On Jan 18, 4:22 am, CBFalconer <cbfalco...@yahoo.com> wrote:
> Malcolm McLean wrote:
> > "Lew Pitcher" <lpitc...@teksavvy.com> wrote in message

> >> OK, first thing is to recognize that there is no such thing as
> >> "standard-compliant C/C++ programming", because there is no
> >> standard that defines something called "C/C++".

> > Standards-compilant C/C++ would be a language that anyone would
> > call C, or C with a few twiddles, but actually conforms to the
> > C++ standard. Many thousands of such programs are written.

> No. Run the following program to get the idea:

> #include <stdio.h>

> int main(void) {
> int C;

> C = rand(void);
> if (C) printf("C / C++ == %d\n", C / C++);
> else printf("C / C++ is undefined\n");
> return 0;
> }

> and I expect the vast majority of results to be the value 1.

Funny, it doesn't compile with my compilers. Neither C nor C++.
Are you sure that there isn't a void too many in there?

Other than that, it may be legal C, but I don't think many C
programmers would intentionally write it that way; there'd be an
include <stdlib.h> in there as well.

(Of course, it would be a rare C program which didn't use malloc
somewhere. And typically, C programmers will *not* cast the
result of malloc(), which will result in a type error in C++. I
agree with you that there are probably very few programs in
existance which will compiler in both languages, but I think you
chose a bad example.)

--
James Kanze (GABI Software) email:james.kanze@gmail.com
Conseils en informatique orientée objet/
Beratung in objektorientierter Datenverarbeitung
9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34

On Jan 18, 4:17 pm, CBFalconer <cbfalco...@yahoo.com> wrote:
> jkherci...@gmx.net wrote:

> ... snip ...

> > All that I have observed is that there are some people who
> > regularly _pretend_ that those using the phrase "C/C++" do refer
> > to some language of that name. That joke was funny the first time
> > I saw it, but it wears off over the years.

> No, I think the users of the "C/C++" phrase tend to be idiots who
> think that C++ is a superset of C, rather than a different
> language. They also commit the crime (usually) of cross posting
> between the groups.

I suspect that it depends on which users, but there issues which
are common, and where it makes sense to speak of C/C++. Any
discussion of sequence points, for example. Or integral types.
Or threading. (The C and C++ committees are working together on
that one, so that the basic guarantees will be the same in both
languages.)

--
James Kanze (GABI Software) email:james.kanze@gmail.com
Conseils en informatique orientée objet/
Beratung in objektorientierter Datenverarbeitung
9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34

James Kanze wrote:
> CBFalconer <cbfalco...@yahoo.com> wrote:
>> Malcolm McLean wrote:
>>> "Lew Pitcher" <lpitc...@teksavvy.com> wrote in message
>
>>>> OK, first thing is to recognize that there is no such thing as
>>>> "standard-compliant C/C++ programming", because there is no
>>>> standard that defines something called "C/C++".
>
>>> Standards-compilant C/C++ would be a language that anyone would
>>> call C, or C with a few twiddles, but actually conforms to the
>>> C++ standard. Many thousands of such programs are written.
>
>> No. Run the following program to get the idea:
>
>> #include <stdio.h>
#include <stdlib.h>
>>
>> int main(void) {
>> int C;
>
>> C = rand(void);
C = rand();
>> if (C) printf("C / C++ == %d\n", C / C++);
>> else printf("C / C++ is undefined\n");
>> return 0;
>> }
>
> > and I expect the vast majority of results to be the value 1.
>
> Funny, it doesn't compile with my compilers. Neither C nor C++.
> Are you sure that there isn't a void too many in there?

And a missing #include of stdlib.h. :-) Also the C/C++ expression
is, strictly speaking, illegal. However it will work as long as
the two components are not calculated at the same time. :-) Bah
humbug. Time to kick the cat.

--
[mail]: Chuck F (cbfalconer at maineline dot net)
[page]: <http://cbfalconer.home.att.net>
Try the download section.



--
Posted via a free Usenet account from http://www.teranews.com

On Jan 19, 3:09 am, CBFalconer <cbfalco...@yahoo.com> wrote:
> James Kanze wrote:
> > CBFalconer <cbfalco...@yahoo.com> wrote:
> >> Malcolm McLean wrote:
> >>> "Lew Pitcher" <lpitc...@teksavvy.com> wrote in message

> >>>> OK, first thing is to recognize that there is no such thing as
> >>>> "standard-compliant C/C++ programming", because there is no
> >>>> standard that defines something called "C/C++".

> >>> Standards-compilant C/C++ would be a language that anyone would
> >>> call C, or C with a few twiddles, but actually conforms to the
> >>> C++ standard. Many thousands of such programs are written.

> >> No. Run the following program to get the idea:

> >> #include <stdio.h>

> #include <stdlib.h>

> >> int main(void) {
> >> int C;

> >> C = rand(void);
> C = rand();
> >> if (C) printf("C / C++ == %d\n", C / C++);
> >> else printf("C / C++ is undefined\n");
> >> return 0;
> >> }

> > > and I expect the vast majority of results to be the value 1.

> > Funny, it doesn't compile with my compilers. Neither C nor C++.
> > Are you sure that there isn't a void too many in there?

> And a missing #include of stdlib.h. :-)

Which shouldn't prevent it from compiling in C. I thought that
that was the point---it looks like C/C++, but in fact, is legal
C, but not legal C++.

> Also the C/C++ expression is, strictly speaking, illegal.
> However it will work as long as the two components are not
> calculated at the same time. :-)

But the results might be different:-). The compiler might do
the C++ first, then read the C for the left operand of the
division.

With the include of <stdio.h>, I think that the program is a
perfect example of the sort of thing C/C++ would be applicable
to---undefined behavior because of a missing sequence point is
one thing where C++ is intentionally compatible with C. (It's
worth noting that the C++ committee is working to add support
for threading. This obviously affects sequence point issues,
and the committee is submitting all work in this regard to the C
committee for approval---the goal is that when the C committee
adds threading, it will be 100% compatible with C++.)

> Bah humbug. Time to kick the cat.

Don't do that. I like cats.

--
James Kanze (GABI Software) email:james.kanze@gmail.com
Conseils en informatique orientée objet/
Beratung in objektorientierter Datenverarbeitung
9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34

James Kanze wrote:
> CBFalconer <cbfalco...@yahoo.com> wrote:
>
.... snip ...
>
>> Bah humbug. Time to kick the cat.
>
> Don't do that. I like cats.

My daughter is away until Monday, so I have to go and feed my
grandsons cat, and clean the kitty litter, and pet the beast. Lots
of motivation for kicking. :-)

--
[mail]: Chuck F (cbfalconer at maineline dot net)
[page]: <http://cbfalconer.home.att.net>
Try the download section.



--
Posted via a free Usenet account from http://www.teranews.com

In article <fmo8s6$imc$1@aioe.org>, nleschov@gmail.com says...
> Hello,
>
> I have done some programming, mostly on embedded systems but now I would
> feel like I have to learn good development practices (i.e programmer's
> good manners). Let's say I want to build an application using some
> established open source library or application. Do I just download the
> source and hack away? Or I'd better do it in some controlled manner, i.e
> set up some source control system so that I can always compute the diffs
> from the original code? Or maybe I should try to keep my changes
> separate from the original code base? How do I do it?

Almost regardless of what you're working with (your own code or
something you downloaded), a bit of control such as a source code
control system is usually a good idea.

You can't normally keep your changes entirely separate -- but you can
create a fork in any competent control system, and do diffs to keep
track of exactly what changes you're making.

> Where do I learn about things like that? I'm sorry for asking this here,
> where we dicuss standard-compliant C/C++ programming, but I really don't
> know any other place now. I do program in C/C++.

comp.software-eng is the most obvious place.

--
Later,
Jerry.

The universe is a figment of its own imagination.