comp.lang.ada
 help / color / mirror / Atom feed
From: John Harbaugh <john.s.harbaugh2@boeing.com>
Subject: Re: Aonix bugs
Date: 1997/04/17
Date: 1997-04-17T00:00:00+00:00	[thread overview]
Message-ID: <335648D2.174A@boeing.com> (raw)
In-Reply-To: 334A8801.7054@gsfc.nasa.gov

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2703 bytes --]


Stephen Leake wrote:
> 
> S�bastien Auclair wrote:
> >
> > ( French if possible )------- ( French if possible )--------- ( French
> > if possible )
> >
> > I'd like to know if their's anyone who could explain me why my
> > "ObjectAda for Windows V7.0.232 from Aonix" doesn't seems to function
> > properly.
> 
> I have Aonix at home, and I'm at work now, so I don't know exactly which
> version I have. However, there is a LARGE patch file on the Aonix web
> site, that fixed lots of bugs when I downloaded and installed it. You
> can download it from:
> 
> http://www.aonix.com/Support/ada_index.html
> 
> >
> > Is there any other Ada 95 compiler available to work with.
> >
> 
> gnat works well under Windows 95, although I haven't used it for much
> actual Win32 coding yet. gnat didn't link the first time I tried, and
> Aoniz did, so I gave up on gnat. However, gnat has MUCH better error
> messages for first-time-thru code, so I run all my code thru gnat till
> it compiles, then link with Aonix.
> 
> Assuming Aonix is listening, I'll post my pet peeve with error messages:
> 
> procedure test is
>         procedure foo (A : in Integer; B : in Float);
> 
> begin
>         foo (1.0, 2);
> end;
> 
> This gives an error message "error in procedure call"; it does NOT say
> that parameter 1 should be Integer, but is a real type. gnat gives:
> 
> test.adb:4:09: expected type "Standard.integer"
> test.adb:4:09: found type universal real
> gnatmake: "test.adb" compilation error
> 
> It is amazing how much nicer Ada feels when you get the gnat message
> instead of the Aonix one (should gnat be capitalized here?).
> 
> > Thanks in advance...
> 
> You're welcome
> >
> > S�bastien Auclair
> > jg791569@er.uqam.ca
> 
> --
> - Stephe

I have been struggling to install the U2A and U2B patch files.  It turns
out that all unzip routines are not created equal.  Specifically, do not
use PKUnzip!  I was directed by Aonix tech. support to use WinZip95. 
This finally installed the patch.  In many cases, the patch is an
improvement - delay until no longer raises an NYI exception, and you can
call Ada.Interrupts routines without raising exceptions.

However, in some cases, Aonix has traded one set of bugs for another. 
For example, running build on a unit requiring recompilation generates
an illegal operation warning, rather than an object file.  Also, while
you can create projects in any directory, attempting to link with them
invokes the MS linker which dies with a fatal link error LLNll04: cannot
open file "C:\Ada95\New".  The only projects wich now work are those
created in the ObjectAda directory.

ObjectAda is getting better and will be a really nice product, maybe by
V7.1.




  reply	other threads:[~1997-04-17  0:00 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1997-04-07  0:00 Aonix bugs S�bastien Auclair
1997-04-08  0:00 ` Stephen Leake
1997-04-17  0:00   ` John Harbaugh [this message]
1997-04-18  0:00     ` Ben Brosgol
  -- strict thread matches above, loose matches on Subject: below --
1997-04-09  0:00 Dave Wood
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox