Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!ames!hc!lanl!jlg
From: jlg@lanl.gov (Jim Giles)
Newsgroups: comp.lang.fortran
Subject: Re: Fortran vs C for computations
Message-ID: <3644@lanl.gov>
Date: 15 Sep 88 22:50:39 GMT
References: <1510@ficc.uu.net>
Organization: Los Alamos National Laboratory
Lines: 27

From article <1510@ficc.uu.net>, by peter@ficc.uu.net (Peter da Silva):
> Let's spell this out.
> 
> There is no standard program available with all Fortran compilers that
> does preprocessing of the program (conditional compilation, etc). This
> means you can't write and distribute a Fortran version of Emacs, say,
> with all the different environments conditionally compiled in. Why not?
> because you can't depend on the guy at the other end having it.

Let's spell this out.  

Since the preprocessor can be written in _standard_Fortran_77, it doesn't
matter that it's not distributed with the compiler.  I can distribute
the preprocessor _with_ the code.  Preprocessors are _very_ simple to
write - I've written them before - I'll probably write them again.
The only thing the 'guy at the other end' needs to have is a standard
conforming compiler.  If he hasn't got _that_ he won't be able to 
run my code anyway.  The _same_ complaint can be directed against
C - if the target machine hasn't got a correctly working C compiler,
you can't port!

The main reason _not_ to put the preprocessor into Fortran itself
is that they might do something like C did - leave out many useful
features and include some which are incompatible with the ones I
want.  I have to write my own preprocessors for C too. 8-(

J. Giles