W. J. Metzger wrote:
> So, what we, or in any case I, have learned is
>
> F requires all functions to be PURE, except that print statements are
> allowed but write(unit=*,fmt=*) is not.
Yes; the idea is to allow quick 'n' dirty print * statements
as a debugging aid. Such statements don't really affect the
state of anything in the program. It would be possible (and
perhaps more consistent) to also allow the WRITE statement,
but it is more complicated to figure out that both unit and
fmt are "*", so it didn't seem worth the effort. A different
choice would be to allow any WRITE statement; we simply
didn't make that choice.
> Further, it seems a good idea to update the version frequently.
> So, I immediately updated my linux box to release 030323
> But it doesn't work, giving the message
> /lib/libc.so.6: version `GLIBC_2.3' not found (required by F)
> and indeed, glibc on my machine is 2.2.5
> So, I guess I go back to 030106. Or is it better to go further back to
> 020224, which is still on the F download site?
I don't think I said it is always a good idea to update
frequently. I only said that *I* update the compiler
frequently. If the new version works for you, great;
if not, don't do it. That is exactly why I left version
020224 on the web site; it is pretty stable, but does not
correctly enforce the rules about the OPTIONAL and access
statements.
Of course, others complained when I *didn't* use the latest
glibc, so I can't win on that one. If updating gcc is not an
option, then don't update F either.
I do appreciate the comments and feedback. Thanks again to
all who provide it.
--
Walt Brainerd +1-877-355-6640 (voice & fax)
The Fortran Company +1-520-760-1397 (outside USA)
6025 N. Wilmot Road [log in to unmask]
Tucson, AZ 85750 USA http://www.fortran.com
|