Fortran - Fortran 2003 -- when?

Author

Fortran 2003 -- when?

Michael Prager

2007-06-26, 10:06 pm

Is there any indication from Fortran vendors of imminent release
of a Fortran 2003 compiler? I don't mean some features included
in an F95 compiler, but a true F2003 compiler.

If not imminent, any guesses as to when it might be? If readers
of this group can't give some educated guesses, I don't know who
can.

--
Mike Prager, NOAA, Beaufort, NC
Address spam-trapped; remove color to reply.
* Opinions expressed are personal and not represented otherwise.
* Any use of tradenames does not constitute a NOAA endorsement.

Beliavsky

2007-06-26, 10:06 pm

On Jun 26, 10:35 am, Michael Prager <Mike.Prager.ind...@noaa.gov>
wrote:
> Is there any indication from Fortran vendors of imminent release
> of a Fortran 2003 compiler? I don't mean some features included
> in an F95 compiler, but a true F2003 compiler.
>
> If not imminent, any guesses as to when it might be? If readers
> of this group can't give some educated guesses, I don't know who
> can.

This newsgroup is a good place for your question, but some Fortran
experts, including people who develop compilers, participate in the
"Fortran 90 list" http://www.jiscmail.ac.uk/lists/comp-fortran-90.html
and not this one, so you could also ask there. Registration is
required, but the list is open to anyone interested in Fortran 90 and
later versions of the language.

Richard Maine

2007-06-26, 10:06 pm

Michael Prager <Mike.Prager.indigo@noaa.gov> wrote:

> Is there any indication from Fortran vendors of imminent release
> of a Fortran 2003 compiler? I don't mean some features included
> in an F95 compiler, but a true F2003 compiler.
>
> If not imminent, any guesses as to when it might be? If readers
> of this group can't give some educated guesses, I don't know who
> can.

Well, I saw a note about IBM's newly released version of XLF here
recently. Checking the advertising material on it, it seems to be darned
close. It claims to implement all the major features other than
parameterized derived types. While that "other than" keeps it from quite
counting as an f2003 compiler, it is getting pretty close. I'd say
that's the first one that comes close enough that I might argue that it
is "almost an f2003 compiler" instead of being "an f95 compiler with
several f2003 features."

Pity that I don't have any suitable machines here at home to test it on.
I'm unlikely to get any either. :-)

--
Richard Maine | Good judgement comes from experience;
email: last name at domain . net | experience comes from bad judgement.
domain: summertriangle | -- Mark Twain

Michael Prager

2007-06-26, 10:07 pm

nospam@see.signature (Richard Maine) wrote:

>
> Well, I saw a note about IBM's newly released version of XLF here
> recently. Checking the advertising material on it, it seems to be darned
> close. It claims to implement all the major features other than
> parameterized derived types. [...]
>
> Pity that I don't have any suitable machines here at home to test it on.
> I'm unlikely to get any either. :-)

Richard, with that fat Federal retirement check, you should be
able to buy a mainframe or two! ;-)

Mike

--
Mike Prager, NOAA, Beaufort, NC
Address spam-trapped; remove color to reply.
* Opinions expressed are personal and not represented otherwise.
* Any use of tradenames does not constitute a NOAA endorsement.

Richard Maine

2007-06-26, 10:07 pm

Michael Prager <Mike.Prager.indigo@noaa.gov> wrote:

> nospam@see.signature (Richard Maine) wrote:
>
>
> Richard, with that fat Federal retirement check, you should be
> able to buy a mainframe or two! ;-)

IBM also has some PPC workstations that I actually could afford without
a horrible stretch. I haven't checked recently, but I think you can
still get modest ones without going over 4 digits, which is a lot in
some senses, but within the range of "big boy toys". (As in "the main
difference between men and boys is the price of their toys".) Heck, I
blew 2.5k on this 24" iMac, which is hard to justify except that I like
it and can afford to indulge that like. A PPC workstation just doesn't
seem to make it up very high on my priority list for personal funds. :-)

--
Richard Maine | Good judgement comes from experience;
email: last name at domain . net | experience comes from bad judgement.
domain: summertriangle | -- Mark Twain

Paul van Delst

2007-06-26, 10:07 pm

Michael Prager wrote:
> nospam@see.signature (Richard Maine) wrote:
>
>
> Richard, with that fat Federal retirement check, you should be
> able to buy a mainframe or two! ;-)

There's no need to buy - just lease one. Isn't that how it works already? :o)

cheers,

paulv

--
Paul van Delst Ride lots.
CIMSS @ NOAA/NCEP/EMC Eddy Merckx

ian_d_chivers@yahoo.co.uk

2007-06-26, 10:07 pm

The last edition (April) of Fortran Forum had an article on the f2003
status of
Cray, gfortran, g95 and nag.

The august edition has added intel and ibm.
i posted the paper copy to the acm today and the pdf version
went electronically on sunday (I think).

it is due out in august.

Ian Chivers

Fortran Forum Editor

On 26 Jun, 15:35, Michael Prager <Mike.Prager.ind...@noaa.gov> wrote:
> Is there any indication from Fortran vendors of imminent release
> of a Fortran 2003 compiler? I don't mean some features included
> in an F95 compiler, but a true F2003 compiler.
>
> If not imminent, any guesses as to when it might be? If readers
> of this group can't give some educated guesses, I don't know who
> can.
>
> --
> Mike Prager, NOAA, Beaufort, NC
> Address spam-trapped; remove color to reply.
> * Opinions expressed are personal and not represented otherwise.
> * Any use of tradenames does not constitute a NOAA endorsement.

Beliavsky

2007-06-26, 10:07 pm

On Jun 26, 5:24 pm, ian_d_chiv...@yahoo.co.uk wrote:
> The last edition (April) of Fortran Forum had an article on the f2003
> status of
> Cray, gfortran, g95 and nag.
>
> The august edition has added intel and ibm.
> i posted the paper copy to the acm today and the pdf version
> went electronically on sunday (I think).
>
> it is due out in august.

I doubt that subscriptions to Fortran Forum generates significant
revenue for the ACM. Could you ask them if PDF files could be posted
somewhere?

Steven G. Kargl

2007-06-26, 10:07 pm

In article <1182893077.929854.214400@w5g2000hsg.googlegroups.com>,
ian_d_chivers@yahoo.co.uk writes:
> The last edition (April) of Fortran Forum had an article on the f2003
> status of
> Cray, gfortran, g95 and nag.
>
> The august edition has added intel and ibm.
> i posted the paper copy to the acm today and the pdf version
> went electronically on sunday (I think).
>
> it is due out in august.
>
> Ian Chivers
>
> Fortran Forum Editor

By the time it's published, the gfortran information will
probably be out dated. :)

I'm in the final testing stage of a patch that implements
most of the ISO C binding feature. The only missing parts
of ISO C binding are those that depend on other F2003 features
(that gfortran has implement, yet; most notably procedure
pointers).

PS: A Google SoC project is addressing procedure pointers.

--
Steve
http://troutmask.apl.washington.edu/~kargl/

Phillip Helbig---remove CLOTHES to reply

2007-07-01, 8:05 am

In article <fs8283dv8o33atna8hm3mnpt40rhrkrr6s@4ax.com>, Michael Prager
<Mike.Prager.indigo@noaa.gov> writes:

> Is there any indication from Fortran vendors of imminent release
> of a Fortran 2003 compiler? I don't mean some features included
> in an F95 compiler, but a true F2003 compiler.
>
> If not imminent, any guesses as to when it might be? If readers
> of this group can't give some educated guesses, I don't know who
> can.

I realise that, traditionally, programming languages use a year as a
"version number". However, the fact that now YEARS are going by between
a "standard year" and the FIRST appearance of a corresponding compiler
really does make a bad impression on the folks who exclaim "is that
still legal?" when they learn that I am using Fortran.

There was a Dilbert cartoon where the PHB told Dilbert to just shift the
product now and move all the stuff in the "known bugs" list into the
"future features" list. I'm not suggesting such drastic measures, but
dedicating a fraction of the standardisation effort to "politics" or
"marketing" might more than pay for itself. Sadly, rumours of death
become self-fulfilling prophecies all to easily.

Dan Nagle

2007-07-01, 8:05 am

Hello,

Phillip Helbig---remove CLOTHES to reply wrote:

> Sadly, rumours of death
> become self-fulfilling prophecies all to easily.

In that situation, I usually ask the C++ programmer when they expect
to use a C++ 2003 compiler. Many don't know there was
a C++ 03 revision.

--

Dan Nagle
Purple Sage Computing Solutions, Inc.

Phillip Helbig---remove CLOTHES to reply

2007-07-01, 8:05 am

In article <f67tp4$5f4$1@online.de>, helbig@astro.multiCLOTHESvax.de
(Phillip Helbig---remove CLOTHES to reply) writes:

> There was a Dilbert cartoon where the PHB told Dilbert to just shift the

I meant "ship" not "shift" but I suppose it doesn't make much
difference.

Colin Paul Gloster

2007-07-18, 7:12 pm

In news:1182894329.376421.22870@c77g2000hse.googlegroups.com
timestamped Tue, 26 Jun 2007 14:45:29 -0700,
Beliavsky <beliavsky@aol.com> posted:
|-----------------------------------------------------------------------|
|"[..] |
| |
|I doubt that subscriptions to Fortran Forum generates significant |
|revenue for the ACM. Could you ask them if PDF files could be posted |
|somewhere?" |
|-----------------------------------------------------------------------|

Frankly, could you pay for a subscription?

Regards,
Colin Paul Gloster

Colin Paul Gloster

2007-07-18, 7:12 pm

In news:f67tp4$5f4$1@online.de timestamped Sun, 1 Jul 2007 09:56:52
+0000 (UTC), helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove
CLOTHES to reply) posted:
|------------------------------------------------------------------------|
|"In article <fs8283dv8o33atna8hm3mnpt40rhrkrr6s@4ax.com>, Michael Prager|
|<Mike.Prager.indigo@noaa.gov> writes: |
| |
|[..] |
| |
|I realise that, traditionally, programming languages use a year as a |
|"version number". However, the fact that now YEARS are going by between|
|a "standard year" and the FIRST appearance of a corresponding compiler |
|really does make a bad impression on the folks who exclaim "is that |
|still legal?" when they learn that I am using Fortran. |
| |
|[..]" |
|------------------------------------------------------------------------|

How does this differ from C; C++; Ada; or VHDL? What language or
languages do they deem to be fine?

Regards,
Colin Paul Gloster

Sponsored Links