. For example, if you download
this binding and do NOT have licensed access to the SDK, you are violating
the copyright. This is true even though there is no technical requirement
for you to need the SDK.

Similarly anyone mirroring an FTP site where these bindings resides is
also violating the Microsoft copyright.

It is these copyright problems incidentally that caused trouble in ARA
endorsing these bindings for general use.

Re: win32ada with Object Ada

[Robert Dewar]

Pascal asks

<<I have made modifications to the Win32Ada binding for GNAT 3.09 and
GNAT 3.10 (a prerealease not yet publicly released). I could distribute a
patch if you are interested.
 
The only question is "have I the right to distribute a patch ?" Looking at
the copyright it seem that I can't, what do you think :>>

You can certainly distribute a patch if you are a licensed user of the
original product. Of course anyone using your patch must be a properly
licensed user of the bindings (and have a proper SDK license), but there
is no restriction on the patch. In fact you can copyright the patch itself
however you like.

Re: win32ada with Object Ada

[Robert Dewar]

<<Actually two compilers that are fully compliant with the LRM may compile
the same code such that only one gives an error.  For example, Object Ada
specifies System.Address as an access type, thus making NULL a legal
value.  Gnat does not, thus requiring the use of System.Null_Address
rather than NULL.>>


generally this is true, but this is not the issue in the illegalities that
were discussed here. Those were true illegalities, relating to freezing issues,
and any correct Ada 95 compiler must reject those errors. I assume that the
Object Ada version has simply fixed these errors.

Re: win32ada with Object Ada

[Pascal Obry]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 922 bytes --]



Robert,

Thanks for the clarification.


I will distribute a patch for GNAT 3.09 / GNAT 3.10 (not yet publicly
released) soon.

Pascal.

--

--|------------------------------------------------------------
--| Pascal Obry                               Team-Ada Member |
--|                                                           |
--| EDF-DER-IPN-SID- Ing�nierie des Syst�mes d'Informations   |
--|                                                           |
--| Bureau G1-010           e-mail: pascal.obry@der.edfgdf.fr |
--| 1 Av G�n�ral de Gaulle  voice : +33-1-47.65.50.91         |
--| 92141 Clamart CEDEX     fax   : +33-1-47.65.50.07         |
--| FRANCE                                                    |
--|------------------------------------------------------------
--|
--|   http://ourworld.compuserve.com/homepages/pascal_obry
--|
--|   "The best way to travel is by means of imagination"

Re: win32ada with Object Ada

[Geert Bosch]

Pascal Obry (pascal.obry@der.edfgdf.fr) wrote:


   I have made modifications to the Win32Ada binding for GNAT 3.09 [...]

   The only question is "have I the right to distribute a patch ?" Looking at
   the copyright it seem that I can't, what do you think :

When you have licensed the Win32 SDK, you can do so when there
are no other restrictions in the End-User License Agreement.

In practise it comes down to this: although you technically do
not need the SDK, you are forced to buy it to get access to
(part of) the Win32 API information. 

Regards,
   Geert
    

Re: win32ada with Object Ada

[Ginny Wood]

(someone) wrote:
> 
> >I have got Object Ada7 (Beta relaease) and Gnat 3.09 and both of the
> >compilers are failed to compile win32ada bindings.  Object Ada does not
> >recognize the project files available in the bindings.  However I opened a
> >new project and included some files in the project.  Both of the compilers
> >have problems with the code in winnt.ads.

The ObjectAda *beta*?!?  This is at least 1.5 years old.  Why are you
even bothering to look at it?

By the way, your right to look at it expired one year ago.  If you 
want to play with the real ObjectAda for free, you should pick up
a "Special Edition" CD from Aonix.

Re: win32ada with Object Ada

[Heinz-Joachim Staerke]

Saif-Ali Butt wrote:
> 
> I have got Object Ada7 (Beta relaease) and Gnat 3.09 and both of the
> compilers are failed to compile win32ada bindings.  Object Ada does not
> recognize the project files available in the bindings.  However I opened a
> new project and included some files in the project.  Both of the compilers
> have problems with the code in winnt.ads.
> 
> If the code in winnt.ads really illegal, I ask, "How  could it be compiled
> on Gnat 3.02 (as is mentioned in the readme file)?" Normally all the
> validated compilers should be able to compile a code or not to compile a
> code.  If some compilers can compile it and other not, then one cannot
> talk about compatibility of compilers.
> 
> With best wishes
> 
> Saif Ali Butt
none

Re: win32ada with Object Ada

[Robert Dewar]

H-J S asks

<<> If the code in winnt.ads really illegal, I ask, "How  could it be compiled
> on Gnat 3.02 (as is mentioned in the readme file)?" Normally all the
> validated compilers should be able to compile a code or not to compile a
> code.  If some compilers can compile it and other not, then one cannot
> talk about compatibility of compilers.>>

Yes, the code is really illegal.

Yes, GNAT 3.02, an unvalidated Ada 95 compiler that became obsoleted by
subsequent versions well over a year ago did compile it.

No, validation does not guarantee that all compiler compile exactly the
same code. Validation is a form of testing. Testing does not guarantee
absence of bugs.

Yes, normally reasonably conformant and mature compilers, one useful, but
not totally adequate indication of which is validation, will have this
problem less often.

Now that Object Ada and GNAT are both validated, they disagree less often,
but they do not promise to agree 100%