Re: [Groff] Flexible whitespace, columns, and the .h register

[Peter Schaffter]

On Wed, Jan 04, 2017, Ralph Corderoy wrote:
> > On re-reading cstr54, I see .h is defined as the "Text baseline
> > high-water mark on current page or diversion," which is somewhat
> > ambiguous
> 
> How so?

"Somewhat" is the key word, above, not "ambiguous", IYKWIM.
Probably just my way of reading English.  Either "...high water mark
*of* the current page or diversion" or "...on the current page or in
the current diversion" are what make it clear as a bell to my eyes.

> > The issue doesn't affect my problem, but should we consider amending
> > the info docs entry?  "...corresponds to the text baseline of the
> > lowest line on the page, or, if in a diversion, of the lowest text
> > baseline in the diversion, measured from the start of the
> > diversion."  Words to that effect, anyway.
> 
> I think the info doc should mention diversion in "...on the page".  Is
> the measured bit needed?  Isn't that the origin of all diversions
> measurements?

Quite so.  I was writing off the top of my head. :)  

> Does .h being diversion-specific make it easy for you to work around .h
> being read-only?

No.  There would insurmountable problems with nested diversions
containing material that may have to be deferred to the next
column/page (floats).

> Is there a problem with making it writable?

Of that, I'm not sure.  Someone more familiar with the code might be
able to say quickly.

> Are there other things that it might be nice to be able to reset
> with some new command as if we were hovering above the start of
> this virgin page?

.trunc.  If, in a footer macro, one wants to find out whether an
.sp tripped the footer, '.if \n[.trunc]' is ineffective because it
can't be zeroed at the start of a new page/column.

-- 
Peter Schaffter
http://www.schaffter.ca