RE: [xsl] Summarising XML datasets

[<Jarno.Elovirta@xxxxxxxxx>]

Hi,

> I've tried following the example from Jeni's site and come up with the
> following:

snip

> <xsl:template match="results">
>     <xsl:for-each select="photograph[count(. | key('idkey',
> id)[1])=1]">
>         <h3><xsl:value-of select="id" /></h3>
>         <h4><xsl:value-of select="name"/></h4>
>         <h5><xsl:value-of select="description"/></h5>
> 	<xsl:for-each select="key('subjectkey', id)">
>             <TABLE border="0" width="75%">
>                 <tr>
>                     <th width="10%" align="right">Subject</th>
>                     <td width="90%" align="left"><xsl:value-of
> select="subject" /></td>
>                 </tr>
>             </TABLE>
>             <hr width="75%" align="left"/>
>          </xsl:for-each>
>
>     </xsl:for-each>
> </xsl:template>
> </xsl:stylesheet>
>
> This works unless I have a duplicate subject node, which actually can
> happen.
>
> I'm a little bit stuck with how to only display unique subjects.

Then you need to generate a second grouping key, which is a concatenation of
ID and subject.

  <xsl:key name="subjectkey" match="photograph" use="concat(id, ' ',
subject)"/>

and list subjects with

  <xsl:for-each select="key('idkey', id)[generate-id(.) =
generate-id(key('subjectkey', concat(id, ' ', subject)))]">
    <xsl:if test="not(position() = 1)">, </xsl:if>
    <xsl:value-of select="subject" />
  </xsl:for-each>

Cheers,

Jarno