XSLT to remove duplicate values while concating - xslt

Question

My XML looks like folloing:
<?xml version="1.0" encoding="utf-8" standalone="no"?>
<BATCHES>
<item>
<Material>1000000079</Material>
<Description>330 Bulk</Description>
<Tank>T123</Tank>
<Batch>2013225287</Batch>
<Quantity>510</Quantity>
</item>
<item>
<Material>1000000079</Material>
<Description>330 Bulk</Description>
<Tank>T123</Tank>
<Batch>2013225301</Batch>
<Quantity>520</Quantity>
</item>
<item>
<Material>1000000196</Material>
<Description>340R Bulk</Description>
<Tank>T700</Tank>
<Batch>1000188378</Batch>
<Quantity>510</Quantity>
</item>
<item>
<Material>1000002754</Material>
<Description>43 Bulk</Description>
<Tank>T515</Tank>
<Batch>2013180125</Batch>
<Quantity>300</Quantity>
</item>
<item>
<Material>1000002754</Material>
<Description>43 Bulk</Description>
<Tank>T515</Tank>
<Batch>2013203124</Batch>
<Quantity>200</Quantity>
</item>
<item>
<Material>1000002754</Material>
<Description>43 Bulk</Description>
<Tank>T515</Tank>
<Batch>2013214839</Batch>
<Quantity>700</Quantity>
</item>
<item>
<Material>1000002754</Material>
<Description>43 Bulk</Description>
<Tank>T517</Tank>
<Batch>2013214342</Batch>
<Quantity>890</Quantity>
</item>
</BATCHES>
My original XSLT look like this:
<?xml version="1.0" encoding="utf-8" standalone="no"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:output encoding="UTF-8" indent="yes" method="xml" version="1.0"/>
<xsl:template match="/">
<Rowsets>
<Rowset>
<xsl:variable name="materials" select=".//item[Tank!='RECV' and Tank!='PARK'] "/>
<xsl:for-each select="$materials">
<xsl:if test="generate-id(.)= generate-id($materials[Material=current()/Material])">
<Row>
<Material>
<xsl:value-of select="Material"/>
</Material>
<Description>
<xsl:value-of select="Description"/>
</Description>
<Value>
<xsl:for-each select="$materials[Material=current()/Material]/Tank">
<xsl:if test="node()">
<xsl:value-of select="concat(.,'||')"/>
</xsl:if>
</xsl:for-each>
</Value>
</Row>
</xsl:if>
</xsl:for-each>
</Rowset>
</Rowsets>
</xsl:template>
</xsl:stylesheet>
The result of this XSLT looks like this:
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<Rowsets>
<Rowset>
<Row>
<Material>1000000079</Material>
<Description>330 Bulk</Description>
<Value>T123||T123||</Value>
</Row>
<Row>
<Material>1000000196</Material>
<Description>340R Bulk</Description>
<Value>T700||</Value>
</Row>
<Row>
<Material>1000002754</Material>
<Description>43 Bulk</Description>
<Value>T515||T517||</Value>
</Row>
</Rowset>
</Rowsets>
I wanted to remove duplicate tanks while concatinating it in Value field. So I changed my XSLT to follwoing:
<?xml version="1.0" encoding="utf-8" standalone="no"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:output encoding="UTF-8" indent="yes" method="xml" version="1.0"/>
<xsl:template match="/">
<Rowsets>
<Rowset>
<xsl:variable name="materials" select=".//item[Tank!='RECV' and Tank!='PARK' and Quantity > 500]"/>
<xsl:for-each select="$materials">
<xsl:if test="generate-id(.)= generate-id($materials[Material=current()/Material])">
<Row>
<Material>
<xsl:value-of select="Material"/>
</Material>
<Description>
<xsl:value-of select="Description"/>
</Description>
<Value>
<xsl:for-each select="$materials[Material=current()/Material]/Tank[not(.=preceding::Tank)]">
<xsl:if test="node()">
<xsl:value-of select="concat(.,'||')"/>
</xsl:if>
</xsl:for-each>
</Value>
</Row>
</xsl:if>
</xsl:for-each>
</Rowset>
</Rowsets>
</xsl:template>
</xsl:stylesheet>
My result now looks like this:
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<Rowsets>
<Rowset>
<Row>
<Material>1000000079</Material>
<Description>330 Bulk</Description>
<Value>T123||</Value>
</Row>
<Row>
<Material>1000000196</Material>
<Description>340R Bulk</Description>
<Value>T700||</Value>
</Row>
<Row>
<Material>1000002754</Material>
<Description>43 Bulk</Description>
<Value>T517||</Value>
</Row>
</Rowset>
</Rowsets>
It removed the duplicate tank T123 for material 1000000079 but for material 1000002754, it even removed T515which should appear in Value field as its quantity is greater than 500 for folloing:
<item>
<Material>1000002754</Material>
<Description>43 Bulk</Description>
<Tank>T515</Tank>
<Batch>2013214839</Batch>
<Quantity>700</Quantity>
</item>
what am I doing wrong here?

Answer 1

Ok I see your problem.
For this a key based solution Using Keys to Group: The Muenchian Method:
Try this:
<?xml version="1.0" encoding="utf-8" standalone="no"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:output encoding="UTF-8" indent="yes" method="xml" version="1.0"/>
<xsl:key name="kMaterial" match="item[Tank!='RECV' and Tank!='PARK' and Quantity > 500]"
use="Material"/>
<xsl:key name="kMaterialTank" match="item[Tank!='RECV' and Tank!='PARK' and Quantity > 500]"
use="concat(Material,'|', Tank)"/>
<xsl:template match="/">
<Rowsets>
<Rowset>
<xsl:variable name="materials" select=".//item[Tank!='RECV' and Tank!='PARK'] "/>
<xsl:for-each select="//item[ generate-id(.) = generate-id( key('kMaterial', Material)[1]) ]" >
<xsl:variable name="m" select="Material" />
<Row>
<Material>
<xsl:value-of select="Material"/>,<xsl:value-of select="$m"/>
</Material>
<Description>
<xsl:value-of select="Description"/>
</Description>
<Value>
<xsl:for-each select="//item[ generate-id(.) =
generate-id( key('kMaterialTank', concat( $m,'|', Tank))[1])]" mode="tank" >
<xsl:apply-templates select="." mode="tank" />
</xsl:for-each>
</Value>
</Row>
</xsl:for-each>
</Rowset>
</Rowsets>
</xsl:template>
<xsl:template match="item" mode="tank" >
<xsl:value-of select="Tank"/>
<xsl:text>||</xsl:text>
</xsl:template>
</xsl:stylesheet>
Which will generate the following output:
<?xml version="1.0" encoding="UTF-8"?>
<Rowsets>
<Rowset>
<Row>
<Material>1000000079,1000000079</Material>
<Description>330 Bulk</Description>
<Value>T123||</Value>
</Row>
<Row>
<Material>1000000196,1000000196</Material>
<Description>340R Bulk</Description>
<Value>T700||</Value>
</Row>
<Row>
<Material>1000002754,1000002754</Material>
<Description>43 Bulk</Description>
<Value>T515||T517||</Value>
</Row>
</Rowset>
</Rowsets>