Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!rutgers!ucsd!ucsdhub!cuuxun!attu3g!ecf_csak
From: ecf_csak@attu3g.JHU.EDU (Steve Klenda)
Newsgroups: u3g.wwb,comp.text
Subject: Re: grap
Keywords: grap, dwb, troff
Message-ID: <175@attu3g.JHU.EDU>
Date: 15 Oct 88 18:51:21 GMT
References: <168@attu3g.JHU.EDU>
Reply-To: ecf_csak@attu3g.UUCP (Steve Klenda)
Organization: The Johns Hopkins University
Lines: 23

In article <168@attu3g.JHU.EDU> ecf_csak@attu3g.JHU.EDU (Steve Klenda) writes:
>Each of the graphs printed properly by themselves, 
>but when they were included in the text only the first
>graph printed correctly.  The second, and all subsequent graphs were 
>badly mangled...

Thanks to Eric Wilson of Bell Communications Research, who pointed out 
the problem.  Each graph was followed by a key which was created with PIC.  
Since it was easier to work in points than in inches, the "scale=" command
was used to change the default scale.  This carried over to 
subsequent GRAP input.  Since GRAP generates distances based upon the
defualt scale, which had been changed by PIC, the output was mangled.

Placing a

.PS
scale=1;
.PE

before each grap .G1 solved the problem.  Of course if GRAP reset the
scale to the defualt, as it should, then this fix wouldn't be
necessary.

Steve Klenda