groff
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: tbl: off-by-one line error at table ends with grotty?


From: Ingo Schwarze
Subject: Re: tbl: off-by-one line error at table ends with grotty?
Date: Sat, 25 Jul 2020 13:07:08 +0200
User-agent: Mutt/1.12.2 (2019-09-21)

Hi Branden,

this is not specific to -Tascii/-Tutf8 output.  Even in -Tps/-Tpdf
output, observe that the vertical distance from the first line in
the table to the second line in the table is the same as the vertical
distance from the second line in the table to the first line after
the table.  That does not depend on the presence or absence of a
box at all.  Only, in PostScript output, the box is close enough
to the preceding text that it doesn't overlap the subsequent text,
but it almost touches it there, too.  So .TE immediately followed
by a text line does not look good in any output mode.

This behaviour has been around long enough that by now, i think
almost all authors have become used to inserting .PP, .Pp, or the
equivalent in whatever macro set they are using, or some .sp if
they are using no macro set, right after .TE.  It looks better
either way, even without a box, and makes more sense logically,
too.  Why would you want a table *inside* a paragraph rather than
starting a new paragraph after a table?

In terminal output, the logical consequence is that it necessarily
looks odd without a paragraph break because the box can't possibly
be closer to the table content than one full line down, so it is
not surprising that it ends up on the same line as the subsequent
text, unless there is a paragraph break.

Admittedly, it does look like a quirk until you think about it,
and i did not reproduce that quirk in mandoc.  Even without the
usual paragraph break, mandoc renders:

       |with some stuff in it. |
       +-----------------------+
       after-table

But that may be hard to reproduce in groff terminal output without
breaking groff typeset output, even if we wanted to, given that
both output variants need to be generated from the same intermediate
low-level output.

Yours,
  Ingo



reply via email to

[Prev in Thread] Current Thread [Next in Thread]