Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!petrus!bellcore!decvax!decwrl!pyramid!hplabs!sdcrdcf!ucla-cs!dberry
From: dberry@ucla-cs.UUCP
Newsgroups: net.text
Subject: problems with transparent output in ditroff diversions
Message-ID: <15520@ucla-cs.ARPA>
Date: Wed, 11-Jun-86 12:24:55 EDT
Article-I.D.: ucla-cs.15520
Posted: Wed Jun 11 12:24:55 1986
Date-Received: Sun, 15-Jun-86 07:17:44 EDT
Reply-To: dberry@ucla-cs.UUCP (Dan Berry)
Distribution: net
Organization: UCLA Computer Science Dept.
Lines: 36

This is a problem that has a student (Martine Schlag) and I totally
stumped.
It seems hard to get ditroff to not obey transparent output immediately
as opposed to in its relative position among input
When we give:
------------------------
.pp
To obtain a layout we must have the exact dimensions of the elements
in the IF.  This information must be provided by the designer. 
The following is the layout of the \f3mxor\fP example used before.
.fs
\&
.sp 4i
.in +2i
\ \"unpaddable blank
.br
\!\!\!&martine
.in -2i
.ce
Figure mxor8 Layout of the mxor.
.fe
.pp
And now we continue.
----------------------------------
to dtroff (here .fs is the start of a floating keep macro (using the
standard two diversion technique and .fe is its ender) we see from the
dtroff output that &martine\ shows up BEFORE the movements corresponding
to .sp 4i and .in +2i!!!!????
I recall seeing a suggestion to use .fl but that does not seem to work..
Does anyone know what to do?????
Thanks
Dan and Martine
martine@neptune.cs.ucla.edu
-- 
Daniel M. Berry         UCLA Computer Science Department
dberry@ucla-locus.arpa  ..!{sdcrdcf,ihnp4,cepu,trwspp,ucbvax}!ucla-cs!dberry