Stefan Kleineikenscheidt wrote:
> Hi all,
> 
> i'm trying to convert an HTML page to a hierachical structure, but I am
> stuck.  Consider a page like that:
> 
> <h1>First Heading1</h1>
>   <p>some text</p>
>   <p>more text</p>
> 
>   <h2>First Heading2</h2>
>     <p>more text</p>
> 
>   <h2>Second Heading2</h2>
>     ...
> <h1>Second Heading1</h1>
>   ...
>   <h2>Third Heading2</h2>
>     ...

First of all you would need to make it well-formed XHTML (use W3C Tidy
for that). This ensures that any subsequent XSLT process won't gag.

> This is my 'h1' template, where i try to process all elements between
> two 'h1' elements:
> 
>   <xsl:template match="//h:h1">
>     <section>
>       <title><xsl:value-of select="text()" /></title>
>       <xsl:variable name="nexth1" select="position(parent::*/*[(name()
> = 'h1')])" />
>       <xsl:apply-templates select="following-sibling::*[position()
> &lt;= $nexth1]" />
>     </section>
>   </xsl:template>

<?xml version="1.0" encoding="iso-8859-1"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
                 version="1.0">

   <xsl:output method="xml" indent="yes"/>

   <xsl:template match="h1|h2|h3|h4">
     <xsl:variable name="id" select="generate-id(.)"/>
     <xsl:variable name="level">
       <xsl:value-of select="number(translate(name(),'h',''))"/>
     </xsl:variable>
     <xsl:variable name="gi" select="name()"/>
     <xsl:element name="{concat('sect',$level)}">
       <xsl:attribute name="id" select="$id"/>
       <title>
         <xsl:apply-templates/>
       </title>
       <xsl:apply-templates select="following-sibling::*
           [generate-id(preceding-sibling::*[name()=$gi][1])=$id]
           [not(substring(name(),1,1)='h' and name()!='hr' and
            number(translate(substring(name(),1,1),'h',''))&lt;$level)]
 
[not(number(translate(name(preceding-sibling::*[substring(name(),1,1)='h' 
and name()!='hr'][1]),'h',''))&lt;$level)]"/>
     </xsl:element>
   </xsl:template>

   <xsl:template match="p">
     <para>
       <xsl:apply-templates/>
     </para>
   </xsl:template>

</xsl:stylesheet>

This needs some more work: it's not subsetting out the higher-level H* 
element types, but I've run out of time here.

///Peter
-- 
XML FAQ: http://xml.silmaril.ie/