Re: fortran 2003 compiler
- From: kargl@xxxxxxxxxxxxxxxxxxxxxxxxxxxx (Steven G. Kargl)
- Date: Fri, 2 Mar 2007 18:31:20 +0000 (UTC)
In article <1172843815.454589.157670@xxxxxxxxxxxxxxxxxxxxxxxxxxx>,
"Beliavsky" <beliavsky@xxxxxxx> writes:
On Mar 2, 8:21 am, "Tobias Burnus" <bur...@xxxxxxxx> wrote:
I think all Fortran compiler vendors to plan to fully support Fortran
2003, it only takes time.
That is not true, I am afraid. I think Salford/Silverfrost and Lahey/
Fujitsu for Windows are in maintenance mode. Someone recently wrote
here that
"According to one of their representatives that I communicated with
recently, Lahey is getting out of the Windows compiler business."
I would bet against a Lahey/Fujitsu Fortran 2003 compiler on any
platform, because they would want to amortize an investment in
developing a Fortran 2003 compiler over as many platforms as possible.
Well, if Lahey is leaving the vendor business, then of course they
aren't going to update their product. Do you expect Watcom to
release a Fortran 95 or 2003 compiler even though Watcom left the
vendor business years ago?
For some projects such as gfortran it depends on interest of a
developer (all do it in their spare time); their interest can be
directed by popular demand and by the number of other compilers which
support this feature. (Popular demand and many competing compilers
which support a given feature also speeds up the development for
commercial compilers.)
None of the gfortran developers are interested enough in Windows to
produce Mingw gfortran binaries in a timely manner, as Doug Cox does
for g95. Nor do I see instructions on the Gfortran wiki on how to
build the compiler from source.
Sound like you may want to volunteer to provide this service. Send
email to fortran@gcc and request help in setting up the compiler
scripts. FX and Jerry will probably be thrilled with the help.
For some Windows Fortranners, the ability to create GUI programs using
g95 or gfortran using something similar to Quickwin for DVF/CVF/IVF (I
think Absoft, Lahey, and Salford have analogous facilities) is more
important than F2003 features. I doubt the developer(s) of g95 and
gfortran want to invest their time in assisting Windows-only GUI
development. I wonder how many man-years it would take to develop
something like Quickwin for g95 and gfortran (I would want it to work
on both). I'd be willing to pitch in some money to pay a developer.
Interesting. Well, my last major contribution to gfortran will be
aiding the merge of the ISO C binding code into the mainline. After
that occurs, I plan to write a module that wraps GTK+ for my own
personal use. One may also be able to do something similar for
quickwin. I freely admit that I haven't looked at the quickwin API,
so perhaps it is a PITA to deal with MS libs.
--
Steve
http://troutmask.apl.washington.edu/~kargl/
.
- Follow-Ups:
- Re: fortran 2003 compiler
- From: Beliavsky
- Re: fortran 2003 compiler
- References:
- Re: fortran 2003 compiler
- From: Tobias Burnus
- Re: fortran 2003 compiler
- From: Beliavsky
- Re: fortran 2003 compiler
- Prev by Date: Re: Array intrinsics with zero-size array argument
- Next by Date: Re: Division of large negative number leads to integer overflow
- Previous by thread: Re: fortran 2003 compiler
- Next by thread: Re: fortran 2003 compiler
- Index(es):
Relevant Pages
|
