[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Groff] footer macro
Re: [Groff] footer macro
Tue, 16 Dec 2003 17:29:51 +0100
> This is the groff list; people don't get annoyed. :-)
Am Montag, 15. Dezember 2003 13:22 schrieb Ralph Corderoy:
> Hi Erich,
> > The code burns down to this:
> > .de fo
> > 'br
> > ..
> I found putting a 'bp instead of 'br began the next page when at -1i.
> I think the one word of output appears because of the 'br you're doing.
Uh, ahem, emm-hemm-rhmm...
I have to be a bit verbose.
The ['br] was a silly typo, the original source code had an ortodox
['bp] request. That was caused by a certain nervosity, because I did
not see how the [.fo .'br ..] macro could produce the effect I had,
and all tests inside a raw groff code didn't show up with it.
You could not possibly know that I have been experimenting with
a special tmac file. Some month ago I wanted to write a chess
documentation, got some help and even encouragement from the group,
and realized that before starting a project like that I should gather
more experience with groff first. The best way to do that was to
write a tmac file from scratch, I thought.
But I got lost in the file. It worked so far, but the longer it
became, the more problems I had with orientation. So I decided that I
needed a tool first and started writing a tmac file the output of which
was again a tmac file, this time with a certain structure inside of it.
Perhaps this idea makes more sense than it seems at first sight.
The output has two environments, the first is embraced in [.ig
end]-[.end] requests and contains comments, numbered headings, cross
references, three indexes and multiple tables of contents. The other
environment is a sort of verbatim environment in which the control
lines stay identical in the output. For this reason the escape &
control mechanisms have to be suspended.
As my output was an ascii file, I didn't have to bother with page
transition traps and I totally forgot that the ['] character is a
control character as well and has to be suspended! And *this* caused
the effect I showed you. I fixed that and it works now. (A tmac file
with such a load of comments is not very handy for practical use,
but the bare control lines are easily extracted with a single sed
command into a slim useable tmac file.)
My "ac.tmac" file, as i call it, is now in the testing phase and I'm
writing a real text producing tmac file with it - the first problem
I encountered is that of the falsified output of the 'bp request.
On the other hand I have the headings, tocs, cross references, indexes
and they help very much, and last not least I gathered my experience.
The ac.tmac seems to work. Ha, no one will believe me after this!
Thank you again, and cheers,