Re: DOCBOOK-APPS: Getting Rid of Character Entities in DocBook XML Docs

[Norman Walsh <ndw at nwalsh dot com>]

/ Holger Rauch <Holger.Rauch@heitec.de> was heard to say:
| > document, though. Or, at least, the parser has to read the external
| > subset.  Could that be the problem?
| 
| Yes, this seems to be the problem. Unfortunately, I can't think of an XSLT
| processor that automatically reads in the entity files.

I can't think of one that doesn't. I was sure that Xalan did a
validating parse by default.

| How does XML
| "know" about these files, by the way? AFAIK, these files are referenced in
| the catalog files, which are only used by DSSSL processors, like jade or
| openjade. Or am I wrong on this issue?

Two wrong issues :-)

First, the files are referenced in the DTD, docbookx.dtd pulls in
dbcentx.mod which refers to each of the files.

Second, catalog files are by no means limited to SGML or DSSSL
processors.  I've got catalog support in XT and Xerces and I'm working
on getting it in Xalan. I'll probably post something officially about
this work in the next month or two.

| The xalan processing fails:
| 
| java org.apache.xalan.xslt.Process -in 
| evaluation_textverarbeitungssysteme.xml -xsl

Do you have a <!DOCTYPE declaration in this file?

This test file works fine for me:

<!DOCTYPE para SYSTEM "/share/doctypes/docbook/xml/docbookx.dtd">
<para>This is a&nbsp;test.</para>

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | As charms are nonsense, nonsense is a
http://nwalsh.com/            | charm.--Benjamin Franklin