Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!yale!cmcl2!lanl!jlg
From: jlg@lanl.gov (Jim Giles)
Newsgroups: comp.lang.fortran
Subject: Re: global data
Message-ID: <10553@lanl.gov>
Date: 7 Jan 91 19:40:28 GMT
References: <1991Jan7.131620.27789@msuinfo.cl.msu.edu>
Distribution: usa
Organization: Los Alamos Natl Lab, Los Alamos, N.M.
Lines: 23

From article <1991Jan7.131620.27789@msuinfo.cl.msu.edu>, by fox@DASHER.NSCL.MSU.EDU:
> [...]
> .. and since INCLUDE is not part of the FORTRAN-77 standard, there's no
> portable way to get the effect of a global parameter statement.  [...]

Well, write your INCLUDE preprocessor in Fortran 77 (a simple task) and
the capability will be present on every machine that Fortran 77 runs on.
There are 'off the shelf' preprocessors written in Fortran 77 listed in
many books.

> [...]                                                            A question
> to those following FORTRAN-9x.. how is this problem dealt with (if at all)
> in that language?

Fortran Extended (the most recent, and let's hope final name) has both
INCLUDE and MODULE in it.  Either one will support global named constants
(parameters).  MODULES also support private constants, variables, and
procedures (so a library can have internal structure that cannot be
interfered with from outside).  MODULEs are among the few things that
the new standard does well.  INCLUDE is basically redundant when MODULEs
are available.

J. Giles