Re: [xsl] position behavior in nested grouping

Thanks Martin,

It is working like a charm.

Regards,
Ganesh


On Thu, Jul 30, 2009 at 7:12 PM, Martin Honnen<Martin.Honnen@xxxxxx> wrote:
> Ganesh Babu N wrote:
>>
>> all those should be clubbed into single section by taking group-adjcent.
>>
>> The output should be
>>
>>                       <sections>
>>                              <section-title>Contents</section-title>
>>                          <section>
>>                               <section-head>Abstracts of the Series
>> of Fascia Congress - Part
>> 4</section-head>
>>                               <para>
>>                                       <page>270</page>
>>                               </para>
>>                          </section>
>>                         <section>
>>                            <section-head/>
>>                              <para>
>>                                       <page>271</page>
>>                               </para>
>>                               <para>
>>                                       <page>271</page>
>>                               </para>
>>                          </section>
>>                          <section>
>>                               <section-head>Abstracts of the Series
>> of Fascia Congress - Part
>> 4</section-head>
>>                               <para>
>>                                       <page>272</page>
>>                               </para>
>>                       </section>
>>                       <section>
>>                               <section-head>Prevention &amp;
>> Rehabilitation</section-head>
>>                               <section-subhead>Editorial</section-subhead>
>>                               <para>
>>                                       <page>273</page>
>>                               </para>
>>                               <para>
>>                                       <page>277</page>
>>                               </para>
>>                       </section>
>>                       <section>
>>                               <section-subhead>Learning
>> Methodology</section-subhead>
>>                               <para>
>>                                       <page>284</page>
>>                               </para>
>>                       </section>
>>               </sections>
>
> If you know that you want group-adjacent then it is only a slight change
> from the earlier stylesheet:
>
> <xsl:stylesheet
>  xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>  version="2.0">
>
>  <xsl:output indent="yes"/>
>
>  <xsl:template match="articles">
>    <sections>
>      <section-title>Contents</section-title>
>      <xsl:for-each-group select="row" group-adjacent="if (col[2] = '') then
> '' else tokenize(col[2], '\|')[1]">
>        <xsl:choose>
>          <xsl:when test="contains(col[2], '|')">
>            <xsl:variable name="outer-key" select="current-grouping-key()"/>
>            <xsl:for-each-group select="current-group()"
> group-by="tokenize(col[2], '\|')[2]">
>              <section>
>                <xsl:if test="position() eq 1">
>                  <section-head><xsl:value-of
> select="$outer-key"/></section-head>
>                </xsl:if>
>                <section-subhead><xsl:value-of
> select="current-grouping-key()"/></section-subhead>
>                <xsl:apply-templates select="current-group()"/>
>              </section>
>            </xsl:for-each-group>
>          </xsl:when>
>          <xsl:otherwise>
>            <section>
>              <section-head><xsl:value-of
> select="current-grouping-key()"/></section-head>
>              <xsl:apply-templates select="current-group()"/>
>            </section>
>          </xsl:otherwise>
>        </xsl:choose>
>      </xsl:for-each-group>
>    </sections>
>  </xsl:template>
>
>  <xsl:template match="row">
>    <para>
>      <page><xsl:value-of select="col[3]"/></page>
>    </para>
>  </xsl:template>
> </xsl:stylesheet>
>
> The problem why the previous solution skips the row elements with an empty
> col[2] child element is that the tokenize function returns an empty
sequence
> for the empty string and that does not result in a group. But
> group-adjacent="if (col[2] = '') then '' else tokenize(col[2], '\|')[1]"
> avoids that problem and simply form a group for those elements with an
empty
> col[2].
>
> --
>
>        Martin Honnen
>        http://msmvps.com/blogs/martin_honnen/