Hi John,
On Tue, Jun 5, 2012 at 7:52 PM, John Harper <[log in to unmask]> wrote:
> I have also produced a web page, mainly about Fortran but with a little
> Maple as well. It includes notes on F95 for F77 users (who still exist
> in large numbers over 20 years after F77 was superseded.) It's at
>
> http://homepages.ecs.vuw.ac.nz/~harper/
>
> As my university has no compiler that implements all of F2003, but several
> that implement all of F95, with support for different F2003 features, I try
> to avoid non-F95 features.
>
> I agree with the comment that Metcalf, Reid & Cohen's book is good, but I
> would add Adams, Brainerd, Hendrickson, Maine, Martin & Smith "The Fortran
> 2003 Handbook" and Adams, Brainerd, Martin, Smith & Wagener "Fortran 95
> Handbook" on those two Fortran versions. I find that MR&C is often better
> for learning things the first time, but the Handbooks are often better for
> looking up something afterwards.
Thanks for your email. And for the book tips. I've added the
"The Fortran 2003 Handbook" here:
http://fortran90.org/src/faq.html#what-are-good-books-to-learn-fortran-from
(You may need to refresh your browser to see the change.)
I think that in general, almost all compilers now support at least F2003,
but I agree with you about F95, in fact for a library that I am about
to publish I also decided to stick to F95 for maximum portability,
as I happened to only be using minimum features from F2003.
However, my aim with the web pages is to document the most
modern "canonical" way of doing things, because in a few years
all compilers will support it and the F2003 and F2008 bring
a lot of little things that make working with Fortran comfortable
and natural (E.g. for example the array syntax [1, 2, 3].)
Ondrej
|