Re: [Groff] Heirloom TBL problem

[Mike Bianchi]

On Sun, Jun 21, 2015 at 08:35:34PM -0500, Blake McBride wrote:
> On Sat, Jun 20, 2015 at 3:52 PM, <address@hidden> wrote:
> > The interface to .ll is \nW or .PGFORM.  At first my plan was to implement
> > .PGFORM.  But *maybe* using W like MS's LL could also make sense.  But for
> > compatibility with groff .PGFORM should be prefered. (?)
> 
> Adding .PGFORM is fine, but I would prefer just having .ll work like it
> does on groff.   This way the original docs work and produce as expected.


Asking for  .ll  to work like it does in groff is an oximoron.  The commands
documented in  groff(7)  are the assembly language of groff.  They are the
commands on which all extensions to groff are built.

So when you use the MM or MS macros, you are using extension macros that are
built on the groff commands.  Likewise, programs like tbl(1), eqn(1), pic(1),
etc. process the input and emit the input combined with more groff commands
that groff then processes to create the desired outcomes.

Most macro packages, and certainly MM and MS, have preferred ways of specifying
line length that ultimately emit  .ll  commands to implement the desired
results.

Using the groff commands within other macro packages often produces confusing
and unexpected results.

So the recommended practice is to stick strictly to the "higher-level" macros
of the macro package, or packages, you are using.

Mixing the package macros with groff macros is discouraged, except to those
who imagine themselves to be experts in groff _and_ the macro packages.
I am one such person, and have many sad tails to tell of my ignorance.

-- 
 Mike Bianchi
 Foveal Systems

 973 822-2085

 address@hidden
 http://www.AutoAuditorium.com
 http://www.FovealMounts.com