[GAP] Binding Problem with GNAT

[Riehle, Richard (CIV)]

John,

That did the trick.   The adjustment of the pragmas
make it work just fine.

I hope others who are using JEWL will find this
a useful tip.

------------------------------------------------------------

Sorry I cannot be at the conference.  Ada is a minor part
of my duties these days.   I am barely able to keep it in
any part of the curriculum.   

NPS students are discouraged from doing anything with Ada.
>From the perspective of our faculty, Ada is now deprecated 
for most DoD projects.   Some think it is forbidden.  Others
think it is a tired, old, and obsolete language.   

We have one course where Ada is included, if I teach that 
course.  If someone else teaches it, Ada is likely to be
mentioned in passing, but given very little support.

It would be great if I could get a funded research project
that required Ada, but those are no longer available from
DoD sources either -- with the folding of AJPO.  In fact, if
I suggest Ada as the language for a research project, I am
likely to be laughed out of the room.   

This is really a sad state of affairs.  When I first came to 
NPS, Ada was required for the Navy students.  Now it is not even
an option.  I attended a lecture by a high-level Navy Admiral a
few years ago, delivered to NPS students and faculty, in which
he went on at some length lamenting the experience with Ada and
expressing his delight that it was no longer a problem with the
abandonment of the mandate. This did not help with my efforts to
reinvigorate interest in it.

I wish there were some way I could get an on-going list of active
Ada projects in the DoD and elsewhere.  I have one short list, but
I know that some projects on that list, while licensing GNAT, are
also giving preference to other languages.   

So far, no one has written an article for widespread publication
announcing the advent of Ada 2005.   I have not even seen such an
article in Crosstalk.  There is never any mention of Ada in any 
of the DoD publications.  Not every Federal Computer Week has had
an announcement, as far as I know.

It is all very well to have a conference of people who have a 
continued interest in Ada.  However, I would hope that those who
are at that conference would be able to collaborate on some strategy
for bringing Ada back into visibility in the many forums and 
publications that decision-makers use.  

One of the most important weapon systems projects (which I am not
allowed to mention) that ought to be using Ada is included in the
research of some of our faculty members.  They have been explicitly
told that the project will use real-time Java, not Ada.  This is
truly a stupid decision, but one that typifies the attitudes that
seem to prevail in today's DoD decision-making environment.  

If there is any topic that ought to be a focus of attention at the
SigAda conference, it is how we resurrect interest in Ada within
the DoD and among DoD contractors.

Sorry for the rant.  Living in the desert (the Ada desert) has 
probably affected my mind.  I know I am bit deranged from having
to live with my darling Ada for so long.  :-)

Richard

    Why would anyone choose a language that is inherently error-prone
    such as C++ and expect a result that is error-free.

===================================================================

-----Original Message-----
From: John McCormick [mailto:mccormick at cs.uni.edu] 
Sent: Wednesday, November 07, 2007 4:00 PM
To: GNAT Academic Program discussion list; Riehle, Richard (CIV)
Subject: Re: [GAP] Binding Problem with GNAT

Richard,

You need to change pragma linker in one of the files.  The directions 
I give to my students are at http://cns2.uni.edu/~mccormic/188/

p.s.  I'm having a great time at the SIGAda conference.  Great 
student paper by a West Point student.  We need some more from the NPS.

John



>Subject:  Binding GNAT using JEWL and AdaGide
>
>To:       Anyone using GNAT
>
>From:     Richard Riehle
>
>Date:     7 November 2007
>
>
>Using AdaGide v 7.420003, I am trying to compile and Link some of the
>John English JEWL programs.
>
>There was a time, with earlier versions of GNAT, that these would
>compile
>and link just fine with AdaGide.
>
>The error message, on calling gnatlink.exe is:
>
>    c:\gnat\2007\libexec\gcc\pentium-mingw32msv\4.1.3\ld.exe:
>                cannot find -luser32   -lgdi32   -lcomdlg32   -lwinnmm
>
>    collect 2:  ID returned 1 exit status
>    gnatlink: error when calling c:\gnat\2007\bin\gcc.exe
>    gnatmake: *** link failed
>    Done: error detected
>
>We use AdaGide in our curriculum because it is so much easier for an
>academic environment than GPL, so we'd like to continue with it.
>Also, JEWL has proven to be one of the most valuable tools for
>making Ada attractive to our students.
>
>Is there a fix for this problem?
>
>
>Richard Riehle
>Computer Science Department
>Naval Postgraduate School
>Monterey, CA
>7 November 2007
>_______________________________________________
>GAP mailing list
>GAP at gnat.info
>/no-more-mailman.html
>To unsubscribe from this list, please contact the GAP GNAT Tracker 
>administrator
>within your organization.


John W. McCormick                mccormick at cs.uni.edu
Computer Science Department
University of Northern Iowa        voice (319) 273-6056
Cedar Falls, IA 50614-0507       fax (319) 273-7123
http://www.cs.uni.edu/~mccormic/