Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!ames!ucbcad!ucbvax!vangogh.Berkeley.EDU!gnu
From: gnu@vangogh.Berkeley.EDU (John Gilmore)
Newsgroups: comp.lang.fortran
Subject: Anybody fixed EFL since 1982?
Message-ID: <22376@ucbvax.BERKELEY.EDU>
Date: 3 Jan 88 03:25:34 GMT
References: <253@hub.ucsb.edu>
Sender: usenet@ucbvax.BERKELEY.EDU
Reply-To: gnu@toad.com (John Gilmore)
Organization: Nebula Consultants in San Francisco
Lines: 22
Keywords: dinosaur

In article <253@hub.ucsb.edu> Matt Wette, mwette%gauss@hub.ucsb.edu, writes:
>                      ...EFL seems to be very buggy.  Try a "do loop" with
> the lower bound 0.  I also encountered lots of problems with "write"'s.
> My feeling is that this is probably not a well maintained piece of software
> and I advise using caution with it.

I just tried compiling efl with the GNU C compiler (ansi draft standard).
The sources at UC Berkeley haven't been touched since 1982.  The code
assumes that C is a typeless language -- it declares all its pointers as
"typedef int *ptr;"s and then references structure members with them.
This does not work in ANSI C -- and shouldn't have in old C.  I suspect
that the code (or the mind of the author, Stu Feldman) was directly
translated from BCPL.

I'm wondering if anybody actually uses this beast, and/or if anyone
has fixed it up so it passes "lint".  If so, could you ship a copy
to Keith Bostic at Berkeley (bostic@okeeffe.berkeley.edu)?  I suspect
it will be dropped from the releases unless somebody has either a
working version or a strong need for it.

	John Gilmore
{pyramid,ptsfa,amdahl,sun,ihnp4}!hoptoad!gnu                      gnu@toad.com
                "Watch me change my world..." -- Liquid Theatre