Michael Metcalf wrote:
>
> FYI.
>
> Mike Metcalf
>
> ----- Original Message -----
> From: "Richard Townsend" <[log in to unmask]>
> Newsgroups: comp.lang.fortran
> Sent: 12 August 2003 22:37
> Subject: New ISO_VARYING_STRING implementation (without memory leaks)
>
> >
> > Dear All,
> >
> > I've just finished implementing an ISO_VARYING_STRING module, conformant
> > to ISO/IEC 1539-2:1999, to enable the use of varying-length strings in
> > Fortran 95. Although there are already implementations of this module
> > available on the web (for instance,
> >
> ftp://ftp.nag.co.uk/sc22wg5/ISO_VARYING_STRING/Sample_Module/iso_vst.f95.gz),
> > these appear to have a problem with memory leaks.
> >
> > To avoid the possibility of leaks, my implementation of the module relies
> > upon ISO/IEC TR15581, the "official" extension to Fortran 95 which
> > permits allocatable derived-type components. While I'm no Fortran expert,
> > I'd be prepared to stake quite a bit of money that my implementation won't
> > leak memory on any non-broken compiler. Of course, the compiler will have
> > to support TR 15581; but nowadays most appear to do so, the Intel Fortran
> > compiler being a notable exception (hint hint, please sort that out,
> > Intel!)
Does TR 15581 requires or implies that an allocatable component of a variable
of a derived type shall deallocate itself? I am asking because I could not
find any deallocate() statement in this new ISO_VARYING_STRING module.
Automatic deallocation would be the only explaination in my opinion.
How can one tell if an allocatable variable has been deallocated in this case?
In another word, how can one tell if a compiler is non-broken?
Jing
> >
> > I'm releasing the module under the GPL, with the hope that some may find
> > it useful and/or correct my bugs and/or laugh at my coding style and/or
> > optimize it (its currently written for simplicity, not speed). If anyone
> > in this NG works for a compiler vendor, and wants to bundle it with their
> > product but can't due to the GPL, I'd be more than happy to release it
> > under a different licence -- just drop me an email.
> >
> > The source code for the module can be downloaded here:
> >
> > http://www.star.ucl.ac.uk/~rhdt/download/#iso
> >
> > (please feel free to put up mirror copies)
> >
> > cheers,
> >
> > Rich
> >
> > --
> >
> +-----------------------------------+-------------------------------------+
> > | Dr Richard Townsend |
> |
> > | Department of Physics & Astronomy | Do not send me HTML-formatted
> |
> > | University College London | messages; they go directly to
> |
> > | Gower Street | my spam folder, and will
> |
> > | London WC1E 6BT | not be read...
> |
> > | Work: +44 20 7419 3410 |
> |
> > | Mob.: +44 7703 439 009 |
> |
> >
> +-----------------------------------+-------------------------------------+
> >
--
________________________________ _-__-_-_ _-___--- _____________________
Jing Guo, [log in to unmask], (301)614-6172(o), (301)614-6297(fx)
Global Modeling and Assimilation Office, Code 910.3, NASA/GSFC
Greenbelt, MD 20771
|