Re: [docbook-apps] Escape DTD snippet

[Karl Eichwalder]

Susanne Wunsch  writes:

> I searched for a possibility to quote DTD snippets that aren't parsed
> but integrated "as is". I didn't find some appropriate hints.
>
>   

Use CDATA marked sectionsv (this is a general SGML/XML feature):



Your example:

 

-- 
Karl Eichwalder

-
To unsubscribe, e-mail: docbook-apps-unsubscr...@lists.oasis-open.org
For additional commands, e-mail: docbook-apps-h...@lists.oasis-open.org

DOCBOOK-APPS: Re: writing a condensed form of docbook

[Karl Eichwalder]

Markus Spath <[EMAIL PROTECTED]> writes:

> Robert P. J. Day wrote:
>>   i like to write my documents and manuals in condensed
>> form, and it would be nice to quickly to go from that to final docbook.
>> 
>
> Since you stay with the xml-syntax anyway

This sounds like a arguable requirement...  If you are interested in
minimization look for SGML minimization features!

Of course, it's also possible to "rename" elements and process the file
using DocBook stylesheets nevertheless.  Just do Architectural Forms (AF)
processing -- this will even work for XML files.

> I'd say XSLT would be a convenient way;

That's a possbility, too.  Or use the Transformation feature defined by
DSSSL.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.gnu.franken.de/ke/|  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: comments on "Using the DocBook XSL Stylesheets"

[Karl Eichwalder]

Daniel Veillard <[EMAIL PROTECTED]> writes:

> I can't do any automated catalog management

I don't believe in this kind of magic ;-)

Question: where shall we install the TEI P4 DTD?  One and the same is
used for SGML and XML (the same is basically true for DocBook and the
DocBook DSSSL stylesheets).  If you are interested in the XML flavor of
the TEI DTD just say in your internal Doctype declaration:



Okay, we can try to install the same files twice or provide
sophisticated link farms--and that's what I already proposed the last
time dicussed this mess.

As long as there is no written and approved standard (part of the LSB)
I'm not going to change directory names under /usr/share/sgml -- I once
did on account of the LSB SGML appendix and then it disappeared ;-(

When a concise description is part of the LSB I'll support it, of
course.  Marc Johnson(?) said he wanted to work on this issue.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.gnu.franken.de/ke/|  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: comments on "Using the DocBook XSL Stylesheets"

[Karl Eichwalder]

Adam DiCarlo <[EMAIL PROTECTED]> writes:

> Ideally the file locations should comply with the LSB/XML standard,
> although I can no longer find that standard (!! maybe it's dead?).

The SGML "standard" (better: proposal or recommendation) once was added
to the LSB (around version 1.0) as an appendix; thus I think it was non
mandatory.  It didn't took a long time and it was dropped again--I
don't know the reason why.

Then "we" discussed it again.  Unfortunately, the last discussion
happened at the wrong moment: before the XML Entity Resolution
standard/proposal (AKA: XML Catalog Standard) were released.  People
became unpatient and no format decision were taken.

Since the SGML "standard" didn't solve a single problem there's no need
to complain ;)

Now we have two full featured tools: libxml2/libxslt (plus helper tools)
and openjade (with DTDDECL support!); both are able to access resources
thru catalogs.  No need for a standard any longer.  We just will have to
make sure proper master catalogs are available:

/etc/sgml/catalog
/etc/xml/catalog

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.gnu.franken.de/ke/|  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: comments on "Using the DocBook XSL Stylesheets"

[Karl Eichwalder]

Adam DiCarlo <[EMAIL PROTECTED]> writes:

> No. It's is because I always get it right. :) If you tried to load the
> SGML ISO entities from and XML doc, that means you're using the wrong
> public identifier.

Yes.

There is a new problem: The TEI group use the same identifiers for SGML
and XML ;-(  I already complained--thus far, nobody answered.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.gnu.franken.de/ke/|  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: XInclude doesn't validate with xmllint

[Karl Eichwalder]

Daniel Veillard <[EMAIL PROTECTED]> writes:

>   You should really start accepting that URL can be used as identifiers
> like any other string, sigh ...

It can, but a SYSTEM identifiers is not canonical (= it is not fixed).
Let's hope somebody will finish the LSB some time soon regarding XML.

Scrollkeeper seems to prefer the SYSTEM identifier, though.

All in all, XML is nicely complex (and limited nevertheless ;) ).

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: XInclude doesn't validate with xmllint

[Karl Eichwalder]

Peter Ring <[EMAIL PROTECTED]> writes:

> First I'd like to lobby for keeping your local copies of DTDs and
> stylesheets as a mirror or cache of the canonical URLs.

There is no such thing like a canonical URL.  Use a catalog and map
PUBLIC identifiers to local resources.

Note, PSGML implements a concept how to map PULIC identifiers to local
resources and to read it from there without a catalog.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: Uniq values of "ID" attributes

[Karl Eichwalder]

Dave Pawson <[EMAIL PROTECTED]> writes:

>>Probably.  Or just add a dummy chapter with all missing IDs to the
>>respective document (cf. David Megginson's book on XML).

> Its presently my bedside table reading.. chapter reference please,
> I haven't got to it.

I'm talking about "Structuring XML Documents" (1998).  Chap. 7
"Exchanging Document Fragments".


> I'm curious about the Architectural forms approach though.
>   Eliot Kimber had a ruminant moment on id idrefs on the xslt list

I'll have to lookup the archive (date and URL please ;) ).  Within
Megginson's book you can read Part 4 for a good introduction on AF.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: Uniq values of "ID" attributes

[Karl Eichwalder]

[I'm escaping to the apps list]

Vitaly Ostanin <[EMAIL PROTECTED]> writes:

>>>Yes, but for separated documents we must to use:
>>>id="full.path.with.sub.parts.of.set.document-name.pgs.ex.1"
>>>for it be unique.
>> Not if you make the id's relate to the document subject?
>> id values do not need the document path.
>
> For example, we may have chapters "network" in "user" and "admin" parts.

Delay the issue until combining the docs.

Then prefix every ID in "user" with "u." and in "admin" with "a.",
say.  This could be done using Architectural Forms processing or using
sed/awk.

> You right, olink is the best way.

Probably.  Or just add a dummy chapter with all missing IDs to the
respective document (cf. David Megginson's book on XML).

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: Announce: DocBook XSL Stylesheets V1.50.0

[Karl Eichwalder]

Camille Bégnis <[EMAIL PROTECTED]> writes:

> Up to now, each new release of a DSSSL or XSL stylesheets had its
> corresponding RPM package but with the same name (docbook-style-dsssl
> resp. docbook-style-xsl). Therefore installing a new version would
> overwrite the previous one.

Not necessarily.  Use -i instead of -U to install the package.

> Another policy, based on new version numbering scheme, is to suffix the
> name with the version number minus the last dot (i.e.
> docbook-style-xsl-1.50) so that different versions with different public
> APIs could coexist on a single system.

The decision (LSB) was to install every version into a separate
directory.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: Editors and IDREFs

[Karl Eichwalder]

Markus Hoenicka <[EMAIL PROTECTED]> writes:

> have a look at RefDB (http://refdb.sourceforge.net).

I have two "concerns" about refdb:

. You're forced to use one and only database system (unfortunately
not PostgreSQL).  There are generic frameworks like
libgda/GNOME-DB, and there are also Python or Perl solutions.

. I didn't find a description how the database tables are
designed.  And queries seem to be hidden in C code only.
Probably I didn't try hard enough.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: "Abbreviated" DocBook

[Karl Eichwalder]

Kevin Atkinson <[EMAIL PROTECTED]> writes:

> DocBook is too dam verbose, providing an entry for just about
> everything under the sun.

Try to enable SGML minimization features; but it looks as if you want
to write an m4-to-xml macro system :)

> Although you may disagree with me, it is so verbose that just reading
> a DocBook source document becomes rather difficult.

Emacs/psgml.el is able to hide tags and using xxml.el nifty colorings
are possible.

> At least to me the former is far easier to read and to enter in than
> the latter.  So do you guys know of any utilities.  I looked all over the
> net, but came up empty.  If none exists it is highly likely that I whip up
> something myself.  I honestly would like to know how people can stand
> editing raw DocBook.

It depends ;)  Consider to use Texinfo and convert your Texinfo files
to DocBook using makeinfo from the latest pre-release of the texinfo
package (don't fear to report bugs or other shortcomings!).

> Even with emacs help it is still a bitch to both read and edit.

Some tempo.el templates can do wonders; for reading convert the document
to HTML and use your favorite browser for online reading.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: [szucker@mac.com: Re: docbook.dcl related errorswith openjade]

[Karl Eichwalder]

Sasha Zucker <[EMAIL PROTECTED]> writes:

> CATALOG "/sw/share/sgml/dsssl/docbook-dsssl-nwalsh/catalog"
> CATALOG "/sw/share/sgml/dtd/docbook/catalog"
> CATALOG "/sw/share/sgml/entities/iso8879/catalog"
> CATALOG "/sw/share/sgml/openjade-1.3/dsssl/catalog"
> CATALOG "/sw/share/xml/dtd/docbookx/catalog"

As long as your software doesn't support DELEGATE you must not mix
catalogs; esp. don't mix non-XML and XML catalogs.  Without supplying a
proper SGMLDECL for every flavor you are not able to verify your
documents properly -- maybe, verfication isn't important to you, though.

> On the bright side, the package finally installs correctly. All I have
> to do post install is comment out the DTDDECL directives.

Since these are just warnings you can leave them in (and better patch
(o)nsgmls instead).

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: Catalog dor DocBook and URI for the XSL stylesheets

[Karl Eichwalder]

Kevin Conder <[EMAIL PROTECTED]> writes:

>   Please consider using another site. Sourceforge uses a strange 
> port-redirection scheme that doesn't work well with corporate
> firewalls.

Thus far I didn't encounter this problem.  Another reason: the docbook
community doesn't "own" the proposed sub-domain.  Why not using
docbook.org or oasis-open.org?  Okay, these domain are not in the public
domain either, but I looks much more safe to me to choose one of those.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)


To subscribe or unsubscribe from this elist use the subscription
manager: 

DOCBOOK-APPS: Re: Solution to funky TOCs in HTML Help...

[Karl Eichwalder]

"Bradford, Denis" <[EMAIL PROTECTED]> writes:

> According to my XSLT book, XSLT has lots of ways to control the
> generation of white-space nodes - elements like space, preserve-space,
> and strip-space elements - in contrast to HTML where the spec is not
> very precise,

Rules are precise but "difficult"; HTML's SGML declarations defines how
to treat whitespace.

> and so different browsers handle whitespace differently.

That's the truth. And to some degree it's their right to treat
whitespace differently; often browsers get whitespace handling wrong
('cos HTML is often wrong ('cos the browsers are wrong)).  The way out
is to use xhtml and to tell browser vendors to fix their products.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

--
To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: [EMAIL PROTECTED]

DOCBOOK-APPS: Re: db2latex

[Karl Eichwalder]

"Sebastian Rahtz" <[EMAIL PROTECTED]> writes:

> I have been around this world for a good few years, and I have only
> once met anyone who used architectural forms. Unless you can show me
> otherwise, I put it to you that AFs are pure theory.

I tried it along David Megginson's book 'Structuring XML Documents'
using nsgmls and it works.  I don't want to argue whether there are some
bugs left or not.  IIRC, a Python based implementation exits, too --
maybe, imcomplete (I didn't check since nsgmls worked).

> Someone will mention shortly how HyTime will solve everything. 

Feel free to ignore it (and use XLink, the reinvented wheel) ;)

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

--
To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: [EMAIL PROTECTED]

DOCBOOK-APPS: Re: db2latex

[Karl Eichwalder]

Michel Goossens <[EMAIL PROTECTED]> writes:

> DC promised to have a look at it a few weeks back... I agree that for
> a general DTD writing the TeX macro code necessary to support it might
> not be a good investment.

Note, such an approach is _never_ required.  You can use Architectural
Forms Processing to "map" the elements and attributes; cf. David
Megginsons' Book on XML -- it covers SGML, too :)

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

--
To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: [EMAIL PROTECTED]

DOCBOOK-APPS: Re: Jadetex and Turkish

[Karl Eichwalder]

Togan Muftuoglu <[EMAIL PROTECTED]> writes:

> db2tex -j "-b iso-8859-9" -s
> /usr/share/sgml/docbook-toys/suse-both.dsl ornek.sgml

Please, send me a small example file ([EMAIL PROTECTED]).

Jorge Luiz Godoy Filho <[EMAIL PROTECTED]> writes:

> I think that jadetex 2.20 isn't maintained anymore...
> 
> Could you please try with Jadetex 3.10? It's available at
> http://jadetex.sourceforge.net 
> 
> If the problem isnt' solved, please, try upgrading your DSSSL
> stylesheets to 1.71 (http://docbook.sourceforge.net).
> 
> If nothing works, please come back with these new versions. These
> upgrades will make it easier to diagnose the problem.

I don't thinks the user is considered such an update on his own (even if
it doesn't look too difficult to do so)...  But if there's interest I'll
make RPMs for evaluation available; warning: the packages might work on
SuSE Linux (version 7.2) only.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

--
To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: [EMAIL PROTECTED]

DOCBOOK-APPS: Re: Converting from DocBook/SGML to DocBook/XML

[Karl Eichwalder]

Dan York <[EMAIL PROTECTED]> writes:

> H interesting command.  Definitely looks like it could have helped
> me. No, I didn't know about it.  The only thing is that the one on my
> system (Red Hat Linux 6.2) converts all tags to UPPERCASE, which is exactly
> NOT what I want to do.  Perhaps there is a newer version somewhere.

>From the very beginnings of sp's XML support there is sx (often
installed as sgml2xml or s2x), James Clark at his best!  sx knows about
the switch -xlower (output option).  Use sgmlnorm and then run the
resulting file through sx.

-- 
[EMAIL PROTECTED] (work) / [EMAIL PROTECTED] (home):  |
http://www.suse.de/~ke/  |  ,__o
Free Translation Project:|_-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/ |   (*)/'(*)

DOCBOOK-APPS: Re: Non SGML Character Number?

[Karl Eichwalder]

Holger Rauch <[EMAIL PROTECTED]> writes:

> Ok, but I've never had any problems so far.

As long as you're using Emacs it doesn't matter that much; PSGML assumes
Reference Concrete Syntax without limits.  PSGML has it's own way to
provide the correct defaults for XML (hardcoded).

Using nsgmls the first SGML declaration will win (if no declaration is
set parser defaults are used); if you're lucky it will work.

-- 
work : [EMAIL PROTECTED]  |   ,__o
 : http://www.suse.de/~ke/ | _-\_<,
home : [EMAIL PROTECTED] |(*)/'(*)

DOCBOOK-APPS: Re: Non SGML Character Number?

[Karl Eichwalder]

Holger Rauch <[EMAIL PROTECTED]> writes:

> My SGML_CATALOG_FILES variable contains (the newline characters are just
> here for readability):

Don't use SGML_CATALOG_FILES.  Using SGML_CATALOG_FILES without knowing
what you do is asking for trouble :)

> /usr/share/sgml/CATALOG.docbk30:/usr/share/sgml/CATALOG.docbk31:
> /usr/share/sgml/CATALOG.gnome:/usr/share/sgml/CATALOG.iso-ent:
> /usr/share/sgml/CATALOG.tei-3:/usr/share/sgml/CATALOG.tei-3.system:
> /usr/share/sgml/CATALOG.teixlite:/usr/share/sgml/jade_dsl/catalog:
> /usr/share/sgml/openjade/catalog:/usr/local/docbook/dtds/sgml/4.1/docbook.cat:
> /usr/local/docbook/dtds/xml/4.1.2/docbook.cat:
> /usr/local/docbook/style/dsssl/1.70/catalog
> 
> What's going on here?

Don't mix CATALOG files.  Different DTDs use different SGML
declarations.  Esp. don't mix DocBook 3.0 with DocBook 4.1 XML.

TEI and DocBook 3.0 are using different SGML declarations, too.  Don't
mix DocBook catalogs with TEI catalogs.

Intentionally, SGML_CATALOG_FILES isn't set on SuSE Linux.

-- 
work : [EMAIL PROTECTED]  |   ,__o
 : http://www.suse.de/~ke/ | _-\_<,
home : [EMAIL PROTECTED] |(*)/'(*)

DOCBOOK-APPS: Re: What are the steps to convert to DocBook XML?

[Karl Eichwalder]

Sebastian Rahtz <[EMAIL PROTECTED]> writes:

> Jade can happily process XML files. switching to XML is a win every
> time, as you can use all the old tools, *and* the new ones

...and you're force to code more verbose: you'll have to quote
attributes, you'll loose some minimazation features and -- most
important -- the marked section feature is more or less gone.

For authoring purposes XML is a step backward.  To use so called XML
tools it's always possible to do an sgml2xml on the fly.

-- 
work : [EMAIL PROTECTED]  |   ,__o
 : http://www.suse.de/~ke/ | _-\_<,
home : [EMAIL PROTECTED] |(*)/'(*)

--
To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: [EMAIL PROTECTED]

DOCBOOK-APPS: Re: PDF and single file HTML output

[Karl Eichwalder]

Bernd Kreimeier <[EMAIL PROTECTED]> writes:

> db2html has no --help, but there is a comment in the
> script:

Did you try db2xxx on SuSE Linux?  ;^)

>  # it's best to make a new directory with several html files in it
>  #cat $* | jade -d $DB_STYLESHEET -t sgml -V nochunks > $TMPFN

tux:~ # db2html --help | grep -1 ^HTML

HTML:
  --nochunks  don't produce HTML subdocuments; output to stdout

> Can I trivially reactivate this line? Is there a db2shtml?

docbook-toys coming with SuSE Linux are considered to work on other
pattforms, too; please, ask google.  I do accept bug reports (and
patches).  Reminds me to upload version 0.14 or better next monday:
http://www.suse.de/~ke/docbook-tools/

> This is why I don't like the db2* scripts, they hide too much from the
> end user, and (IMHO) they're not sufficiently customisable.

Not necessarily...  I rewrote them on users request for SL --
personally, I do like the Makefile approach more ;)

-- 
work : [EMAIL PROTECTED]  |   ,__o
 : http://www.suse.de/~ke/ | _-\_<,
home : [EMAIL PROTECTED] |(*)/'(*)

DOCBOOK-APPS: Re: Modifying parameters in the DocBook stylesheets

[Karl Eichwalder]

Norman Walsh <[EMAIL PROTECTED]> writes:

> [Follow-ups to docbook-apps, please]

;-)

> / "Juan R. Migoya" <[EMAIL PROTECTED]> was heard to say:
> | You can't avoid to modify something if you want to customize
> | your stylesheet. Other thing is that you might avoid to modify
> | the original dsl files, other than the main one.

Shameless plug: The db2x.sh script (to be called as db2html, db2pdf,
etc.) coming with SuSE Linux offers some hooks to do customizations from
the command line.

> No, you don't have to modify the original stylesheets at all.
> See http://nwalsh.com/docbook/dsssl/doc/custom.html

Yes, thanks for all your work!

-- 
work : [EMAIL PROTECTED]  |   ,__o
 : http://www.suse.de/~ke/ | _-\_<,
home : [EMAIL PROTECTED] |(*)/'(*)

--
To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: [EMAIL PROTECTED]

DOCBOOK-APPS: Re: Simple example to generate a TOC in a DVI and aPDFfile

[Karl Eichwalder]

Jirka Kosek <[EMAIL PROTECTED]> writes:

> 2.
> - Get LaTeX source from DocBook and include hyperref package. 
> - process it with pdftex -> you will get PDF file with bookmarks

[...]

> Ad 2. If you want to get LaTeX source from DocBook, there is
> experimental stylesheet 
> http://www.infres.enst.fr/~casellas/docbook.html
> Stylesheet is written in XSLT, so you would need to convert your SGML
> file to XML before processing. Including hyperref package and processing
> with pdflatex should produce desired output (but I personally did not
> tested LaTeX stylesheets and whole processes).

Around Xmas holidays I tried Norman Gray's patch to openjade with good
success http://www.astro.gla.ac.uk/users/norman/distrib/jade.html; after
some cleanup I can make available latex transformation stylesheet

This way you can stay away from XSLT and use DSSSL.

-- 
work : [EMAIL PROTECTED]  |   ,__o
 : http://www.suse.de/~ke/ | _-\_<,
home : [EMAIL PROTECTED] |(*)/'(*)

DOCBOOK-APPS: Re: Problem getting backslashes

[Karl Eichwalder]

Les Schaefer <[EMAIL PROTECTED]> writes:

> My environment:
> 
> SuSE Linux 7.0

[...]

> i.e. backslashes are produced as dashes from jade, which isn't quite
> what I need. Also the RTF output is also correct - i.e. the problem
> really seems to be restricted to the jadetex system (or its
> configuration).

Wrongly patched TeXFOTBuilder, I'd say.

> Can anyone help me to get my backslashes without using entities?

Thanks for the detailed report.  I guess you'll need the update packages
from http://www.suse.de/de/support/download/updates/70_i386.html:

ftp://ftp.suse.com/pub/suse/i386/update/7.0/sgm1/jade_dsl.rpm
ftp://ftp.suse.com/pub/suse/i386/update/7.0/sgm1/sp.rpm
ftp://ftp.suse.com/pub/suse/i386/update/7.0/sgm1/sp_libs.rpm

It doesn't hurt also to update
ftp://ftp.suse.com/pub/suse/i386/update/7.0/sgm1/iso_ent.rpm 

To fix the '—' vs. '--' (= '-\/-') problem I introduced another
bug :-(  Sorry for inconveniences.

-- 
work : [EMAIL PROTECTED]  |   ,__o
 : http://www.suse.de/~ke/ | _-\_<,
home : [EMAIL PROTECTED] |(*)/'(*)