Some Questions.

[Kenny McCormack]

Note also that you sometimes see the wording "... written (entirely) in
ISO C ..." (most recently, I believe, in posts from Heathfield), and
whenever I see this wording, I think "but, but, that's redundant...
According to some of the nutbars here, there's no other kind."

Wrong .....

The Ansi C standard (mostly used here since you know who hates c99)
was adopted by ISO.

So it is MORE correct to say "100% Ansi C" ... in order to show off your
credentials and historical knowledge of the standards.....[/QUOTE]

Now, who's nitpicking...?

P.S. upthread (in the part I clipped), I said "<x> is not C - where <x>
is something not mentioned in the standard". I should have said:

"<that> is not C - where <that> contains, however tangentially, some <x>
that is not mentioned in the standard".

I.e., in the nutbar view, one tiny bit of OT poisons the whole work.

 

[Kenneth Brody]

Kenny McCormack wrote:
[...]

The point (and a very important point it is) is that many of the
regulars (e.g., CBF), not all of them, but some of them, have explicitly
stated both that "If it isn't in the C standard, it doesn't exist" (1) and
"<x> is not C" (2) (where <x> is something not mentioned in the standard).

Both statements ((1) and (2) above) are obviously bogus, but we, those
of us in the know, know what it means and accept it as poetic license.
But, and this is another important point, newbies don't know how to take
it, and the whole things has a rather unpleasant smell.

I'd have to see such a quote in context to see exactly what was said.
Of course, what the (probably) really mean is "it doesn't exist _as_
_far_as_the_standard_is_concerned_".

I have a cup of water next to me. It obviously exists, despite the
fact that the standard makes no mention of cups or water. (Heck, it
doesn't even mention "me".)

Note also that you sometimes see the wording "... written (entirely) in
ISO C ..." (most recently, I believe, in posts from Heathfield), and
whenever I see this wording, I think "but, but, that's redundant...
According to some of the nutbars here, there's no other kind."

I think they would agree that there are other forms out there. There
is "Microsoft C", there is "gcc C", there is "Watcom C", and so on.

The point is, that in clc, the unqualified use of "C" means "ISO C",
but some people need to be told that explicitly. I would probably
say that the software I write is "written entirely in C", but
technically (as far as clc is concerned) it should really be "written
in C, using extensions which are defined in other standards such as
POSIX, plus some other system-specific extensions". The language and
syntax of my code is pretty close to pure "ISO C", but it calls
functions which are not defined by the standard, and those functions
do things which by there very nature are system-specific.

--
+-------------------------+--------------------+-----------------------+
| Kenneth J. Brody | www.hvcomputer.com | #include |
| kenbrody/at\spamcop.net | www.fptech.com | <std_disclaimer.h> |
+-------------------------+--------------------+-----------------------+
Don't e-mail me at: <mailto:[email protected]>

 

[Keith Thompson]

The Ansi C standard (mostly used here since you know who hates c99)
was adopted by ISO.

So it is MORE correct to say "100% Ansi C" ... in order to show off your
credentials and historical knowledge of the standards.....

ANSI published a C standard in 1989 (C89).

ISO published a C standard, essentially identical to C89, in 1990.
ANSI then adopted the ISO C90 standard document, which superseded the
ANSI C89 document.

Then, of course, ISO published a revised C standard in 1999 (C99), and
ANSI adopted it.

The term "ANSI C" is, in my opinion, only of historical interest.
There was an ANSI C standard other than one adopted from ISO only for
about a year, 17 years ago.

"ISO C", strictly speaking, refers to the C99 standard, which
officially replaced the ISO C90 standard but has not yet replaced it
in actual usage. The correct way to refer to the older standard is
"C90" or "ISO C90" (or "ISO/IEC 9899:1990" if you want to be painfully
precise). I wouldn't use the unqualified phrase "ISO C" unless the
distinction between C90 and C99 were either unimportant or obvious
from the context.

The term "ANSI C" is still in common use due to historical inertia,
and it generally refersto the language defined by the C89 and C90
standards, but there's no good reason not to use the correct
terminology. It's a waste of time to have to explain that, yes, the
current official C standard recognized by ANSI is C99, but yes, what I
really mean is C89/C90.

 

[Kenny McCormack]

Kenny McCormack wrote:
[...]

The point (and a very important point it is) is that many of the
regulars (e.g., CBF), not all of them, but some of them, have explicitly
stated both that "If it isn't in the C standard, it doesn't exist" (1) and
"<x> is not C" (2) (where <x> is something not mentioned in the standard).

Both statements ((1) and (2) above) are obviously bogus, but we, those
of us in the know, know what it means and accept it as poetic license.
But, and this is another important point, newbies don't know how to take
it, and the whole things has a rather unpleasant smell.

I'd have to see such a quote in context to see exactly what was said.
Of course, what the (probably) really mean is "it doesn't exist _as_
_far_as_the_standard_is_concerned_".

I stand by my post, though it is to be granted that such things are
always phrased with enough weasel words so that they can claim that that
was not what they meant. But it is designed to say what it says
(Aesopean language).