Re: [Scrollkeeper-devel] db2omf

[David M. <da...@lu...>]

On Thu, Mar 08, 2001 at 02:33:35PM -0800, Dan Mueth wrote:
> 
> On Thu, 8 Mar 2001, David Merrill wrote:
> 
> > On Thu, Mar 08, 2001 at 01:22:06PM -0800, Dan Mueth wrote:
> > > 
> > > 
> > > On Thu, 8 Mar 2001, David Merrill wrote:
> > > 
> > > > Has anyone done any work toward creating omf from docbook sgml/xml
> > > > sources? I want to do some testing of LDP packaging for use with
> > > > scrollkeeper. I would like to not have to maintain omf separately from
> > > > our docbook if that is possible.
> > > > 
> > > > I'd be happy to help expand the sk documentation as I go along.
> > > > 
> > > > Any other tips gratefully accepted!
> > > 
> > > Dave,
> > > 
> > > Nobody has started work on this yet and this would be a valuable addition
> > > to scrollkeeper.  This is especially true since most documentation is in
> > > DocBook.  There are two problems that come to mind:
> > > 1) DocBook does not yet support all of the OMF tags
> > > 2) The OMF DTD will be changing somewhat, since it is not in
> > > complete agreement with the OMF specification and because we are
> > > discussing adding a couple new elements or attributes.
> > > 
> > > However, both of these are very solvable problems, and really need to be
> > > solved.  Norman Walsh has agreed to help get the necessary tags into
> > > DocBook.  And fixing the DTD is a matter of somebody (Gregory Leblanc I
> > > think is the maintainer of the DTD) putting in some time and resolving the
> > > outstanding issues on the ldp-meta mailing list.
> > > 
> > > If you'd like to get started anyway, realizing that things will be in a
> > > state of flux for a little while, that would be great.  In the short term,
> > > people could run it manually to generate partially completed OMF file
> > > templates and then manually fill in the missing bits.  In the longer term,
> > > we can have scrollkeeper-preinstall run db2omf on DocBook files so that
> > > the OMF files are not created until build time.
> > > 
> > > BTW: I wasn't sure what you meant by "testing of LDP packaging", but I
> > > have almost finished a new example package which illustrates one of the
> > > simpler ways to set up a package to use scrollkeeper and handle DocBook
> > > properly, and it provides templates and some documentation to make things
> > > easier.  I'm hoping to release it today.
> > 
> > My immediate goal is to establish docbook guidelines for our authors,
> > so that their docbook is omf-ready, and to position us to fully
> > support scrollkeeper. I keep the LDP up to date with what sk is doing,
> > or at least as up to date as *I* am.
> 
> This is good to hear.
> 
> > An LDP volunteer is going to work on a db2omf script, and I will send
> > it to you when it is ready. I asked him to license GPL so you can use
> > it without license issues.
> 
> Thanks.  We have licensed scrollkeeper under the LGPL because it will
> become a proper library.  Since it makes sense to have db2omf be a part of
> scrollkeeper so that the OMF is generated from the DocBook at install
> time, we would need to license it under the LGPL.  Would this be okay with
> you?

Sure.

> > If scrollkeeper is going to generate omf from db directly, then does
> > that mean the LDP won't need to generate omf for you? That was the
> > basis of my thinking, and if that goes away then we don't really need
> > to do anything extra. But we'll still work on the db2omf script.
> 
> I think it will be a while until we get to this step.  First we have to
> finalize the OMF DTD and then we have to get it accepted into DocBook and
> wait for a DocBook release to incorporate it.  I suspect the DocBook
> process will take a long time.
> 
> I'm hoping to have the OMF DTD figured out within the next month or two
> and then start the DocBook process.

For the time being, we will create the script for generating omf from
docbook as it stands now - db3.1 or db4.x sgml or xml. When docbook
gets updated to support omf natively, we can deal with it. When sk
goes beta, I want to have the LDP documentation included. We've done a
lot of work, as have the other documentation projects, to create the
content, but the delivery right now sucks. We hope sk will change
that, and I personally consider it very important that you succeed.
Good luck!

-- 
Dr. David C. Merrill                     http://www.lupercalia.net
Linux Documentation Project                   da...@lu...
Collection Editor & Coordinator            http://www.linuxdoc.org
                                       Finger me for my public key

Postmen never die, they just lose their zip.