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: [docbook-apps] Callouts and numbering


Hi Andreas,
This template in fo/callout.xsl establishes the numbering of co:
 
<xsl:template match="co" mode="callout-bug">
  <xsl:call-template name="callout-bug">
    <xsl:with-param name="conum">
      <xsl:number count="co"
                  level="any"
                  from="programlisting|screen|literallayout|synopsis"
                  format="1"/>
    </xsl:with-param>
  </xsl:call-template>
</xsl:template>
It resets within the four listed elements, which are text display elements.  The other elements in which co are permitted are inlines. Since co is intended to add callouts to a text display, I believe the intention was that the inlines were to be within one of these four display elements.  There is no way to express such a restriction in DTD syntax, so it was left loose enough to permit it.  But the stock stylesheets don't support co outside of those four elements.
 
You are of course free to customize this template as needed.
 
Bob Stayton
Sagehill Enterprises
DocBook Consulting
bobs@sagehill.net
 
 
----- Original Message -----
Sent: Friday, April 29, 2005 1:01 AM
Subject: [docbook-apps] Callouts and numbering

Hello!
 
I've encountered a situation where one of the writers has somehow produced a <calloutlist> (and corresponding <co>) in an appendix which continues its numeration from a previous callout, in a previous chapter. I minimized the example to a
 <para>
  <userinput> text <co id="a"/></userinput>
 </para>
 
still showing the problem. Changing the <para> to <screen> makes the callout being numbered from 1 again, like I want it.
 
The documentation of <co> says it can be the child of a <userinput> or a <screen> - which it is in both variants - but I cannot find any information under which conditions the callout numbering is reset to start from 1 again. Does anybody know the details?
 
In this case the writer uses the <co> in a <procedure> <step> <userinput> construct which validates fine, but gives wrong output. We are using DTD 4.3, xslt 1.67.2 (xsltproc-RenderX processing).
 
   /Andreas
 

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