Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!ucbvax!madis.af.mil!hansen
From: hansen@madis.af.mil (John Hansen)
Newsgroups: comp.soft-sys.andrew
Subject: Re: troff and transcript
Message-ID: <9007301353.AA17159@madis.af.mil>
Date: 29 Jul 90 13:23:00 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 31

With regard to printing without ditroff and Typescript and the possibility
of using groff instead, I have semi-successfully done just that.  I used g++
to compile groff (make sure you have cperf or gperf for gpic) and have
tested ez with no insets and it works OK.  I had to preference the 
formatcommand to:
groff -T$PRINTERTYPE -e /tmp/$s.n |

I also had to ln -s a devpsc directory to the devps directory which groff
uses.  One of the files I tried to print attempted to use a couple of fonts
that it couldn't find (I think C and CD) so I cp'd the CR (courier) font
file and edited the name entry for C & CD --- I don't know enough about troff
to tell what negative effects this may have on output.  Also I edited the
groff shell script to allow for a -Tpsc argument (is this hardwired into ATK?)
Unfortunately, I haven't been able to test ZIP cause it keeps dumping core
(patchlevel 5) and TABLE works but the test I ran printed the inset EZ cell
properly but all the other cells were printed in very small print. (ie TINY -
but barely readable)  AND the boxes I drew around the cells had lines sticking
up at the tl & tr corners.  I don't know it this is groffs fault or Andrews.
I haven't yet attempted to test EZ with insets (other than ZIP which dumped)
so I can't report on any problems that might exist with using groff on such
a file.  Nor have I tested printing from any of the other programs.

For reference sake, my machine is a Vaxstation 3200 with Ultrix 3.2 (X11.3)

Hope someone finds this info interesting and useful.  Getting groff to work
was not TOO hard but it did require a lot more effort than it should have.

Thanks,

John Hansen
madis!hansen@lognet2.af.mil