[Top][All Lists]

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

Re: [Groff] Introduction

From: Werner LEMBERG
Subject: Re: [Groff] Introduction
Date: Tue, 18 Oct 2005 22:43:28 +0200 (CEST)

> > I really would like to see the UTP improved, this is, all
> > references to dead features/programs should be removed, and the
> > new groff features should be incorporated as extensions.
> Yes, I know we talked about that...  It's just a question of time.
> There's a small number of you who really know the information...  Is
> there some efficient way that we could share the work, with those of
> us who are less knowlegeable picking up some of the scribal work?

There's a CVS repository for the UTP -- which isn't publicly available
currently for unfortunate reasons.  This should be moved into the
public again -- IIRC, Larry McVoy has offered this a longer time ago.

It's up to Larry Koller to proceed since he was (and hopefully still
is) the driving force behind the UTP project.

> > With other words, the groff_man(7) page and/or the corresponding
> > section in groff.texinfo isn't as nice as it should be, right?
> They are actually quite nice...  Perhaps we could add some
> command-line and simple formatting-definition examples as a quick
> fix?  Is that the way to go given everyones time constraints?

Can you send patches?  Ideally with `diff -u'...

> > Please discuss it here.  There have been plans to make groff emit
> > XML too but...
> Another one of those projects for when you have copious free time,
> right?

Not me, but Gaius :-)

> Actually, would there be a purpose for having groff emit XML over
> having a filter that converts groff source to XML source?  In other
> words, is there a purpose other than converting existing groff
> documents to XML?  The difficulty with a filter, of course, is that
> XML has to have the beginning and ending defined (say, for a
> section) whereas groff generally just defines where something
> begins...  If the groff is really well-structured, it's probably
> doable, but we all know that a lot of existing groff documents are
> not so well-structured...

A groff-to-xml translation basically suffers the same problems as
groff-to-html or rtf-to-latex: There can be a lot of low-level
operations with no equivalent high-level code.

If you follow strict rules while writing your groff documents, a
source-to-source translation might work, giving much better results.
For example, the `mdoc' macros for man pages, as used on FreeBSD and
other BSD flavours, are far easier to translate into different formats
than ordinary man pages written with `man': `man' is often too simple
to get typographically pleasing results, even on a modest level.


reply via email to

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