[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 19990107: Re: ncvgt subroutines

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 19990107: Re: ncvgt subroutines
  • Date: Thu, 07 Jan 1999 08:59:42 -0700
Steve,

>Date: 7 Jan 1999 09:42:12 -0500
>From: "Steve Mauget" <address@hidden>
>Organization: USDA/ARS
>To: "Steve Emmerson" <address@hidden>
>Subject: Re: 19990106: Re: ncvgt sub
>Keywords: 199812301538.IAA00805

In the above message, you wrote:

>     This problem with compiling *.F files was something that I had problems
> with installing v.3.3.1. As of about 18 months ago, the people at NAG  claim
> that *.F are an exotic, non-standard extension (see your reply below).

Preprocessing .F files is non-standard, but hardly exotic.

> The f90
> compiler doesn't know what to do with them and forwards them to the loader.
> The earlier fix was to use c89 -E to preprocess *.F files. The previous
> version of our C compiler was able to pre-process *.F files (i.e., c89 -E
> worked  on *.F files), but for some reason the current version can't. At this
> point c89 -E ftest.F seems to go looking for cmplrs/mfe77 (an f77 utility ?)
> the we don't have or are not licensed for. No mention in the man pages for cc
> and cpp of an option to preprocess *.F files. Our SGI support contract (about
> $2000 for 1 year, or $250/hr, min 2 hrs) has expired, and we won't be deciding
> till the end of the month whether to extend it. For $2000 we could buy a PC,
> install Linux, f77, and netcdf and be on our way, which is another option.
> This SGI - NAG f90 - netCDF combination seems cursed. Anyway, thanks for the
> help and I'll let you know how it turns out.

If you can contrive a single command that will run the C preprocessor on
a .F file and write the results to standard output, then the configure
script can be told to use this command via the FPP environment variable,
e.g.

    setenv FPP "preprocF -E -Ireallymeanit"
    ./configure >&! configure.log

> The NAG people are correct in that the .F extension is used only by
> exotic systems.  Such systems include (but are not limited to) the
> following:
> 
>     System      Compiler
>     ------      --------
>     AIX         xlf
>     BSD/OS      fort77
>     HP-UX       fort77
>     IRIX        f77
>     IRIX64      f77
>     Linux       fort77
>     OSF/1       f77
>     SunOS 4     f77
>     SunOS 5     f77
>     UNICOS      f90
>     ULTRIX      f77

Did I write this?  Such sarcasm!  I like it!

I don't suppose using another "Fortran" compiler (e.g. fort77) is a
viable option?

--------
Steve Emmerson   <http://www.unidata.ucar.edu>