Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 (Tek) 9/28/84 based on 9/17/84; site aeolus.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!vax135!cornell!uw-beaver!tektronix!teklds!aeolus!bobl
From: bobl@aeolus.UUCP (Bob Lewis)
Newsgroups: net.sources,net.wanted.sources
Subject: Re: Re: RATFOR sources needed
Message-ID: <27@aeolus.UUCP>
Date: Wed, 1-May-85 16:18:43 EDT
Article-I.D.: aeolus.27
Posted: Wed May  1 16:18:43 1985
Date-Received: Fri, 3-May-85 04:21:56 EDT
References: <352@creare.uucp> <25159@lanl.ARPA> <366@cvl.UUCP>
Reply-To: bobl@aeolus.UUCP (Bob Lewis)
Organization: Tektronix, Beaverton OR
Lines: 28
Keywords: ratfor, software tools
Xref: watmath net.sources:2867 net.wanted.sources:789

--------
Since the original query, I've seen a couple of requests for the same
information on the net, so I thought I'd post the response I sent
creare!gda.  Here goes:

I don't have F77 RATFOR sources, but I recall having to deal with this
problem at the last place I worked.  In that case, our target was an Apollo
(strict F77) from a VAX (VMS Fortran).  There is an easier approach than
getting a whole new Ratfor, though.

Given that you have one host on which DECUS RATFOR works and the source for
it, what you can do is create your own F77 boot file for the target on that
host by concatenating your *target* RATDEFS file to RATFOR.RAT (or whatever
you call them) and passing them through the host RATFOR to produce a
RATFOR.F77 that you can then transfer to and compile on the target.  The
only tough part, then, is modifying the RATDEFS file.

In the particular case of RATFOR characters, you'll probably want to

	define(character, INTEGER*2)

I know it takes more space, but I think you'll find that FORTRAN 77
CHARACTER*n's just don't have the same flexibility as integer arrays.

Good luck.

	- Bob Lewis
	  ...!tektronix!teklds!bobl