refdb-users — List for RefDB users

Re: [Refdb-users] User feedback 3

[Markus H. <mar...@mh...>]

David Nebauer writes:
 > At this point may I make a bold suggestion: the addition to refdb of a 
 > refdbxml configuration/init file, presumably .refdbxmlrc.  This would 
 > have the following advantages:
 > 
 > 1. Rather than directly editing refdbxml, as is necessary now if you 
 > want to changes xslt processors, you can instead nominate the processor 
 > to use in the config file.  This makes your choices upgrade-proof.  You 
 > could either have saxon and saxon-xerces as choices for the processor, 
 > or you could have a separate variable (use-xerces) in the init file.
 > 
 > 2. If you add support for FOP, there needs to be a way to nominate which 
 > FO processor to use.  An init file makes this easy.
 > 
 > 3. I would dearly love to see it possible to nominate, in this init 
 > file, a classpath specific to each java application.  Perhaps the use of 
 > variables such as: saxon-classpath, xalan-classpath, xerces-classpath 
 > and fop-classpath.  These classpaths could be extracted by refdbxml and 
 > used as a -cp command-line option when invoking the relevant processor.
 > 
 > My thinking here is that, with 4 java-based apps in potential use 
 > (saxon, xalan, xerces and fop) the current "common classpath" stands to 
 > become increasingly complex.  Relying on a single classpath variable for 
 > increasing numbers of java applications is difficult to debug, difficult 
 > to maintain, unwieldy, inelegant and just plain _messy_ .
 > 

This is my favourite rant about Java ;-)
 > 
 > My powers of shellscript-fu are a little stronger than my (pathetic) 
 > xsl-fu, and I could try to knock up a "proof-of-concept" if you are 
 > inclined to look into my suggestion.
 > 

I strongly encourage you to whip up a proof of concept. The shell
magic to read config files is not that hard BTW, the db2ris script
shows an example how to do this.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] User feedback 3

[Markus H. <mar...@mh...>]

David Nebauer writes:
 > The situations it reports are accurate.  The template starting at line 
 > 1821 _is_ the only template where <xsl:choose> appears in a different 
 > location to all the others.  It also appears to be the only template 
 > where xsl:withparam _is_ a direct child of xsl:when.
 > 
[...]
 > What Saxon is saying about this part of the stylesheet is certainly 
 > true.  I don't know enough about xsl to know whether this stylesheet is 
 > strictly legal.  If this xsl is strictly legal and Saxon is mistakenly 
 > choking on it, then I'll need to raise this issue on the Saxon forums.  
 > It is, however, possible that Saxon is picking up malformed xsl that 
 > xsltproc is happy to live with.  In that case, this stylesheet needs to 
 > be corrected.
 > 

Your analysis is correct. I've investigated this and found out that
the .xsl file represents the status of some unfinished hacking. I've
updated the Perl script that generates the .xsl file and hope that the
generated file works for Saxon as well. I've sent you an updated
version by private mail. Please let me know whether or not it works.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] User feedback 2

[Markus H. <mar...@mh...>]

David Nebauer writes:
 > A small point: the refdbibrc.example file contains a bit of old comment:
 > 
 > # Bibliography output type. Currently only "db31" available
 > #outtype    db31
 > 

Well, this probably reflects the status as of 2001. Thanks for
pointing this out. It is fixed in CVS.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] User feedback 2

[Markus H. <mar...@mh...>]

David Nebauer writes:
 > Hi Markus,
 > 
 > > I'm afraid you did not set up a .refdbibrc config file. This should
 > > contain the username (if it is different from the current login name)
 > > and either the password or a '*' to get the password prompt. runbib is
 > > just a script that calls refdbib at some point.
 > 
 > Hmm, quite right.  In my defence, may I plead a _slight_ lack of clarity
 > in the manual.  In the section "The non-interactive mode of refdbib"
 > (http://refdb.sourceforge.net/manual/x2084.html) you refer readers to
 > "Starting refdbib" (http://refdb.sourceforge.net/manual/x6660.html) which
 > refers back to "The mystery of the configuration files"
 > (http://refdb.sourceforge.net/manual/x575.html#SECT1-MYSTERY-INIT-FILES)
 > for a general discussion of config files.
 > 
 > However, in that first section I mentioned, "The non-interactive mode of
 > refdbib" (http://refdb.sourceforge.net/manual/x2084.html), you recommend
 > that users instead use runbib and refer them to "The runbib shell script"
 > (http://refdb.sourceforge.net/manual/x8190.html).  No mention is made in
 > this page of the need to set up a .refdbibrc config file.
 > 

No need to defend yourself. If reading the docs didn't do the trick,
then the docs need to get better.
 
 > Many users like myself may skip the details about refdbib configuration,
 > instead going straight to runbib as you recommend.
 > 
 > Perhaps you could add a short para to the runbib page mentioning the need
 > to set up .refdbibrc file?
 > 

Sure. I put this on my list.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] User feedback 1

[Markus H. <mar...@mh...>]

David Nebauer writes:
 > The best way to get usable pdf output from a DocBook XML-generated .fo
 > file, IMHO, would be to use fop (as you mention trying out in another
 > reply).  The install is simple -- just decompress the files into the
 > desired directory (see
 > http://apache.planetmirror.com/dist/xml/fop/README.html).  I used the
 > instructions in Bob Stayton's book
 > (http://www.sagehill.net/docbookxsl/InstallingAnFO.html) to set it up.  He
 > mentions a number of ways to invoke fop, but I use a command like:
 > java -cp "CLASSPATH_AS_PER_BOB_STAYTONS_INSTRUCTIONS" \
 >     org.apache.fop.apps.Fop  \
 >     -fo  myfile.fo  \
 >     -pdf myfile.pdf
 > 

Thanks David, I'll give it a try.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] User feedback 1

[David N. <dav...@bi...>]

Hi Markus,

> > 2. The table of contents has a an entry for the Reference List.  The 
> > page number and leading dots are there, but the heading "Reference List" 
> > is not.  The leading dots start at the left margin.  Again, html output 
> > renders correctly.
> > 
> > Perhaps there is a workaround in how the xml source is encoded.  I hope so.
>  
>
I cannot explain it, but this problem has spontaneously fixed itself.

The document in which I observed this phenomenon is a test document in 
which I am exploring the different features of DocBook XML.  It has been 
growing in size and complexity and I periodically output the document as 
html and pdf to assess the results.  At one point I suddenly noticed the 
Reference List entry in the Table of Contents was displaying correctly.  
It has continued to do so for several iterations of the document.

As I said, I have no explanation for this.  Without an earlier version 
of the document, I have no way to reproduce the problem.  I quickly 
created a new document and it, too, displayed the ToC correctly.  For 
now, I must declare this problem fixed.

Regards,
David.

P.S.  The other Reference List problem (second printing of words 
"Reference List" under "Reference List" heading, but only in pdf output) 
remains.

Re: [Refdb-users] User feedback 1

[Bruce D'A. <bd...@fa...>]

On Mar 2, 2004, at 8:58 AM, David Nebauer wrote:

> As I see it with PassiveTeX, the problem occurs when a document 
> _created in XML_ is converted to TeX for outputting.  The problem 
> occurs at the conversion step and is nothing to do with the underlying 
> TeX formatting engine.

The problem is that the FO --> TeX route cuts out all the macro power 
in LaTeX.  So, the code (both in the fo file, and in passivetex itself) 
gets rather complex, and prone to breakage.  This is why I've started 
to just output my XML docs directly to LaTeX code, though I still need 
to solve the bib issue (a big one!).

Bruce

Re: [Refdb-users] User feedback 1

[David N. <dav...@bi...>]

Hi Bruce,

>
> My problem with the fo professors is that there output quality is an 
> obvious step below that of TeX.
>
I have to agree that a document written in TeX/LaTeX/LyX and outputted 
using a TeX engine is going to look fantastic.  There is just no denying 
how well TeX formats documents.  There must be hundreds, if not 
thousands, of man-years invested in the whole TeX "ecology".  I 
personally have been using LyX/BibTeX for my assignments for years.

As I see it with PassiveTeX, the problem occurs when a document _created 
in XML_ is converted to TeX for outputting.  The problem occurs at the 
conversion step and is nothing to do with the underlying TeX formatting 
engine.

At the moment FO seems to be a more "natural" fit with XML.

I have to admit, though, that there are a _lot_ of rough edges with FO 
processors and it will be years, if ever, before they have same 
refinement as TeX.

Regards,
David.

[Refdb-users] Using xsl customisation layer with refdb

[David N. <dav...@bi...>]

After some help from Markus I have managed to implement an xsl 
customisation layer with refdb.

Here's how in case there are any other newbies wondering.

I use DocBook XML on a Debian Sarge/testing box.


First, the theory:

Usually, at configure time the --with-docbook-xsl option is set to point 
at the root directory of the original ("Norman Walsh") docbook 
stylesheets.  On my box that is: 
/usr/share/sgml/docbook/stylesheet/xsl/nwalsh.

The refdb install sets the refdb stylesheets to import those original 
docbook stylesheets.

So, when you create a document and type in "make html|pdf|whatever", the 
makefile creates a stylesheet in the document directory --> it calls the 
refdb stylesheets --> which call the original docbook stylesheets.

What we are going to do is create an extra link in that chain: the 
makefile will create a stylesheet in the document directory --> it  
calls the refdb stylesheets --> which call our customisation layer 
stylesheets --> which call the original docbook stylesheets.


Now, the practice:

The first thing to do is create the customisation layer.  You must 
create the following directory/file structure in a directory of your choice:

     BASE_DIR/fo
     BASE_DIR/fo/docbook.xsl
     BASE_DIR/html
     BASE_DIR/html/docbook.xsl
     BASE_DIR/xhtml
     BASE_DIR/xhtml/docbook.xsl

Each docbook.xsl file must include an appropriate import statement 
importing the matching original docbook stylesheet.

Here is my fo/docbook.xsl file:

----------------------------------------------------------------------------------------
<!--
FO customisation layer for refdb stylesheets

First imports original DocBook stylesheets

Finally imports my custom stylesheets
-->

<xsl:stylesheet
  xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"
  xmlns:fo="http://www.w3.org/1999/XSL/Format">

    <!-- the original docbook stylesheets for FO -->
    <xsl:import 
href="/usr/share/sgml/docbook/stylesheet/xsl/nwalsh/fo/docbook.xsl"/>

    <!-- my customisation layer -->
    <xsl:import 
href="/home/david/data/conf/xsl/custom-layer/fo-custom.xsl"/>

</xsl:stylesheet>
----------------------------------------------------------------------------------------

You must include a file called VERSION in the base directory of your 
customisation layer.  It is to keep the configure test happy.  The file 
can be empty.  I created mine using the command: touch VERSION.

Now you are ready to install (or re-install) refdb with the proper 
configuration.  At the ./configure step you must use the 
--with-docbook-xsl option and specify the full path to your 
customisation layer base directory.

On my box, that option looked like:
--with-docbook-xsl=/home/david/data/conf/refdb/xsl

Hope that helps,
David.

Re: [Refdb-users] User feedback 1

[Bruce D'A. <bd...@fa...>]

On Mar 1, 2004, at 10:20 PM, David Nebauer wrote:

> Bob has some strong opinions about the current compatibility of 
> passivetex
> and the DocBook XSL stylesheets, and recommends fop over passivetex.

My problem with the fo professors is that there output quality is an 
obvious step below that of TeX.

Bruce

Re: [Refdb-users] User feedback 4

[David N. <dav...@bi...>]

Hi Markus,

>I haven't tried this yet, but why don't you try and point configure
>--with--docbook-xsl to your customization layer instead of to the
>official stylesheets? This requires a file "VERSION" in your stylesheet
>directory to make the configure test happy. It should work ok
>otherwise if your stylesheets are arranged in the same directory
>structure as the official stylesheets.
>  
>
Thanks for that suggestion.  After playing around a little I managed to 
get it working.

I'll post a message to the forum explaining what I did.  I'll give it a 
relevant heading so anyone in the future trawling through the archive 
will find it.

Regards,
David.

Re: [Refdb-users] User feedback 3

[David N. <dav...@bi...>]

Hi Markus,

> > 7. Staying with saxon, if it is ever able to work with refdb, it would 
> > be good to have an option to run it with the xerces parser.  This 
> > requires a (lengthy) command line option, but it should be rather doable.
> > 
>
>I'm reluctant to fiddle with this stuff as the Java installations are
>even more diverse in terms of install directories between systems than
>standard C/C++ programs are. A simple configure test can verify the
>installation of xsltproc, but not of saxon and other Java tools (at
>least I haven't seen a configure script that I could steal code
>from). If you think there is a generally applicable way to handle
>this, please let me know. I'll be happy to include this feature.
>  
>
Providing the class path has been set up correctly in each case, the 
command to invoke the Saxon processor with or without the Xerces parser 
is the same regardless of the installation configuration of Saxon or 
Xerces -- this is part of the promise of Java, after all ;-)

Here is the command to invoke vanilla Saxon (I'm cribbing from Bob 
Stayton's book again) for html output:

java  -cp $CLASSPATH \
    com.icl.saxon.StyleSheet  \
    -o myfile.html  \
    myfile.xml  \
    docbook-xsl/html/docbook.xsl

Here is the same command, but specifying use of the Xerces parser:

java -cp $CLASSPATH \ 
  -Djavax.xml.parsers.DocumentBuilderFactory=org.apache.xerces.jaxp.DocumentBuilderFactoryImpl \
  -Djavax.xml.parsers.SAXParserFactory=org.apache.xerces.jaxp.SAXParserFactoryImpl \
  com.icl.saxon.StyleSheet  \
  -o myfile.html  \
  myfile.xml  \
  docbook-xsl/html/docbook.xsl

At first glance it looks hideous, but there are simply two additional 
command line options specifying the parser, with the rest being the same.

I don't know if it's necessary to worry about java-based apps like 
Xalan, Saxon, Xerces or FOP at configure time.  You can keep xsltproc as 
the default and, as you do now, make it clear in the manual that use of 
Saxon and Xalan require a working stallation of these packages.

At this point may I make a bold suggestion: the addition to refdb of a 
refdbxml configuration/init file, presumably .refdbxmlrc.  This would 
have the following advantages:

1. Rather than directly editing refdbxml, as is necessary now if you 
want to changes xslt processors, you can instead nominate the processor 
to use in the config file.  This makes your choices upgrade-proof.  You 
could either have saxon and saxon-xerces as choices for the processor, 
or you could have a separate variable (use-xerces) in the init file.

2. If you add support for FOP, there needs to be a way to nominate which 
FO processor to use.  An init file makes this easy.

3. I would dearly love to see it possible to nominate, in this init 
file, a classpath specific to each java application.  Perhaps the use of 
variables such as: saxon-classpath, xalan-classpath, xerces-classpath 
and fop-classpath.  These classpaths could be extracted by refdbxml and 
used as a -cp command-line option when invoking the relevant processor.

My thinking here is that, with 4 java-based apps in potential use 
(saxon, xalan, xerces and fop) the current "common classpath" stands to 
become increasingly complex.  Relying on a single classpath variable for 
increasing numbers of java applications is difficult to debug, difficult 
to maintain, unwieldy, inelegant and just plain _messy_ .


I realise you can nominate the desired processor on the command line 
when invoking refdbxml, but using the init file approach I have 
suggested would be consistent with the refdbnd/"make" approach, which 
relies heavily on config files.  (And I just _love_ refdbnd and the 
makefile approach).

My powers of shellscript-fu are a little stronger than my (pathetic) 
xsl-fu, and I could try to knock up a "proof-of-concept" if you are 
inclined to look into my suggestion.

Regards,
David.

Re: [Refdb-users] User feedback 3

[David N. <dav...@bi...>]

Hi Markus,

> > refdbxml -s Eur.J.Pharmacol.html.xsl -t html basic-doc.xml
> > Error at xsl:with-param on line 74 of 
> > file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-html/docbk-refdb-html.xsl:
> >   Variable ADVSTURLSTYLE has not been declared
>
>This is a typo in docbk-refdb-html.xsl. Change this to ADVSURLSTYLE
>(without the 'T' after ADVS). I'm not an XSLT guru but fixing this
>might fix the other error messages as well. This does not exactly
>explain why xsltproc does not complain, though. I've fixed this in CVS
>anyway.
>  
>
I'm afraid it does not fix the following messages.  Here is the current 
error output:

$ make html
refdbxp -t db31x < basic-doc.short.xml > basic-doc.xml
runbib -d refs_computing -S Eur.J.Pharmacol. -t db31x basic-doc.xml
5 reference(s) formatted, 0 failed

refdbxml -s Eur.J.Pharmacol.html.xsl -t html basic-doc.xml
Error at xsl:choose on line 1826 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  Element must only be used within a template body
Error at xsl:with-param on line 1828 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1831 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1834 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1837 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1840 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1843 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1846 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1849 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1852 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1855 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1858 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1861 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1864 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1867 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1870 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1873 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1876 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1879 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1882 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1885 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1888 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1891 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1894 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1897 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1900 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1903 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1906 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1909 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1912 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1915 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1918 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1921 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1924 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1927 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1930 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Transformation failed: Failed to compile stylesheet. 36 errors detected.

The situations it reports are accurate.  The template starting at line 
1821 _is_ the only template where <xsl:choose> appears in a different 
location to all the others.  It also appears to be the only template 
where xsl:withparam _is_ a direct child of xsl:when.

A representative snippet of the "non-erroring" templates:

<xsl:template match="bibliomset[@relation='seditor']" mode="refdb">
  <xsl:variable name="reftype" select="ancestor::bibliomixed/@role"/>
      <xsl:choose>                        <--- **
        <xsl:when test="$reftype='ABST'">
          <xsl:call-template name="refdb-process-inline">      <--**
            <xsl:with-param name="style" select="$ABSTSEDITORLISTSTYLE"/>
          </xsl:call-template>
        </xsl:when>

The "erroring" template:

  <xsl:template match="ulink" mode="refdb">
    <xsl:variable name="reftype" select="ancestor::bibliomixed/@role"/>
      <xsl:variable name="target" select="@url"/>
      <a href='#{$target}'>
      <xsl:call-template name="refdb-literal">
        <xsl:choose>                            <--**
          <xsl:when test="$reftype='ABST'">
            <xsl:with-param name="style" select="$ABSTURLSTYLE"/>     <--**
          </xsl:when>

What Saxon is saying about this part of the stylesheet is certainly 
true.  I don't know enough about xsl to know whether this stylesheet is 
strictly legal.  If this xsl is strictly legal and Saxon is mistakenly 
choking on it, then I'll need to raise this issue on the Saxon forums.  
It is, however, possible that Saxon is picking up malformed xsl that 
xsltproc is happy to live with.  In that case, this stylesheet needs to 
be corrected.

Without knowing anything about xsl, I attempted to use xslint to check 
the offending stylesheet.  It output a vast number of cryptic error 
messages and I quickly abandoned that line of inquiry.  Of course, I may 
have invoked it incorrectly or something.  It will require someone with 
more xsl knowledge than I currently possess to decide on this 
styleheet's correctness.

Regards,
David.

Re: [Refdb-users] User feedback 2

[David N. <dav...@bi...>]

Hi Markus,

>I'm afraid you did not set up a .refdbibrc config file. This should
>contain the username (if it is different from the current login name)
>and either the password or a '*' to get the password prompt. runbib is
>just a script that calls refdbib at some point.
>  
>
A small point: the refdbibrc.example file contains a bit of old comment:

# Bibliography output type. Currently only "db31" available
#outtype    db31

Regards,
David.

Re: [Refdb-users] User feedback 2

[David N. <dav...@bi...>]

Hi Markus,

> I'm afraid you did not set up a .refdbibrc config file. This should
> contain the username (if it is different from the current login name)
> and either the password or a '*' to get the password prompt. runbib is
> just a script that calls refdbib at some point.

Hmm, quite right.  In my defence, may I plead a _slight_ lack of clarity
in the manual.  In the section "The non-interactive mode of refdbib"
(http://refdb.sourceforge.net/manual/x2084.html) you refer readers to
"Starting refdbib" (http://refdb.sourceforge.net/manual/x6660.html) which
refers back to "The mystery of the configuration files"
(http://refdb.sourceforge.net/manual/x575.html#SECT1-MYSTERY-INIT-FILES)
for a general discussion of config files.

However, in that first section I mentioned, "The non-interactive mode of
refdbib" (http://refdb.sourceforge.net/manual/x2084.html), you recommend
that users instead use runbib and refer them to "The runbib shell script"
(http://refdb.sourceforge.net/manual/x8190.html).  No mention is made in
this page of the need to set up a .refdbibrc config file.

Many users like myself may skip the details about refdbib configuration,
instead going straight to runbib as you recommend.

Perhaps you could add a short para to the runbib page mentioning the need
to set up .refdbibrc file?

Regards,
David.

Re: [Refdb-users] User feedback 1

[David N. <dav...@bi...>]

Hi Markus,

> I hope no one is going to laugh too loud, but I've got a small problem
> here: none of the systems that I currently work with creates any
> printable XML output at all.

FWIW, I'm running Debian testing/Sarge and already had passivetex
installed as a dependency of one of the LaTeX|TeX-based packages I've
previously used (mostly LyX).  When it came to using it with refdb,
passivetex worked "out of the box", requiring no alteration from me (aside
from the config file alterations you mention in the manual).

Presuming you can get a .fo file from xsltproc, there are a couple of ways
of getting PDF output:

The best way to get usable pdf output from a DocBook XML-generated .fo
file, IMHO, would be to use fop (as you mention trying out in another
reply).  The install is simple -- just decompress the files into the
desired directory (see
http://apache.planetmirror.com/dist/xml/fop/README.html).  I used the
instructions in Bob Stayton's book
(http://www.sagehill.net/docbookxsl/InstallingAnFO.html) to set it up.  He
mentions a number of ways to invoke fop, but I use a command like:
java -cp "CLASSPATH_AS_PER_BOB_STAYTONS_INSTRUCTIONS" \
    org.apache.fop.apps.Fop  \
    -fo  myfile.fo  \
    -pdf myfile.pdf

I've actually written a small shell script to make life easier which I'll
happily post if you like.  As a java app, fop should work on any of your
systems.

Bob has some strong opinions about the current compatibility of passivetex
and the DocBook XSL stylesheets, and recommends fop over passivetex.

RenderX's xep comes in a free trial version
(http://xep.xattic.com/xep/editions/trial.html) which prints documents up
to 10 pages.  It is a simple install as well -- decompress into a
directory and run with java.  It, too, works "out of the box" and could be
used for debugging.

Regards,
David.

Re: [Refdb-users] User feedback 4

[Bruce D'A. <bd...@fa...>]

On Mar 1, 2004, at 5:05 PM, Markus Hoenicka wrote:

>> 10. May I suggest adding support for the Apache FO processor: FOP.  I
>> find its output much cleaner than pdfxmltex.  Some of the problems 
>> with
>> the pdfxmltex output of my text document are:
>> - header (document title) is only two words, but is split across two 
>> lines
>> - header and footer separation lines only cross left half of page
>> - revision history output squeezed to left side of page
>> - table of contents header starts _inside_ left margin while other
>> headers start _at_ left margin
>> - inclusion of "rubbish" text at top left of each output page: - -4pc 
>> - -4pc
>> - tables have no vertical or horizontal lines
>> - page numbers not horizontally centred

The problem with DocBook output with passivetex is not exactly 
passivetex per se, but rather the interaction between the DocBook 
stylesheets and passivetex.  Some of this has been fixed recently (the 
"rubbish" thing above), but a lot the rest of it is less-than-ideal 
table handling in passivetex, and the fact that the stylesheets use 
features not support in PT.

I'm personally writing my own XML 2 TeX stylesheets :-)

Bruce

[Refdb-users] User feedback 4

[Markus H. <mar...@mh...>]

Hi again,

David Nebauer writes:
 > [...continued]
 > 
 > 8.  In multiple citations html output contains a single hyperlink to the 
 > first reference.  I seem to recall reading about this in a previous 
 > thread on refdb-users.  IIRC, you said it was non-trivial to correct and 
 > (?) you were not planning to do anything about it in a hurry.  It's not 
 > a major issue for me but I include it in my list.
 > 

I'm still not planning to play with this. If the bibliofile approach
matures, there may be a more elegant solution available for this problem.

 > 9.  make clean doesn't remove .aux or .fo files.
 > 

I've fixed this in CVS.

 > 10. May I suggest adding support for the Apache FO processor: FOP.  I 
 > find its output much cleaner than pdfxmltex.  Some of the problems with 
 > the pdfxmltex output of my text document are:
 > - header (document title) is only two words, but is split across two lines
 > - header and footer separation lines only cross left half of page
 > - revision history output squeezed to left side of page
 > - table of contents header starts _inside_ left margin while other 
 > headers start _at_ left margin
 > - inclusion of "rubbish" text at top left of each output page: - -4pc - -4pc
 > - tables have no vertical or horizontal lines
 > - page numbers not horizontally centred

Now that you mention it, I recall I've seen all these symptoms on
XP/Cygwin, the only platform that gave me printable, albeit broken XML
output.

 > I mention this not because I expect you to correct the deficiencies of 
 > pdfxmltex, but as a reason to support FOP.  In my test document, FOP 
 > outputted pdf with none of the above-mentioned flaws.  In fact, it's 
 > output was indistinguishable from that of RenderX's (commercially 
 > available) xep processor.  In addition, on my machine at least, FOP is 
 > faster than pdfxmltex.
 > 
 > It is easy, of course, to use the .fo file from the make process as 
 > input for FOP, but it would be nice to add the option to refdb and 
 > automate the process.
 > 

The reason why FOP is not included is historic. Last time I checked
(this must be more than a year ago) FOP was in such a broken state
that it seemed useless compared to passivetex (which happened to work
back then on my then-Debian-2.x installation). I've never bothered to
check since, but if it works ok now I'll be happy to include it.

 > 11. I have an XSL customisation layer for Norm Walsh's stylesheets.  
 > With standard docbook it is trivial to use a customisation stylesheet 
 > which imports the main stylesheet and overrides the relevant 
 > parameters.  It is difficult, however, to see how this can easily be 
 > done with refdb.  The base stylesheets are created anew each time make 
 > is run.  I see that these stylesheets basically call on refdb 
 > stylesheets docbk-refdb-XX which themselves call on Norm Walsh's 
 > stylesheets.  If I were to modify the docbk-refdb-XX stylesheets (the 
 > most accessible answer for me) to override certain parameters, I would 
 > have to do the whole thing over next time I upgrade refdb.
 > 
 > I wonder whether a structural solution using refdb is possible.  Perhaps 
 > a build-time configure option to point to fo|html|xhtml customisation 
 > stylesheets which can then be included in the makefile generated 
 > stylesheets.
 > 

I haven't tried this yet, but why don't you try and point configure
--with--docbook-xsl to your customization layer instead of to the
official stylesheets? This requires a file "VERSION" in your stylesheet
directory to make the configure test happy. It should work ok
otherwise if your stylesheets are arranged in the same directory
structure as the official stylesheets.

 > 12.  Finally, I come to the questions asked by everybody who reaches the 
 > point of having a fully functional refdb system -- where are the 
 > styles?!  I have read the refdb-user archives and know your standard 
 > response -- go forth and modify existing ones/write your own.  I respect 
 > that and intend to begin the long journey to XSL mastery.  Can you, 
 > however,  point to any current efforts under way to modify/create refdb 
 > styles that I can plagiarise/piggyback on/learn from?
 > 

/usr/local/share/refdb/styles? Not that many to learn from, but it's a
start...

 > 
 > That's all the feedback I have for now, except to repeat again how 
 > impressed I am with refdb.  For want of a better word, it is an ... 
 > elegant ... solution to a bibliographic problem.
 > 

I guess Homer would say: "sweet"!

Thanks for your input.

Markus


-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

[Refdb-users] User feedback 3

[Markus H. <mar...@mh...>]

David Nebauer writes:
 > [.. continued]
 > 
 > 6. If refdbxml is edited to set saxon as default processor, the make 
 > procedure fails.
 > 
 > The problem occurs with file docbk-refdblib.xsl.  Here is the output:
 > 
 > $ make html
 > refdbxp -t db31x < basic-doc.short.xml > basic-doc.xml
 > runbib -d refs_computing -S Eur.J.Pharmacol. -t db31x basic-doc.xml
 > 5 reference(s) formatted, 0 failed
 > 
 > refdbxml -s Eur.J.Pharmacol.html.xsl -t html basic-doc.xml
 > Error at xsl:with-param on line 74 of 
 > file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-html/docbk-refdb-html.xsl:
 >   Variable ADVSTURLSTYLE has not been declared

This is a typo in docbk-refdb-html.xsl. Change this to ADVSURLSTYLE
(without the 'T' after ADVS). I'm not an XSLT guru but fixing this
might fix the other error messages as well. This does not exactly
explain why xsltproc does not complain, though. I've fixed this in CVS
anyway.

 > 
 > 7. Staying with saxon, if it is ever able to work with refdb, it would 
 > be good to have an option to run it with the xerces parser.  This 
 > requires a (lengthy) command line option, but it should be rather doable.
 > 

I'm reluctant to fiddle with this stuff as the Java installations are
even more diverse in terms of install directories between systems than
standard C/C++ programs are. A simple configure test can verify the
installation of xsltproc, but not of saxon and other Java tools (at
least I haven't seen a configure script that I could steal code
from). If you think there is a generally applicable way to handle
this, please let me know. I'll be happy to include this feature.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

[Refdb-users] User feedback 2

[Markus H. <mar...@mh...>]

David Nebauer writes:
 > [... continued]
 > 
 > 3. When importing files from other formats _with their own citation 
 > keys_, it is possible to end up with refdb citation keys that have 
 > colons in them.  Any such citation will not appear in the final output.
 > 
 > I presume this is to do with parsing for the A: and Y: parts of the 
 > citations.
 > 
 > I imagine possible solutions may include preventing such references 
 > being uploaded, silently removing the colon or simply ignoring the 
 > citation key and adding a new one according to the usual algorithms.
 > 

Yes, this is an oversight. Colons should be stripped from citation
keys. I'll fix that.

 > 4. When using RIS format you must use BT for book titles -- if TI is 
 > used the title doesn't print in the reference list.  The manual (p.44) 
 > says that BT and TI are the same for books, but it doesn't say you 
 > _must_ use BT.
 > 

This is an attempt to make things a little more reasonable and to get
this stuff in line with risx. The manual more or less reflects the tag
usage of Reference Manager which is quite a mess. I'll fix this
section in the manual.

 > 5. The makefile will fail on the runbib command unless the runbib file 
 > is edited to include at least the user's username.  Adding password 
 > prevents runbib from pausing to ask for password.
 > 
 > Here is the output when username is _not_ present in runbib:
 > 
 > $ make html
 > refdbxp -t db31x < basic-doc.short.xml > basic-doc.xml
 > runbib -d refs_computing -S Eur.J.Pharmacol. -t db31x basic-doc.xml
 > Please enter your password:
 > server error: could not connect to database server
 > .
 > mv: cannot stat `Eur.J.Pharmacol.xsl': No such file or directory
 > make: *** [basic-doc.bib.xml] Error 1
 > 
 > The username and password lines in runbib occur _after_ the 
 > user-customisable section, so it may not be readily apparent to users 
 > that they need to amend them.
 > 
 > Perhaps runbib should prompt for username?
 > 

I'm afraid you did not set up a .refdbibrc config file. This should
contain the username (if it is different from the current login name)
and either the password or a '*' to get the password prompt. runbib is
just a script that calls refdbib at some point.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

[Refdb-users] User feedback 1

[Markus H. <mar...@mh...>]

Hi David,

first of all, thanks for your thorough testing and for sharing your
thoughts. Let's go through your quite impressive list of
questions/suggestions.

David Nebauer writes:
 > 1. In printed (fo-->pdf) output the reference list heading appears 
 > twice: once as a major heading and once again just below in normal print 
 > .  I have tried the standard pdftex output (as per the makefile) and 
 > also used fop and RenderX/xep to transform the .fo file: same results.  
 > This is not seen in html output which displays normally.
 > 
 > 2. The table of contents has a an entry for the Reference List.  The 
 > page number and leading dots are there, but the heading "Reference List" 
 > is not.  The leading dots start at the left margin.  Again, html output 
 > renders correctly.
 > 
 > Perhaps there is a workaround in how the xml source is encoded.  I hope so.
 > 

I hope no one is going to laugh too loud, but I've got a small problem
here: none of the systems that I currently work with creates any
printable XML output at all. Therefore I can't debug the passivetex
output, I can't even comment on it. In case you wonder how I print
things, well... I use SGML + openjade. Therefore...

1) Is anyone with a working passivetex installation willing to look
   into this?

2) Is anyone with a working passivetex installation willing to
   enlighten me how to get this going on at least one of my systems?

I'm mainly working on FreeBSD 4.7 with a current ports collection,
and occasionally on Debian 3.0 (mostly stable). If nothing else helps,
I can use my box at work which runs XP + Cygwin.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

[Refdb-users] Initial setup of databases

[Markus H. <mar...@mh...>]

Hi Todd,

Todd Wallentine writes:
 > I am working on getting RefDB 0.9.4 setup for my research group and 
 > needed some clarification.  My understanding is that there is supposed 
 > to be a refdb database that is the "system" database.  I am then 
 > supposed to create a new database that will store the actual reference 
 > information.  Is that right so far?
 > 

Yes that's right. Every RefDB installation needs exactly one system
database which is called "refdb" in the current release. On the
contrary, you may have as many reference databases as you like,
although it's a good idea to keep all references in a single database.

 > Ok, so what tables are supposed to be in the "refdb" database?  Just 
 > t_journal_words, CITSTYLE, REFSTYLE, SEPARATORS, and POSITION?  Or do I 
 > add the other t_* tables to it as well?
 > 
 > Now, when I create a new database with refdba, what tables should this 
 > create?  For example, if I do "create foo" and then look at the tables 
 > in the database, what should I see?
 > 
The manual has more details about it, but you can simply have a look
at the SQL scripts shipped with RefDB. refdb.dump creates the system
database "refdb" and creates the following tables (verify this by
running 'grep "CREATE TABLE" *' in the scripts subdirectory):

refdb.dump:CREATE TABLE CITSTYLE (
refdb.dump:CREATE TABLE POSITIONS (
refdb.dump:CREATE TABLE REFSTYLE (
refdb.dump:CREATE TABLE SEPARATORS (

The reference databases currently contain the following tables, as
seen in empty.mysql.dump:

empty.mysql.dump:CREATE TABLE t_author (
empty.mysql.dump:CREATE TABLE t_keyword (
empty.mysql.dump:CREATE TABLE t_meta (
empty.mysql.dump:CREATE TABLE t_note (
empty.mysql.dump:CREATE TABLE t_periodical (
empty.mysql.dump:CREATE TABLE t_refdb (
empty.mysql.dump:CREATE TABLE t_user (
empty.mysql.dump:CREATE TABLE t_xauthor (
empty.mysql.dump:CREATE TABLE t_xkeyword (
empty.mysql.dump:CREATE TABLE t_xnote (
empty.mysql.dump:CREATE TABLE t_xuser (

regards,
Markus


-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

[Refdb-users] Initial setup of databases

[Todd W. <tc...@ks...>]

I am working on getting RefDB 0.9.4 setup for my research group and 
needed some clarification.  My understanding is that there is supposed 
to be a refdb database that is the "system" database.  I am then 
supposed to create a new database that will store the actual reference 
information.  Is that right so far?

Ok, so what tables are supposed to be in the "refdb" database?  Just 
t_journal_words, CITSTYLE, REFSTYLE, SEPARATORS, and POSITION?  Or do I 
add the other t_* tables to it as well?

Now, when I create a new database with refdba, what tables should this 
create?  For example, if I do "create foo" and then look at the tables 
in the database, what should I see?

Thanks in advance,
todd wallentine

[Refdb-users] User feedback 4

[David N. <dav...@bi...>]

[...continued]

8.  In multiple citations html output contains a single hyperlink to the 
first reference.  I seem to recall reading about this in a previous 
thread on refdb-users.  IIRC, you said it was non-trivial to correct and 
(?) you were not planning to do anything about it in a hurry.  It's not 
a major issue for me but I include it in my list.

9.  make clean doesn't remove .aux or .fo files.

10. May I suggest adding support for the Apache FO processor: FOP.  I 
find its output much cleaner than pdfxmltex.  Some of the problems with 
the pdfxmltex output of my text document are:
- header (document title) is only two words, but is split across two lines
- header and footer separation lines only cross left half of page
- revision history output squeezed to left side of page
- table of contents header starts _inside_ left margin while other 
headers start _at_ left margin
- inclusion of "rubbish" text at top left of each output page: - -4pc - -4pc
- tables have no vertical or horizontal lines
- page numbers not horizontally centred
I mention this not because I expect you to correct the deficiencies of 
pdfxmltex, but as a reason to support FOP.  In my test document, FOP 
outputted pdf with none of the above-mentioned flaws.  In fact, it's 
output was indistinguishable from that of RenderX's (commercially 
available) xep processor.  In addition, on my machine at least, FOP is 
faster than pdfxmltex.

It is easy, of course, to use the .fo file from the make process as 
input for FOP, but it would be nice to add the option to refdb and 
automate the process.

11. I have an XSL customisation layer for Norm Walsh's stylesheets.  
With standard docbook it is trivial to use a customisation stylesheet 
which imports the main stylesheet and overrides the relevant 
parameters.  It is difficult, however, to see how this can easily be 
done with refdb.  The base stylesheets are created anew each time make 
is run.  I see that these stylesheets basically call on refdb 
stylesheets docbk-refdb-XX which themselves call on Norm Walsh's 
stylesheets.  If I were to modify the docbk-refdb-XX stylesheets (the 
most accessible answer for me) to override certain parameters, I would 
have to do the whole thing over next time I upgrade refdb.

I wonder whether a structural solution using refdb is possible.  Perhaps 
a build-time configure option to point to fo|html|xhtml customisation 
stylesheets which can then be included in the makefile generated 
stylesheets.

12.  Finally, I come to the questions asked by everybody who reaches the 
point of having a fully functional refdb system -- where are the 
styles?!  I have read the refdb-user archives and know your standard 
response -- go forth and modify existing ones/write your own.  I respect 
that and intend to begin the long journey to XSL mastery.  Can you, 
however,  point to any current efforts under way to modify/create refdb 
styles that I can plagiarise/piggyback on/learn from?


That's all the feedback I have for now, except to repeat again how 
impressed I am with refdb.  For want of a better word, it is an ... 
elegant ... solution to a bibliographic problem.

Regards,
David.

[Refdb-users] User feedback 3

[David N. <dav...@bi...>]

[.. continued]

6. If refdbxml is edited to set saxon as default processor, the make 
procedure fails.

The problem occurs with file docbk-refdblib.xsl.  Here is the output:

$ make html
refdbxp -t db31x < basic-doc.short.xml > basic-doc.xml
runbib -d refs_computing -S Eur.J.Pharmacol. -t db31x basic-doc.xml
5 reference(s) formatted, 0 failed

refdbxml -s Eur.J.Pharmacol.html.xsl -t html basic-doc.xml
Error at xsl:with-param on line 74 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-html/docbk-refdb-html.xsl:
  Variable ADVSTURLSTYLE has not been declared
Error at xsl:choose on line 1826 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  Element must only be used within a template body
Error at xsl:with-param on line 1828 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1831 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1834 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1837 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1840 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1843 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1846 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1849 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1852 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1855 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1858 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1861 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1864 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1867 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1870 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1873 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1876 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1879 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1882 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1885 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1888 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1891 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1894 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1897 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1900 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1903 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1906 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1909 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1912 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1915 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1918 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1921 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1924 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1927 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Error at xsl:with-param on line 1930 of 
file:/usr/local/share/refdb/xsl/docbk-refdb-xsl/docbk-lib/docbk-refdblib.xsl:
  xsl:with-param cannot appear as a child of xsl:when
Transformation failed: Failed to compile stylesheet. 37 errors detected.


My powers of XSL-fu are weak, so I cannot say why this this occurring 
with saxon and not with xsltproc.  Is saxon being too picky?  It is a 
problem, however, as it is occurring with one of the major xsl 
processors.  This was a large problem for me because I initially set 
refdbxml to use saxon (it was my default parser until refdb) and had no 
end of problems.

7. Staying with saxon, if it is ever able to work with refdb, it would 
be good to have an option to run it with the xerces parser.  This 
requires a (lengthy) command line option, but it should be rather doable.

I have found that using xerces with saxon results in much better 
processing of some entities such as quotes and n/mdashes, especially in 
html output.

[To be continued...]

Regards,
David.