Question

XML/XSLT Newbie here.
I have some sets of data that need to be compared. The output needs to be a list of Previous Members of a Roster that are not in the current Roster. The Node contains both sets of elements and I need to cycle through and put the missing member into a csv file.
In the example below, Kerry Smith is listed a ‘Previous_Members’ but not ‘Member’, so she needs to be listed in the output.

This same structure will occur for multiple teams. And there can be different #’s on Different Teams. (Team Smith may have 4 members, Team Jones may have 7 members, Team Jackson may have 10, etc.) So the solution needs to be able to handle a variable number of elements.

This is a big stretch for me experience-wise, so any guidance would be appreciated.

<?xml version="1.0" encoding="UTF-8"?>
<am:Team>
<am:First_Day>08/24/2012</am:First_Day>
<am:_Date>08/23/2012</am:_Date>
<am:Member am:Descriptor="Clare Smith">
    <am:ID am:type="Member_ID">4588-D4</am:ID>
</am:Member>
<am:Members am:Descriptor="Jack Smith">
    <am:ID am:type="Member_ID">4588-D3</am:ID>
</am:Members>
<am:Members am:Descriptor="Colin Smith">
    <am:ID am:type="Member_ID">4588-D2</am:ID>
</am:Members>
<am:Previous_Members am:Descriptor="Kerry Smith">
    <am:ID am:type="Member_ID">4588-D1</am:ID>
</am:Previous_Members>
<am:Previous_Members am:Descriptor="Colin Smith">
    <am:ID am:type="Member_ID">4588-D2</am:ID>
</am:Previous_Members>
<am:Previous_Members am:Descriptor="Jack Smith">
    <am:ID am:type="Member_ID">4588-D3</am:ID>
</am:Previous_Members>
<am:Previous_Members am:Descriptor="Clare Smith">
    <am:ID am:type="Member_ID">4588-D4</am:ID>
</am:Previous_Members>

Answer 1

Use:

/*/am:Previous_Members[not(am:ID=/*/am:Member/am:ID)]

This selects all previous team members whose ID isn’t an ID of a current member.

Apply templates only to the elements selected by this XPath expression.

Update:

The XML document as provided contains both: am:Member and am:Members — this is probably a typing error.

Correcting the above , so that there are only am:Members, the XML document becomes:

<am:Team xmlns:am="some:am">
    <am:First_Day>08/24/2012</am:First_Day>
    <am:_Date>08/23/2012</am:_Date>
    <am:Members am:Descriptor="Clare Smith">
        <am:ID am:type="Member_ID">4588-D4</am:ID>
    </am:Members>
    <am:Members am:Descriptor="Jack Smith">
        <am:ID am:type="Member_ID">4588-D3</am:ID>
    </am:Members>
    <am:Members am:Descriptor="Colin Smith">
        <am:ID am:type="Member_ID">4588-D2</am:ID>
    </am:Members>
    <am:Previous_Members am:Descriptor="Kerry Smith">
        <am:ID am:type="Member_ID">4588-D1</am:ID>
    </am:Previous_Members>
    <am:Previous_Members am:Descriptor="Colin Smith">
        <am:ID am:type="Member_ID">4588-D2</am:ID>
    </am:Previous_Members>
    <am:Previous_Members am:Descriptor="Jack Smith">
        <am:ID am:type="Member_ID">4588-D3</am:ID>
    </am:Previous_Members>
    <am:Previous_Members am:Descriptor="Clare Smith">
        <am:ID am:type="Member_ID">4588-D4</am:ID>
    </am:Previous_Members>
</am:Team>

The XPath expression becomes:

/*/am:Previous_Members[not(am:ID=/*/am:Members/am:ID)]

And, for the convenience of the OP, here is a complete transformation that copies exactly the wanted nodes:

<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
 xmlns:am="some:am" exclude-result-prefixes="am">
 <xsl:output omit-xml-declaration="yes" indent="yes"/>
 <xsl:strip-space elements="*"/>

 <xsl:template match="am:Team/*"/>

 <xsl:template match="am:Previous_Members[not(am:ID=/*/am:Members/am:ID)] ">
     <xsl:copy-of select="."/>
 </xsl:template>
</xsl:stylesheet>

When this transformation is applied on the above XML document, the wanted, correct result is produced:

<am:Previous_Members xmlns:am="some:am" am:Descriptor="Kerry Smith">
   <am:ID am:type="Member_ID">4588-D1</am:ID>
</am:Previous_Members>