Matching two different blocks in XML using STX parser

Question

I have a large XML file that I am trying to match two blocks in by id using the java joost implementation for STX. This is my XML file:

<?xml version='1.0' encoding='UTF-8'?>
<main>
    <table>
        <description id="1" size="big" />
        <description id="2" size="small" />
        <description id="3" size="medium" />
    </table>
    <products>
        <item id="1" color="red" />
        <item id="2" color="green" />
        <item id="3" color="blue" />
    </products>
</main>

And this is the output I'm trying to create:

<?xml version='1.0' encoding='UTF-8'?>
<total>
    <item id="1" color="red" size="big" />
    <item id="2" color="green" size="small" />
    <item id="3" color="blue" size="medium" />
</total>

The stx:buffer did not give any results, and there was no solution for creating key-value pairs similar to a java HashMap. Is there any solution?

Answer 1

I am not familiar with STX, with standard XSLT 3.0 streaming you can use an accumulator holding a map:

<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
    xmlns:xs="http://www.w3.org/2001/XMLSchema"
    xmlns:map="http://www.w3.org/2005/xpath-functions/map"
    exclude-result-prefixes="#all"
    version="3.0">
  
  <xsl:accumulator name="description" as="map(xs:untypedAtomic, xs:string)" initial-value="map{}" streamable="yes">
    <xsl:accumulator-rule match="table/description" select="map:put($value, @id, string(@size))"/>
  </xsl:accumulator>

  <xsl:mode on-no-match="shallow-copy" use-accumulators="description" streamable="yes"/>

  <xsl:output method="xml" indent="yes"/>
  
  <xsl:template match="main">
    <xsl:apply-templates/>
  </xsl:template>

  <xsl:template match="table"/>
  <xsl:template match="products">
    <total>
      <xsl:apply-templates/>
    </total>
  </xsl:template>
  
  <xsl:template match="item">
    <xsl:copy>
      <xsl:apply-templates select="@*"/>
      <xsl:attribute name="size" select="accumulator-before('description')(@id)"/>
    </xsl:copy>
  </xsl:template>

</xsl:stylesheet>

XSLT 3.0 with streaming seems currently to be only supported by Saxon EE (9.8 and later (i.e. 9.9, 10, 11, 12)).

I have now tried to construct an STX example, the following seems to work:

<stx:transform xmlns:stx="http://stx.sourceforge.net/2002/ns" version="1.0"
               pass-through="all" strip-space="yes">

  <stx:variable name="id"/>
  <stx:variable name="size"/>

  <stx:template match="table/description">
    <stx:assign name="id" select="($id, string(@id))"/>
    <stx:assign name="size" select="($size, string(@size))"/>
  </stx:template>

  <stx:template match="/* | table | table/description">
    <stx:process-children/>
  </stx:template>

  <stx:template match="products">
    <total>
      <stx:process-children/>
    </total>
  </stx:template>

  <stx:template match="item">
    <stx:copy attributes="@*">
      <stx:attribute name="size" select="item-at($size, index-of($id, @id))"/>
    </stx:copy>
  </stx:template>

</stx:transform>

Or using a Java HashMap:

<stx:transform xmlns:stx="http://stx.sourceforge.net/2002/ns" version="1.0"
               xmlns:hm="java:java.util.HashMap"
               exclude-result-prefixes="hm"
               pass-through="all" strip-space="no">

  <stx:variable name="description" select="hm:new()"/>

  <stx:template match="table/description">
    <stx:variable name="oldValue" select="hm:put($description, string(@id), string(@size))"/>
    <!-- <stx:assign name="description" select="$description"/> -->
  </stx:template>

  <stx:template match="/* | table | table/description">
    <stx:process-children/>
  </stx:template>

  <stx:template match="products">
    <total>
      <stx:process-children/>
    </total>
  </stx:template>

  <stx:template match="item">
    <stx:copy attributes="@*">
      <stx:attribute name="size" select="hm:get($description, string(@id))"/>
    </stx:copy>
  </stx:template>

</stx:transform>