> Or does the mention of EXSLT mean you need a 1.0 solution?
Yes, as you may know DataPower Processore is XSLT 1.0 + EXSLT + proprietary
extensions.

> And why exactly is a recursive solution considered unsuitable for the
> task? (Is it because you need to run this on processors that don't
> optimize tail calls?)
It was not unsuitable, I just wondered that the recursive solution was an
overkill.

And below (and now tested version) from Martin Honnen is what I looked for:
$ cat ent2.xsl
<xsl:stylesheet version="1.0"
  xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>
  <xsl:output omit-xml-declaration="yes" />

  <xsl:variable name="N" select="3"/>

  <xsl:template match="Entity">
    <argument><xsl:value-of select="."/></argument>
  </xsl:template>

  <xsl:template match="/root/data">
    <all>
      <xsl:for-each select="Entity[(position() - 1) mod $N = 0]">
        <arguments>
          <xsl:apply-templates select=". |
            following-sibling::Entity[position() &lt; $N]"/>
        </arguments>
      </xsl:for-each>
    </all>
  </xsl:template>

</xsl:stylesheet>
$


Mit besten Gruessen / Best wishes,

Hermann Stamm-Wilbrandt
Developer, XML Compiler, L3
Fixpack team lead
WebSphere DataPower SOA Appliances
----------------------------------------------------------------------
IBM Deutschland Research & Development GmbH
Vorsitzender des Aufsichtsrats: Martin Jetter
Geschaeftsfuehrung: Dirk Wittkopp
Sitz der Gesellschaft: Boeblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294



From:       Michael Kay <mike@xxxxxxxxxxxx>
To:         xsl-list@xxxxxxxxxxxxxxxxxxxxxx
Date:       10/19/2010 08:09 PM
Subject:    Re:  question on EXSLT data partitioning




What's wrong with

<xsl:variable name="Entities" select="child::Entity"/>
<xsl:for-each select="1 to (count($Entities) idiv $PARTITION_SIZE">
<partition>
<xsl:copy-of select="subsequence($Entities, (1 + (. - 1) *
$PARTITION_SIZE), $PARTITION_SIZE)"/>
</partition>
</xsl:for-each>

Or does the mention of EXSLT mean you need a 1.0 solution?

And why exactly is a recursive solution considered unsuitable for the
task? (Is it because you need to run this on processors that don't
optimize tail calls?)

Michael Kay
Saxonica

> Hello,
>
> yesterday I was asked by a colleague on data partitioning.
> He wanted to partition 100000s of Entities in blocks of 1000
> for sending a single Database update for 1000 entities.
>
> Below is the simplified input, partition size is N=3 and the
> requested output. Below that is the solution I provided.
>
> Here are my questions:
> * can this task be done without recursion in EXSLT?
>    [the colleage did not like the idea of doing the partitioning with
>     just XPath (1<=position()<=1000, 1001<=position()<=2000, ...)
>     because of the 6 digit number of entities]
> * is the conversion of Entity to argument by apply-templates
>    the way to go?