This is the mail archive of the docbook-apps@lists.oasis-open.org mailing list .


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Onechunk problem



Okay, all of this information about the refname is fascinating, but I'm
still confused by the problem with onechunk.xsl.

Norm put a fix in CVS for the bug I reported, thinking the problem was in
DocBook. That fix adds the line

      <xsl:param name="onechunk" select="1"/>

to onechunk.xsl. I installed the fix, but I still get the same errors.
Apparently, I cannot use chapters in books or appendices in articles. I
have no refnames with spaces in either of these cases. The filename is
supposed to by taken from the root id, and that contains no spaces or odd
characters. I still keep getting the error that "chapter is not a chunk!"
or "appendix is not a chunk!" as well as the runtime error pointing to line
79 in chunker.xsl. This is the template from chunker that contains the
offending line:

    <xsl:when test="element-available('exsl:document')">
       <exsl:document href="{$filename}"
                      method="{$method}"
                      encoding="{$encoding}"
 79-->                indent="{$indent}">
         <xsl:copy-of select="$content"/>
       </exsl:document>
     </xsl:when>

 Now, am I just misreading the exslt page, or is it noteworthy that this
 particular type of exsl expression is the one listed on
 http://exslt.org/exsl/exsl.html as "not considered stable" and "not part
 of the core of EXSLT-Common"? If this is the source of trouble, is there
 an available alternative to exslt?

 Is there any way to make this work?

 Thanks.

 Dennis Grace

 Information Developer
 IBM Linux Technology Center
 (512) 838-3937  T/L 678-3937  cell: (512)-296-7830
 dgrace@us.ibm.com

 What is this some kind of demonstrative pronoun?




|---------+---------------------------->
|         |           Daniel Veillard  |
|         |           <veillard@redhat.|
|         |           com>             |
|         |                            |
|         |           05/15/02 03:15 AM|
|         |           Please respond to|
|         |           veillard         |
|         |                            |
|---------+---------------------------->
  >--------------------------------------------------------------------------------------------------------------|
  |                                                                                                              |
  |       To:       Michael Smith <smith@xml-doc.org>                                                            |
  |       cc:       docbook-apps@lists.oasis-open.org                                                            |
  |       Subject:  Re: DOCBOOK-APPS: Onechunk problem                                                           |
  |                                                                                                              |
  |                                                                                                              |
  >--------------------------------------------------------------------------------------------------------------|



On Wed, May 15, 2002 at 03:03:34AM -0500, Michael Smith wrote:
> Bob Stayton writes:
>
> > On Fri, May 03, 2002 at 03:30:17PM -0500, Dennis Grace wrote:
> > > > About my error message:
> > > >
> > > > Error chapter is not a chunk!
> > > > Writing for chapter
> > > > runtime error: file
> > > > /usr/share/sgml/docbook/docbook-xsl-1.50.0/html/chunker.xsl
> > > > line 79 element document
> > > > xsltDocumentElem: URL computation failed for
> > [...]
> >
> > The line number reported is the line with the closing ">"
> > for exsl:document element.  I think the xsltproc parser
> > is using that as the element line.  I have several files
> > that do not generate the error, and two that do, so it is
> > not happening with all files.  I haven't had time to
> > investigate what is different about those two files.
>
> It appears that at least one thing that will cause this problem is
> having spaces in a refname. The current db2man code takes the refname

  yes libxslt doesn't cope well with the document extension if there
are unescaped spaces in the target chunk name (in the URI sense because
I assume the name is an URI Reference, and then pass it to libxml2 layer
doing the URI- Ref computation job, which itself fails because "a b" is
not an URI-Reference).
  I don't know what's the best way to handle this, maybe libxslt could
URI escape the returned string automatically ...

Daniel

--
Daniel Veillard      | Red Hat Network https://rhn.redhat.com/
veillard@redhat.com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/






Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]