Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!linus!philabs!cmcl2!harvard!husc6!panda!genrad!decvax!decwrl!glacier!hplabs!hao!nbires!rcd
From: rcd@nbires.UUCP
Newsgroups: net.text
Subject: Re: Orphans & Widows Fix for Troff Wanted
Message-ID: <753@nbires.UUCP>
Date: Fri, 9-May-86 01:24:51 EDT
Article-I.D.: nbires.753
Posted: Fri May  9 01:24:51 1986
Date-Received: Wed, 14-May-86 01:06:55 EDT
References: <312@garth.UUCP> <2331@phri.UUCP>
Distribution: net
Organization: NBI,Inc, Boulder CO
Lines: 25

> > I'm interested in any hacks or purchaseable software that fixes
> > the problems of Orphans and Widows in Troff.
>...	We fixed widow lines by imbedding a ".ne 3" request in the
> paragraph macro (I think -me has this by default).

This only helps sometimes.  If you have a footnote in the first line of the
paragraph, that can push the end-of-page trap up enough that you still end
up with a widow.  It's hard to do much better with troff, or any formatter
where you've got a "single pass" sort of approach.

> 	Orphan lines are harder to deal with; I could see a scheme whereby
> you divert output in the start-paragraph macro, and have an end-paragraph
> macro that sees how long the paragraph is, and how much is left on the page
> and inserts a .bp at the right place.  Yuck!

And the same sort of scheme would work for the footnote-induced widow
problem that I mentioned, if you could stomach writing the macros and
ensuring that nothing gets screwed up when the diversion is re-read, etc.

The use of diversions gives you a second "pass" of limited range.  The
really painful part (ref. "Yuck!" above) is having to process all the text
twice just to get the page breaks right.
-- 
Dick Dunn	{hao,ucbvax,allegra}!nbires!rcd		(303)444-5710 x3086
   ...Cerebus for dictator!