Re: Overload problem

Arjen Markus <arjen.markus@xxxxxxxxxx> wrote:

BTW, the allocatable attribute is not allowed on dummy
arguments, IIRC, unless the compiler allows it as an
extension.

Please note that, while this is literally true of f95, the extension in
question is an formal ISO TR. One might say that it is a formally
standardized extension. Also, the feature is incorporated into the base
language in f2003. I'm aware that there are as yet no f2003 compilers.

I think it important to distinguish between vendor-specific extensions
and ones like this. I certainly agree that it is important to be aware
when one is using any kind of extension; there do still exist f90
compilers that don't yet implement the TR (gfortran, for example, though
I see that the gfortran folk are hard at work on it). But one's
criterion for acceptability of an extension might well be quite
different for vendor-specific extensions vs forward-looking features
from later standards.

I agree with your base points about not being able to use allocatable vs
pointer to disambiguate (though some people have argued for allowing
that - I don't recall whether or not they sold the idea sucessfully for
f2008), and about not needing it anyway for the code shown. It is just
the abovee BTW that I am quibbling with.

I am not quite sure if the pointer attribute should be used to
identify the correct procedure. I expect it is, but others
will be able to confirm this.

I can't follow what you are saying here, so I can neither confirm nor
deny it.

--
Richard Maine | Good judgment comes from experience;
email: my first.last at org.domain| experience comes from bad judgment.
org: nasa, domain: gov | -- Mark Twain
.

Relevant Pages

  • Re: The chart of comp.lang.c
    ... That's topical in the groups for those compilers. ... compiler (e.g., how the extension fits into the general design of C, ... But normally if you implement something like, say, exceptions or arbitrary-sized bitfields, you'd want it to be incorporated in a future standard. ... If you don't it will be something very narrow, like say fuzzy logic ops in place of the standard ones, but that would go in a newsgroup dealing with AI, fuzzy programming, machine learning type problems. ...
    (comp.lang.c)
  • Re: Pro*Cobol moving to Microfocus - getting 1403 & 1405 errors because of precompiler??
    ... Compilers that allow a NEXT SENTENCE with an END-IF are *not* ... extension) than those that don't. ... "Howard Brazee" wrote in message ...
    (comp.lang.cobol)
  • Re: Is this code snippet safe ?
    ... >> char * ... Visual is one of the compilers that keys off of filename (or ... extension, but people like to argue about what extension ... Randy Howard ...
    (comp.programming)
  • Re: Any plans to remove obsolescent features?
    ... and then re-introduce it with the new meaning in the next version of the standard. ... This would give compilers a few years of freedom to choose whether to support the old semantics or the new as an extension. ... I remember it being fairly common. ...
    (comp.std.c)
  • Re: help compiling Python on vs 2008!
    ... the PCBuild8 directory in the Python 2.5 source. ... Building Python using VC++ 8.0 ... (For other Windows platforms and compilers, ... I did not manage to build extension with distutils with Python compiled ...
    (comp.lang.python)