> Date: Fri, 25 Sep 1998 08:15:47 -0700 (PDT)
> Mime-Version: 1.0
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
> References: <[log in to unmask]>
> X-Mailer: VM 6.34 under 19.16 "Lille" XEmacs Lucid
> From: Richard Maine <[log in to unmask]>
> X-List: [log in to unmask]
> X-Unsub: To leave, send text 'leave comp-fortran-90' to [log in to unmask]
> X-List-Unsubscribe: <mailto:[log in to unmask]>
> Reply-To: Richard Maine <[log in to unmask]>
> Sender: [log in to unmask]
> Errors-To: [log in to unmask]
> Precedence: list
> Content-Transfer-Encoding: 7bit
>
> [log in to unmask] writes:
> > In a message dated 9/24/98 8:37:52 AM Pacific Daylight Time, [log in to unmask]
> > darmstadt.de writes:
> >
> > > > In C and also, I think, Fortran (tell me if I'm wrong, please)
> > > > external data (e.g., COMMON) is guaranteed to be initialized
> > > > effectively to 0 anyway.
> > >
> > > I don't think that the `guaranteed' part is true. It might be true
> > > for many of implementations, however.
> > According to ANSI C, named COMMON blocks (but not blank COMMON) which are
> > implemented C style will be initialized to (integer) zero....
>
> ANSI C talks about named COMMON blocks? Well, I'm not going to
> specifically say you are wrong, because I'm not fluent enough in the
> details of ANSI C to be sure (and I'm not prepared to comb the
> standard to check, particularly since I don't have an electronic
> copy to make it easy). But I must say I'd be quite surprised.
>
> Certainly no Fortran standard gives any such guarantee anout named
> COMMON blocks. That part I am confident of.
>
> Perhaps ANSI C talks about some construct which you are equating to
> Fortran's named COMMON because they are equivalent in many
> implementations. That wouldn't surprise me. But unless the standard
> specifically mentions Fortran named COMMON blocks, then any
> association you might make because of "typical implementations" is
> not part of the standard.
>
> I'm curious if you did manage to find something about Fortran named
> COMMON in the ANSI C standard. Stranger things have happened. If
> there is such a reference to Fortran in the ANSI C doc, I'd quite like
> to know the citation - its a citation I'd likely find use for.
Fortran doesn't even appear in the index to ANSI X3.159-1989.
It certainly doesn't make any guarantees of Fortran COMMON blocks.
> --
> Richard Maine
> [log in to unmask]
>
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|