NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/std/c/228

Text File | 1996-08-06 | 3.5 KB | 69 lines

Newsgroups: comp.lang.c,comp.std.c
Path: phcoms4.seri.philips.nl!panther!baynes
From: [e-mail] (Stephen Baynes)
Subject: Re: Integral conversion e.t.c. (was: Re: Hungarian notation)
Sender: [e-mail] (account for localnews)
Message-ID: <[e-mail]>
Date: Tue, 30 Jan 1996 08:42:42 GMT
References: <[e-mail]> <[e-mail]> <4edqh2$[e-mail]> <[e-mail]> <4eindq$[e-mail]>
Organization: Philips Semiconductors, Southampton, UK
X-Newsreader: TIN [version 1.2 PL2]
Followup-To: comp.lang.c,comp.std.c

Peter Seebach ([e-mail]) wrote:
: In article <[e-mail]>,
: James Kanze US/ESC 60/3/141 #40763 <[e-mail]> wrote:
: >In article <4edqh2$[e-mail]> [e-mail]
: >(Peter Seebach) writes:
: >In the particular case which you cite, I do *not* believe that gcc
: >will compile the program without a diagnostic if the program is in a
: >file called a.c.  More ever, the compiler documentation definitly
: >states that only programs in files whose names end in .c will be
: >compiled as C.  (A legal restriction, as far as I can see.)

: That's an interesting question; the standard does not visibly make any
: such requirements, but you could argue that gcc is only a conforming compiler
: when invoked on a file ending in .c.

: Yes, it would *have* to give a diagnostic for a Fortran program; nonetheless,
: the program is *still* conforming if it is accepted.  I think this is a
: botch; the standard should probably distinguish between extensions and
: violations.

I think you need to distinguish between extensions (which a conformant compiler
must give a diagnostic) and parts of the magic incantations needed to run the
compiler (in ansi mode). Nameing a file to end '.c' is an incantation
along with -strict or -Aa. If you get it wrong you it is up to the implementor.
However if having perfomed the correct incantation - if the compiler fails
to give a diagnostic required by the standard then that is a bug in the
compiler - even if it then goes on to generate useful code. The whole point
about a standard compiler is that if code compiles clean on one, it should
compile clean on all.



: >Yes and no.  In comp.std.c, the discussion is exactly about what is
: >legal according to the standard.  In comp.lang.c (to which this
: >article is also cross-posted), the discussion should generally be a
: >bit more open.

: It's debated.  Generally, questions about features of specific compilers or
: OS's are off-topic, even when they relate to conforming programs.  Only
: strictly conforming programs (or attempts at such, when people are asking
: questions) are going to be usable and testable by the whole readership.
: (I doubt you can find a single extension that works on *all* machines
: we use.)

I would regard it as reasonable in comp.std.c to also to permit occasional
discussion on other well defined and reasonably well known standards in
their impact on how one can use the C language. The only standards I can think
of at the moment are POSIX/XOPEN, IEEE floating point and the one that came
up last year to do with arithmetic (the name of which I have forgotton).


--
Stephen Baynes                              [e-mail]
Philips Semiconductors Ltd
Southampton                                 My views are my own.
United Kingdom