"Dewar's compiler", was Re: Learning Windows 95 programming with Ada?

"Dewar's compiler", was Re: Learning Windows 95 programming with Ada?

[Tom Moran]

The original message said:
>I am currently using the GNAT compiler for MS-DOS ...
> ... Any recommendations?

Dewar's response said:
>... the way to proceed is to get hold of the Intermetrics
>Win32 bindings. These are very close to the C bindings, so you can get
>anyone of many books on these bindings to know what the routines do.
>They work with both GNAT and with Object Ada. ...

Now Dewar says:
>... started off with a question from someone who specifically
>said they were currently using GNAT ...
>By what stretch of the imagination is that a "recommendation of GNAT"?

  There are two different things with the same name here.  "GNAT" as
mentioned by the original poster refers to an old compiler he
possesses that was developed by, but is no longer supported by, ACT,
that runs under MS-DOS.  "GNAT" as mentioned as part of "the way to
proceed" by Dewar I understand to be Windows GNAT, a current product
line of ACT.  If I'm mistaken, and the questioner can in fact use his
existing MS-DOS GNAT along with the Intermetrics bindings to produce
Windows 95 programs, and that is what Robert Dewar was suggesting,
please correct me.

  Since "GNAT" is a family name of related compilers of different
vintages, platforms, sources, and prices, perhaps it would be a good
idea in the future to refer to specific subsets of that family to
avoid such tedious discussions.

Re: "Dewar's compiler", was Re: Learning Windows 95 programming with Ada?

[Robert Dewar]

Tom said

<<REPORT ON NILE NIGHTLY RUN (97-12-02)
-------------------------------------

Potential regressions (old)
---------------------------
(none)

Potential regressions (new)
---------------------------
4721-009:CRASH  new crash (new gcc problem?)

OK stuff
--------
1416-008:DIFF  OK, new bogus warning msg, fixed by i-cpoint v1.12
1522-001:DIFF  OK, new bogus warning msg, fixed by i-cpoint v1.12
1618-001:DIFF  OK, new bogus warning msg, fixed by i-cpoint v1.12
1622-005:DIFF  OK, new bogus warning msg, fixed by i-cpoint v1.12
1718-007:DIFF  OK, correct new warning message, new test.out filed
1821-015:DIFF  OK, incorrect test.out, new test.out filed
2028-003:DIFF  OK, correct new warning message, new test.out filed
2317-009:DIFF  OK, correct new warning message, new test.out filed
2405-005:DIFF  OK, incorrect test.out, new test.out filed
2414-006:DIFF  OK, correct new warning message, new test.out filed
2524-007:DIFF  OK, correct new warning message, new test.out filed
2726-005:DIFF  OK, correct new warning message, new test.out filed
3209-014:DIFF  OK, correct new warning message, new test.out filed
3209-015:DIFF  OK, correct new warning message, new test.out filed
3212-002:DIFF  OK, new bogus warning msg, fixed by i-cpoint v1.12
3421-003:DIFF  OK, correct new warning message, new test.out filed
3609-003:DIFF  OK, correct new warning message, new test.out filed
4406-001:DIFF  OK, new bogus warning msg, fixed by i-cpoint v1.12

Re: "Dewar's compiler", was Re: Learning Windows 95 programming with Ada?

[Robert Dewar]

Tom Moran says

<<  There are two different things with the same name here.  "GNAT" as
mentioned by the original poster refers to an old compiler he
possesses that was developed by, but is no longer supported by, ACT,
that runs under MS-DOS.  "GNAT" as mentioned as part of "the way to
proceed" by Dewar I understand to be Windows GNAT, a current product
line of ACT.  If I'm mistaken, and the questioner can in fact use his
existing MS-DOS GNAT along with the Intermetrics bindings to produce
Windows 95 programs, and that is what Robert Dewar was suggesting,
please correct me.
>>

I hate to belabour the issue, but the product I was recommending here
is the generally available intermetrics binding to Win95. I have no
proprietary interest in this product, which as far as I know, works
with all Ada 95 Windows compilers (and I specifically mentioned that
it worked with both GNAT and Object Ada).

Yes, the Intermetrics bindings compete with Claw
Yes, Tom has some kind of interest in Claw
No, Robert has no financial interest in the Intermetrics bindings
No, Robert has no financial interest in Claw
Both Claw, and the Intermetrics bindings work with GNAT

I recommended the Intermetrics bindings because as far as I know there
is no freely available distribution of CLAW, but perhaps I am wrong on
that. If CLAW is now free software, then Tom should correct things.

As for Windows GNAT being a product of ACT, it is most certainly true
that ACT offers commercial support for the Win95/NT version of GNAT,
and we have many customers who pay for this support.

However, public versions of this technology are available for free
downloading at many sites, and I am assuming that this student of
Ada will go the appropriate locations to figure out how to get hold
of the necessary tools to explore further. Note that versions of
Object Ada suitable for this kind of learning are also available
freely, so that is why I mentioned both possibilities. If there are
other Ada compilers for Win95 that are freely available, let me
know, I will include them in my list next time!

Robert Dewar

Re: "Dewar's compiler", was Re: Learning Windows 95 programming with Ada?

[Robert Dewar]

My strange newsreader perversely posted a message with the completely
wrong text on this thread, actually a list of the results of our nightly
run with GNAT last night. I aborted it, but my newsreader did not believe
it :-)

Anyway I canceled it now, if you saw it, you can just ignore it!