Re: [Groff] Ragged Bottoms

[Peter Schaffter]

On Tue, Nov 25, 2003, Greg Pavelcak wrote:
> I've been trying to write some two-pass macros that measure the
> difference between the bottom line and the desired bottom on the first
> pass and applies the adjustments to the flexible spaces in the document
> on the second. It really seems like thi should be easy, but I haven't
> been having much luck. Has anyone else tried to automate ragged bottom
> elimination in this or any other way? If you have, I'd love to hear
> about your results.

While it isn't the precise solution to your problem, you might want
to study om.tmac, particularly the macro TRAPS, which is where
document leading and the bottom margin are worked out.  Very simply,
TRAPS compares the lead supplied by the user with the bottom margin
set by the user, counts how many lines actually fit on the page at
the given lead, then calculates how many machine units (1/1000 pt
for PostScript, which is what om.tmac's geared toward) to add to the
overall leading so that the last line on the page falls precisely on
the established bottom margin.

Another place to look for ideas is the SHIM macro, which puts the
leading "back on track" if a user strays from it.  It does this
simply by subtracting where one is on the current page from the next
legal place one needs to be to ensure flush bottom margins, and
advances the difference.

In a two-pass scenario, it should be possible to spring a trap
near the bottom of the page, one that uses the number of times the
flexible space macro was invoked divided into whatever space is
left over, then outputs the first pass diversion with the appropriate
space being given to .sp as an argument.

Applying Occam's razor at its most trenchant:

.de space
.if \\n[#count]=0 \{\
.   nr #count 0 1
.\}
\\n+[#count]
.sp \\n[#flexible_space]u \"or whatever unit of measure you prefer
..

When you spring the trap near (hopefully short!) of the bottom
margin, you invoke a macro that calculates #flexible_space by
figuring out how much space is left over and dividing it by #count.
This gives you #flexible_space, which, on the second "pass," is the
value given to .sp.  Naturally, you have to remember to reset
#count and #flexible_space for each page.

Don't take this as gospel.  It's just a quick "sketch" to point you
in one possible direction.  My example, for instance, assumes
you're running a version of groff that understands .sp 0.

-- 
Peter Schaffter

Author of _The Schumann Proof_, appearing fall, 2004
(pub. RendezVous Press, Canada)