Not mentioned by the others is the Borland C++ v5.5 that was released
last week.  It seems to be stable and compliant (although I've only
started to test it out).  Best of all it is free.  (For MS Windows.)

-Wayne Pollock,  pollock@acm.org

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

In article <951203464.680850@Stalker.Alfacom.net>, Igor Solodovnikov
<sivsoft@yahoo.com> writes
>4. I do not know what is Koenig lookup so it is not important to me.:)

It is if you do not know what it is, it can completely change the
meaning of your code by selecting an overload that you did not realise
was in the overload set. :)


Francis Glassborow      Journal Editor, Association of C & C++ Users
64 Southfield Rd
Oxford OX4 1PA          +44(0)1865 246490
All opinions are mine and do not represent those of any organisation

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

Mon, 21 Feb 2000 06:38:37 CST, Igor Solodovnikov <sivsoft@yahoo.com> pisze:

> Which c++ compiler is most standard compatible?

There is a comparison between a few at
http://animal.ihug.co.nz/c++/compilers.html

--
 __("<    Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/              GCS/M d- s+:-- a22 C+++$ UL++>++++$ P+++ L++>++++$ E-
  ^^                  W++ N+++ o? K? w(---) O? M- V? PS-- PE++ Y? PGP+ t
QRCZAK                  5? X- R tv-- b+>++ DI D- G+ e>++++ h! r--%>++ y-

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

Pete Forman <gsez020@kryten.bedford.waii.com> writes:

> Try http://animal.ihug.co.nz/c++/compilers
Should be: http://animal.ihug.co.nz/c++/compilers.html

> --
> Pete Forman
> Western Geophysical
> pete.forman@westgeo.com
--
Worik Turei Stanton           The hard disc has crashed       worik@mbmnz.co.nz
Auckland            Your life's work has been destroyed       worik@noggon.com
Aotearoa                      Squeeze trigger (yes/no)?      +64 9 309 8047
(New Zealand)    ----------------------------------------*=

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

Igor Solodovnikov wrote:
>
> Thank you for answer.
> Now i will try to narrow my question.
> 1. Your guess is rigth: i'm using Windows NT.
> 2. I do not care about stdlib implementations.
> 3. I want to know about core language implementation conformance. (And
> especially all features dealing with templates and namespaces. I think those
> features is most exciting relatively new part of the language. (My IMHO
> only))
> 4. I do not know what is Koenig lookup so it is not important to me.:)

3 and 4 are contradicting each other ;-)
Koenig lookup is a namespace feature that allows functions
(esp. operators) to be found if they are _not_ in the current
scope, but in the scope of one of their argument's types.

Example:

namespace matrix
{
  class Matrix { ... };

  Matrix operator+(Matrix, Matrix);
}

int main()
{
  matrix::Matrix a, b, c;
  // ...
  c = a + b;
  // ...
}

Now, operator+ is not in scope. That is, normal lookup won't
find the correct operator+. But Koenig lookup finds it, since
a abd b are of type matrix::Matrix from namespace matrix, and
therefore operator+ is searched for in namespace matrix as
well, and found there.

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

Which c++ compiler is most standard compatible?
If this question is oftopic then please give me new direction to research
this question.


---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

In article <951047060.150113@Stalker.Alfacom.net>, sivsoft@yahoo.com
says...
> Which c++ compiler is most standard compatible?
> If this question is oftopic then please give me new direction to research
> this question.

The good news is that the moderators have considered conformance of
implementations to be topical in the past, so there's a pretty fair
chance that they'll allow the discussion.

The bad news is that there's no single answer, and if there was it
probably wouldn't matter anyway.

To mean much of anything, you usually have to restrict the discussion
to implementations for the platform you're using: if you're writing
programs for a SPARC box running Solaris, chances are that it makes
little difference to you HOW good a compiler is available for a
Macintosh running OS X.

The other part of the problem in answering this question is that it
involves some value judgements: if compilers X and Y support features
A and B respectively (but neither compiler supports both) deciding
which is closer to conforming basically involves deciding whether
feature A or B is the more important one.  This, in turn, can depend
heavily upon the code you write, and is likely to vary from one person
to the next (E.g. Koenig lookup is probably more important to me than
supporting the "export" keyword, but the opposite is undoubtedly true
for many people).

Based on not having said otherwise, I'm going to guess that your
target is an Intel box running some variety of Windows.  In this case,
I think the race really comes down to three or maybe four contenders:
1) Metrowerks Codewarrior
2) Comeau C++
3) IBM VA C++
4) Intel C++

Among those it may be somewhat difficult to pick out a clear winner.
Comeau ships with no standard library of its own, so part of its
conformance depends on the standard library you use with it.  Intel
ships with no standard library of its own either, but is at least
intended to be used strictly with the MS Visual C++ library, which has
a fair number of problems of its own.  Dinkumware has an update to the
VC++ library that probably helps out quite a bit though.

Between those two, it looks to me like Comeau is probably a bit closer
to conformance, but I haven't had it very long so I haven't been able
to test it very thoroughly.

The other two (IBM and Metrowerks) come as complete, pre-packaged
solutions.  I haven't used (or tested) recent versions of either one
enough to say with certainty, but from what I've seen, I'd have to
guess that IBM is probably a bit closer.  In fact, if I HAD to give a
direct answer to your original question, I'd say that THE compiler
that comes the closest to absolute conformance (IMO) is IBM VA C++.
Given any chance at all, I'd add a lot of qualifications trying to say
that 1) depending on what you're doing you might not see it that way
and, 2) it probably doesn't matter anyway.  Despite this, if I had to
give a single answer, that's what it would be...

--
    Later,
    Jerry.

The universe is a figment of its own imagination.

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

Thank you for answer.
Now i will try to narrow my question.
1. Your guess is rigth: i'm using Windows NT.
2. I do not care about stdlib implementations.
3. I want to know about core language implementation conformance. (And
especially all features dealing with templates and namespaces. I think those
features is most exciting relatively new part of the language. (My IMHO
only))
4. I do not know what is Koenig lookup so it is not important to me.:)

"Jerry Coffin" <jcoffin@taeus.com> wrote in message
news:MPG.131b262f9ef09279989b1f@news.mindspring.com...
> Based on not having said otherwise, I'm going to guess that your
> target is an Intel box running some variety of Windows.  In this case,
> I think the race really comes down to three or maybe four contenders:
> 1) Metrowerks Codewarrior
> 2) Comeau C++
> 3) IBM VA C++
> 4) Intel C++



---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

"Igor Solodovnikov" <sivsoft@yahoo.com> writes:

> Which c++ compiler is most standard compatible?

Please have a look at http://animal.ihug.co.nz/c++/compilers.html
which tries to answer this question.

Regards,
Martin

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

Try http://animal.ihug.co.nz/c++/compilers
--
Pete Forman
Western Geophysical
pete.forman@westgeo.com

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]